LIBRARY UNIVERSITY OF CALIFORNI>4 DAVIS -^J^+ , , ■ DEPAH HSITY OF CALIFOiUMA UaRARY , 4 ■AVIS ^ STATE OF CALIFORNIA DEPARTMEN r OF A\A1 ER RESOURCES ISION OF RESOURCES PLANNING BULLETIN No. 63 COPY a SEA WATER INTRUSION IN CALIFORNIA APPENDIX B REPORT BY LOS ANGELES COUNTY FLOOD CONTROL DISTRICT ON INVESTIGATIONAL WORK FOR PREVENTION AND CONTROL OF SEA WATER INTRUSION, WEST COAST BASIN EXPERI- MENTAL PROJECT, LOS ANGELES COUNTY As Directed by Chapter 1500, Statutes of 1951 GOOD\MN J. KNIGHT Governor March, 1957 HARVEY O. BANKS Director of Wa ter Resources „ . i'uNlVERSlTY OF CALIFORNIA I OAV/t'^'i FEB 2 :'» 1959 LIBRARY L \ STATE OF CALIFORNIA DEPARTMENT OF WATER RESOURCES DIVISION OF RESOURCES PLANNING BULLETIN No. 65 SEA WATER INTRUSION IN CALIFORNIA APPENDIX B REPORT BY LOS ANGELES COUNTY FLOOD CONTROL DISTRICT ON INVESTIGATIONAL WORK FOR PREVENTION AND CONTROL OF SEA WATER INTRUSION, WEST COAST BASIN EXPERI- MENTAL PROJECT, LOS ANGELES COUNTY As Directed by Chapter 1500, Statutes of 1951 GOODWIN J. KNIGHT HARVEY O. BANKS Governor Director of Water Resources March, 1957 LIBRARY UNIVERSITY OF CAUFORNIA DAVIS i TABLE OF CONTENTS Page LETTER OF TRANSMITTAL, STATE WATER BOARD 5 ORGANIZATION, STATE DEPARTMENT OF WATER RESOURCES, DIVISION OF RESOURCES PLANTSTING 7 ORGANIZATION, STATE WATER BOARD 7 PART T. INTRODUCTION Authorization 9 Location of West Coast Basin Experimental Project and Conduct of In- vestigation 9 Agreement for Performance of Investigational Work for Prevention and Control of Sea-Water Intrusion 10 Amendment to Agreement for Performance of Investigational Work for Prevention and Control of Sea- Water Intrusion 15 Second Amendment to Agreement for Performance of Investigational Work for Prevention and Control of Sea-Water Intrusion 16 Agreement Permitting Los Angeles County Flood Control District to Take Possession of the Manhattan Beach Experimental Project 17 Extension of Agreement Permitting Los Angeles County Flood Control District to Take Possession of the Manhattan Beach Experimental Project 18 Action by the Los Angeles County Flood Control District 18 PART II. REPORT BY LOS ANGELES COUNTY FLOOD CONTROL DISTRICT ON INVESTIGATIONAL WORK FOR PREVENTION AND CONTROL OF SEA-WATER INTRUSION, WEST COAST BASIN EXPERIMENTAL PROJECT, LOS ANGELES COUNTY 19 (3) Address All Communications TO THE CHArRMAN p. O. BOX 1079 SACRAMENTO 5 Goodwin J. Knight GOVERNOR CLAIR A. HILL. CHAIRMAN REDOING A. FREW. Vice CHAIRMAN KING City STATE OF CALIFORNIA DEPARTMENT OF WATER RESOURCES Harvey O. Banks director 1 120 N STREET SACRAMENTO JOHN P. BUNKER. GUSTINC EVERETT L. GRUBB, ELSINORK W. P. RICH. MARYSVILLE PHIL O. SWING. SAN DiECO KENNETH Q. VOLK, LoS ANGCLes Mai-L-li 1."), 1957 Honorable Goodwin J. Knight, Governor and Members of the Legislature of the State of California Gentlemen : I have the honor to transmit herewith Appendix B entitled, ' ' Re- port by the Los Angeles County Flood Control District on Investigational Work for Prevention and Control of Sea-water Intrusion, West Coast Basin Exjieri- raental Project, Los Angeles County," January 27, 1955. This appendix, author- ized by Chapter 1500, Statutes of 1951, is of several appendixes to accompany Bulletin No. 63 of the Department of "Water Resources, entitled, "Sea-water Intrusion in California. ' ' The West Coast Basin Experimental Project was constructed and operated in the Manhattan Beach-Hermosa Beach area, Los Angeles County, by the Los Angeles County Flood Control District, under supervision of the Division of Water Resources and in accordance with terms of a contract with the State Water Resources Board. The project was designed to determine the feasibility and practicability of creating a pressure ridge in confined aquifers by injection of fresh water through wells for prevention and control of sea-water intrusion. Appendix B describes in considerable detail the results of a two and one-half year experimental study of one of the more important methods of control of fresh water through wells for prevention and control of sea-water intrusion. Very truly yours, Harvey 0. Banks Director (5) ORGANIZATION STATE DEPARTMENT OF WATER RESOURCES D/VISION OF RESOURCES PLANNING HARVEY O. BANKS Director of Wafer Resources M. J. SHELTON Deputy Director WILLIAM L. BERRY Chief, Division of Resources Planning CARL B. MEYER .Chief, Special Activities Branch MAX BOOKMAN District Engineer, Southern California OfFice STAFF MEMBERS ASSISTING IN EXPERIMENTS LAURENCE B. JAMES Chief Engineering Geologist E. C. MARLIAVE (now resigned) Supervising Engineering Geologist RAYMOND C. RICHTER _ _ Supervising Engineering Geologist WILLARD R. SLATER Supervising Hydraulic Engineer PHILIP J. LORENS Senior Engineering Geologist DONALD L. McCANN Assistant Engineering Geologist RESIDENT ENGINEER AT WEST COAST BASIN EXPERIMENTAL PROJECT JACK J. COB Senior Hydraulic Engineer STATE WATER BOARD CLAIR A. HILL, Chairman, Redding A. FREW, Vice Chairman, King City JOHN P. BUNKER, Gustine W. P. RICH, Marysville EVERETT L. GRUBB, Riverside PHIL D. SWING, San Diego KENNETH Q. VOLK, Los Angeles SAM R. LEEDOM, Administrative Assistant Note: Prior fo esfab/isfimenf of fhe Deparfmenf of Wafer Resources on July 5, 1956, the following organizaiional positions were in effect under the Division of Water Re- sources and the State Water Resources Board: DIVISION OF WATER RESOURCES HARVEY O. BANKS * State Engineer L. C. JOPSON Assstant State Engineer HENRY HOLSINGER _ Principal Attorney T. R. MERRYWEATHER Administrative Assistant STATE WATER RESOURCES BOARD CLAIR A. HILL, Chairman, Redding R. V. MIEKLE, Vice Chairman, Turlock A. FREW, King City W. P. RICH, Marysville C. A. GRIFFITH, Azusa W. PENN ROWE, San Bernardino PHIL D. SWING, San Diego ' A. D. Edmonston was State Engineer until his retirement on November 1, 1955. (7) PART I INTRODUCTION Sea-water intrusion is a major problem in ground water basins bordering the California coast. The State Leprislature. in diroetinjr that an investigation be made to determine plans for the prevention and control of sea-water intrusion, recognized this fact and stated that in conducting the study, considera- tion was to be given to the determination of criteria for control of sea-water intrusion by the creation and maintenance of a pressure ridge by the introduction of fresh water through wells into aquifers. AUTHORIZATION Through enactment of Chapter 1500, Statutes of 1951, the California Legislature directed that an ex- perimental program be undertaken to determine criteria for the prevention and control of sea-water intrusion into ground water basins. "The sum of seven hundred fifty thousand dollars ($750,000) is hereby appropriated out of any monej- not otherwise appropriated in the Postwar Unem- ployment and Construction Fund, or, in the event that such amount of money is not available there- from, then to the extent not so available out of any money not otherwise appropriated in the General Fund to the State Water Resources Board for in- vestigational work and design criteria for correc- tion or prevention of damage to underground wa- ters of the State by sea-water intrusion in the "West Coast Basin of Los Angeles County and other critical areas. The cost of such investigation and study shall include the cost of providing water, water injection wells, observation wells, water spreading grounds, pipe lines, equipment, rights of way, and other facilities necessary to introduce water into the water-bearing aquifers. The board is authorized to cooperate and contract with the Los Angeles County Flood Control District, the "West Basin I\Iunieipal "Water District, and any other public or private corporation or agency to the purpose of this act." In order to carry out the intent of this legislation, the State "Water Resources Board on July 6, 1951, re- quested that the Division of "Water Resources investi- gate and submit recommendations as to an experi- mental program for the prevention and control of sea-water intrusion. These recommendations, set forth in a report entitled "Proposed Investigational Work for Control and Prevention of Sea- Water In- trusion into Ground Water Basins," dated August, 11)51, advocated performance of laboratory research concerning well injection, basic parameters of sea- water intrusion, and reduction in aquifer permeabil- ity, to be carried on concurrently and coordinated with a large scale field experiment to investigate the hydraulic feasibility of creating a pressure ridge in confined aquifers. In addition, the report recom- mended that the State Water Resources Board and its staff further investigate construction techniques involved in actual installations of cutoff walls and the effectiveness of such walls in impeding the lateral movement of ground water. A study of the economic factors involved in prevention of sea-water intrusion was also recommended. The State Water Resources Board, in order to im- plement the foregoing investigational program, exe- cuted a contract with the Los Angeles County Flood Control District on October 1, 1951, for execution of certain portions of the pi-ogram. The contract, and subsequent amendments thereto, authorized the Los Angeles County Flood Control Di.strict to install and operate experimental facilities, under supervision of the Division of Water Resources, to ascertain the h.ydraulic feasibility of creating a pressure ridge by use of injection wells and its effectiveness in prevent- ing sea-water intrusion, and to report thereon. LOCATION OF WEST COAST BASIN EXPERI- MENTAL PROJECT AND CONDUCT OF INVESTIGATION The large scale experimental field project was lo- cated in the cities of Manhattan Beach and Hermosa Beach, Los Angeles County, California. The location of this site was selected after careful consideration of alternate sites in other critical areas throughout California. Late in 1951, the Los Angeles County Flood Con- trol District, in close cooperation with the Division of Water Resources completed plans for construction of project facilities, including injection and observa- tion wells and feeder lines, and operation of the field test on an experimental basis. Detailed geologic ex- ploration was also commenced at this time. By Febru- ary, 1953, construction of project facilities was com- pleted, and injection of treated Colorado River water was commenced. Throughout the field investigation, monthly progress reports were submitted to the State Water Resources Board both by the Division of Wa- ter Resources and Los Angeles County Flood Control District. (9) 10 SEA WATER INTRUSION IN CALIFORNIA In accordance with terms of the agreement, the State retained ownership of project facilities until after completion of the prescribed work, at which time certain property was retained by the State Water Resources Board for use by the Division of Water Resources on Board work and the remaining nonexpendable property was sold to the Los Angeles County Flood Conti-ol District. The State Water Resources Board allocated a total of $642,126.30 to Los Angeles County Flood Control District for prosecution of this experimental study. These funds, except for $9,000 reserved for comple- tion of the District's final report, were exhausted in December, 1953. Operation since that time has been financed from local and county taxes and by local contributions. Other phases of the investigational program recom- mended by the Division of Water Resources were completed by the University of California at Berkeley and Los Angeles, and United States Geological Sur- vey, Qualit}' of Water Branch, at Sacramento and Menlo Park. Final reports by these agencies describ- ing the work accomplished appear as Appendixes C, D and E. AGREEMENT FOR PERFORMANCE OF INVESTIGATIONAL WORK FOR PREVENTION AND CONTROL OF SEA-WATER INTRUSION This Agreement, entered into as of October 1, 1951, by and between the State Water Resources Board, hereinafter referred to as the Board, and The Los Angeles County Flood Control District, hereinafter referred to as the District, wituesseth : Whekeas, by Chapter 1500, Statutes of 1951, the sum of seven hundred and fifty thousand dollars ($750,000) is appropriated to the Board for investiga- tional work and study with the objective of formulat- ing plans and design criteria for the correction or pre- vention of damage to underground water of the State by sea-water intrusion in the West Coast Basin of Los Angeles County and other critical coastal areas; and Whereas, a report entitled "Proposed Investiga- tional Work for Control and Prevention of Sea- water Intrusion Into Ground Water Basins," dated August 1951, was prepared at the request of the Board by the State Engineer, and Whereas, said report and recommendations con- tained therein were accepted by the Board and ap- proved at a regular meeting on September 7, 1951, and Whereas, said approved report contains a recom- mended program of investigations, part of which is set forth as f oUows : "2. A sum of $450,000 .should be allocated from funds presently available for installation and one year's operation of a field experimental project to investigate the hydraulic feasibility of creating a pressure ridge in confined aquifers by means of injection wells and the effectiveness of such a ridge in preventing sea-water intrusion. This project should be undertaken in the vicinity of Manhattan Beach in West Coast Basin, Los Angeles County. The initial installation should comprise five injec- tion wells and sufficient observation wells to yield the observational data necessary for complete and conclusive interpretation of the results. Capital cost for feeder and distribution pipe lines should be kept to a minimum and emphasis placed upon experimental techniques and collection of pertinent data so that the results and conclusions therefrom will be applicable on a State-wide basis. "3. Additional funds should be allocated to the Manhattan Beach field experiment if results of the first six months operation analyzed in conjunction with laboratory' research studies indicate that the initial installation of five injection wells is not ex- tensive enough to yield conclusive results." and "7. The State Water Resources Board should carefully supervise the [)lanning and execution of all experimental work in order to assure that the results obtained therefrom are interpretable and usable on a State-wide basis." Whereas, the Board is authorized by Chapter 1500, Statutes of 1951, to cooperate and contract with the Los Angeles County Flood Control District, the West Basin Municipal Water District, aud/or any other public or private corporation or agency for the pur- pose of making such investigations; and Whereas, the Board desires the District to under- take certain portions of the investigational work and study under the direction and supervision of the State Engineer, acting as Engineer for the Board ; Now Therefore, in consideration of the premises and of the several promises to be performed by each as hereinafter set forth, the Board and the District do hereby mutually agree as follows: ARTrCLE l-PROJECT Description of Work The project will consist of providing plans, form- ulating procedures, and conducting investigations which will comprise but not be limited to the follow- ing: A. Preparing plans, specifications and cost esti- mates for furnishing and installing : SEA WATER INTRUSION IN CALIFORNIA 11 1. Approximately 7,800 feet of jtipe line from tlie Metropolitan Water District vault at Manhattan Beach Boulevard and Redondo Avenue in the City of Manhattan Beach ■westerly to the Santa Fe riprht of way at ;Manhattan Beach Boulevard, IManhattan Beach, California; 2. Pipe lines for distribution alon'^ the Santa Fe right of way approximately parallel to the coast line; 3. Five recharcre wells inchidinp: drilling:, cor- ingr. samplinjr, casing:, and developing:; 4. Approximately 30 observation wells includ- ing: drilling:, coring:, sampling, easing, and developing ; 5. Necessary well connections, meters, valves and pressure regulators ; 6. Chlorinating equipment and housing. Advertising, receiving bids, and executing and completing a contract or contracts for the in- stallation of that portion of the work listed in "A," which cannot be advantageously per- formed by District forces, and providing super- vision over that portion of the work to be done under such contracts; Furnishing labor and materials for that portion of the work listed in "A" above which is to be done by the District forces ; Contracting with the Metropolitan Water Dis- trict (or a member agency) for the purchase of Metropolitan Water District water and a tem- porary connection at the vault located at ]\Ian- hattan Beach Boulevard and Redondo Avenue in the City of Manhattan Beach, California, for supply of water for the field experimental work outlined herein. Formulating plans for, and, after approval by the State Engineer, conducting investigations and experimental work to determine the feasi- bility and practicality of creating a pressure ridge in confined aquifers for prevention and control of sea-water intrusion as outlined herein, including collection of the necessary data to de- termine if possible the following factors: 1. Feasible rates of injection through wells as related to thickness, permeability and other properties of the aquifers and variation in rates of injection with pressure head built up in the well and with time ; 2. Height and shape of pressure ridge that can be built up as related to thickness and per- meability of aquifer and hydraulic gradient ; 3. Required height of pressure ridge and amount of water necessary to inject to control intru- sion of sea-water as related to properties and depth of aquifer; 4. Required spacing of injcctioii wells to con- trol sea water intrusion as related to lliick- uess of aquifer, permeability and hydraulic gradient ; 5. Gradient of the piezometric surface in the water-bearing deposits and its cfTect on the quantity of water injected and the shape of the pressure mound ; 6. Rates and amounts of displacement of saline waters and/or degree of dilution of saline waters ; 7. Effect of ground water extractions by pump- ing in the inland areas on the rate of injec- tion necessary to control sea-water intrusion ; 8. Degree of chlorination or other treatment necessary for continued injection of water at feasible rates. 9. Maintenance of wells, including procedures such as sand bailing, surging, de-aeration, and studies of formation of micro-organisms and the effect of chlorination or other methods of disinfection on their growth, and studies of base-exchange reactions and sus- pended solids deposition. F. Preparation of reports on the findings as speci- fied in Article VI herein. Locafion Manhattan Beach, Los Angeles County, California. Amount of Allocation An allocation of $450,000 for the initial program which may be supplemented by additional allocations for further investigation subject to the approval by the Board. ARTICLE ll-SURVEYS AND PLANS The District will make necessary surveys and pre- pare the plans and specifications and cost estimates needed for the work contemplated in this Agreement. Outlines of work and procedures for conduct of the experimental work and plans and specifications and cost estimates as prepared by the District for various items of work shall be submitted by the District to the State Engineer for review and approval prior to commencement of work. No changes shall be made in the work program and plans and specifications after such approval except with the consent of the State Engineer. ARTICLE lll-RIGHTS OF WAY The District shall obtain all necessary rights of way, licenses, permits for entry, or easements necessary for the work hereinabove described under the Project. Approval of the State Engineer wiU be secured prior to making any expenditures for rights of way or ease- ments except for the incidental expense involved therein in accepting and recording same. 10 SEA WATER INTRUSION IN CALIFORNIA lu accordance with terms of the agreement, the State retained ownership of project facilities until after completion of the prescribed work, at which time certain property was retained by the State "Water Resources Board for use by the Division of Water Resources on Board work and the remaining nouespeudable property was sold to the Los Angeles County Flood Control District. The State Water Resources Board allocated a total of $642,126.30 to Los Angeles County Flood Control District for prosecution of this experimental study. These funds, except for $9,000 reserved for comple- tion of the District's final report, were exhaust December, 1953. Operation since that time has financed from local and county taxes and by contributions. Other phases of the investigational program r mended by the Division of Water Resources completed by the University of California at Bei and Los Angeles, and United States Geological vey, Quality of Water Branch, at Sacramentc Menlo Park. Final reports by these agencies de ing the work accomplished appear as Appendix D and E. AGREEMENT FOR PERFORMANCE OF INVESTIGATIONAL WORK FOR PREVENTION CONTROL OF SEA-WATER INTRUSION This Agreement, entered into as of October 1, 1951, by and between the State Water Resources Board, hereinafter referred to as the Board, and The Los Angeles County Flood Control District, hereinafter referred to as the District, wituesseth : Whereas, by Chapter 1500, Statutes of 1951, the sum of seven hundred and fifty thousand dollars ($750,000) is appropriated to the Board for investiga- tional work and study with the objective of formulat- ing plans and design criteria for the correction or pre- vention of damage to underground water of the State by sea-water intrusion in the West Coast Basin of Los Angeles County and other critical coastal areas; and Whereas, a report entitled "Proposed Investiga- tional Work for Control and Prevention of Sea- water Intrusion Into Ground Water Basins," dated August 1951, was prepared at the request of the Board by the State Engineer, and Whereas, said report and recommendations con- tained therein were accepted by the Board and ap- proved at a regular meeting on September 7, 1951, and Whereas, said approved report contains a recom- mended program of investigations, part of which is set forth as follows : "2. A sum of $450,000 should be allocated from funds presently available for installation and one year's operation of a field experimental project to investigate the hj'draulic feasibility of creating a pressure ridge in confined aquifers by means of injection wells and the effectiveness of such a ridge in preventing sea-water intrusion. This project should be undertaken in the vicinity of Manhattan Beach in West Coast Basin, Los Angeles County. The initial installation should comprise five injec- tion wells and sufficient observation wells to yield the observational data necessary for complete and conclusive interpretation of the results. Capital cost for feeder and distribution pipe lines should be kept to a minimum and emphasis placed experimental techniques and collection of pert data so that the results and conclusions there will be applicable on a State-wide basis. "3. Additional funds should be allocated t Manhattan Beach field experiment if results o first six months operation analyzed in conjun with laboratory research studies indicate tha initial installation of five injection wells is nc tensive enough to yield conclusive results." and "7. The State Water Resources Board si carefully supervise the planning and executic all experimental work in order to assure tha results obtained therefrom are interpretable usable on a State-wide basis." Whereas, the Board is authorized by Chapter '. Statutes of 1951, to cooperate and contract witl Los Angeles County Flood Control District, the " Basin Municipal Water District, and/or any c public or private corporation or agency for the pose of making such investigations ; and Whereas, the Board desires the District to ui take certain portions of the investigational work study under the direction and supervision of the !: Engineer, acting as Engineer for the Board ; Now Therefore, in consideration of the prei and of the several promises to be performed by as hereinafter set forth, the Board and the Dis do hereby mutually agree as follows : ARTICLE l-PROJECT Description of Work The project will consist of providing plans, f( ulating procedures, and conducting investigal which will comprise but not be limited to the fol ing: A. Preparing plans, specifications and cost mates for furnishing and installing: SEA WATER INTRUSION IN CALIFORNIA 11 1. Approximately 7.300 feet of pipe lino Iroui the Metropolitan Water District vault at Manhattan Beach Boulevard and Redondo Avenue in the City of Manhattan Beach ■westerly to the Santa Fe rijrht of way at Manhattan Beach Boulevard, ISIanhattan Beach, California; 2. Pipe lines for distribution alour)4, and June 30, 1954, the project was operated by general funds of the District with funds for tlcf raying water costs supplied by the West Basin Water Association through contributions and assessments in the West Basin area. After July 1, 1954, Zone II of the Flood Control District provided funds, through ad valorem taxes, to I)urchase water for continued operation of the barrier. District general funds were again used for operating expenses. CHAPTER II PROJECT FACILITIES A. PROJECT PLANNING General plamiiiig lor iho proji'ct bopan in October, 1951 with the ratification of the contract with the State Water Resources Board. Preliminary plans had been prepared and were included in recommendations in the 1951 report on the Manhattan Beach well re- charge test. Studies were made of water supply as it concerned length of necessary pipelines and of location of re- charge wells. Other considerations relative to the lo- cation of the recharge line were : the amount of re- charge water needed ; possible waste of recharge water to the ocean ; availability of right of way for drilling recharge •wells ; and possible routes of an extension of the recharge line. It wa.s decided that tlie most logical location for the recharge well line would be along the Atchison, To- peka and Santa Fe Railway Company right of way, paralleling the coast along Ardmore Avenue in Man- hattan Beach about i mile inland. The general loca- tion of the project in Los Angeles County is shown on Plate 1 and the detailed layout of facilities is indi- cated on Plate 3. Following this decision, plans for the transmission pipeline proceeded rapidly. A study of utility services on two alternate routes indicated that the most favor- able pipeline location for delivery of fresh water for recharge would be on Manhattan Beach Boulevard. Many permits for access and land use, in lieu of purchasing right of way, were obtained at no cost to the project. Principal among these were those obtained from the A.T. & S.F. Railway Company for the use of its right of way for laying i)ipeline and drilling wells; from the City of Manhattan Beach for laying pipeline; and the cities of Manhattan Beach and Hermosa Beach for drilling observation wells. Pri- vate individuals also cooperated in providing permits for drilling observation wells on their property. Agreements were made with the City of Manhattan Beach, California "Water Service Company, General Chemical Company, and the Standard Oil Company permitting the District to make field and laboratory water quality studies of certain key water wells. At the request of the State "Water Resources Board, negotiations for waiver agreements, holding the State harmless of liability, were initiated in March 1952 with certain property owners in the vicinity of the test site whose wells were possibly subject to damage due to acceleration of saline intrusion of ground water as a result of recharge experiments. Processing of these agreements delayed drilling operations, then in progress. The last of these agreements was completed by the California "Water Service Comjiany on July Ki, 1952. Agreements with Standard Oil Company of Cali- fornia, the City of Manhattan Beach, and the ISfe.ssrs. John, Philip G., and John Shaw of Hermosa Beach had been completed previously. B. APPURTENANT PROJECT INSTALLATIONS Major installations for the project consisted of the laying of a pipeline and the drilling of recharge and observation wells. A combination field office and chlo- rination building was also constructed. The transmission line design was prepared by the Design Division of the District. In order to minimize capital outlay, asphalt lined, thin gauge welded steel pipe was used. To provide the water supply estimated to be needed for the test, and to allow for the contingency of a possible additional required supply, the pipe line was designed to carry 15 cfs. The contract for the pipeline installation was awarded in the fall of 1952 and com- pleted in March, 1953. The connection with a source of imported water for the test was made in a vault near the intersection of Redondo Avenue and Manhattan Beach Boulevard. At this point the City of El Segundo and the City of Manhattan Beach have connections with the "West Basin Feeder of the Metropolitan "Water District. The transmission line was connected to the City of El Se- gundo line in such a manner as to allow for metering and billing of the test water supply through the facil- ities of the City of El Segundo. The transmission line was laid westerly along the south side of Manhattan Beach Boulevard to its inter- section with the A. T. & S. F. Railway right of way and thence southerly along the railroad right of way paralleling Ardmore Avenue to the vicinity of 28th Street in the City of Hermosa Beach. (See Plate 3) Pipeline construction involved : (a) Installation of 12,098 feet of mechanically coupled, ten and twelve gauge standard asphalt dipped welded steel pipe, valves, and fittings. (b) Jacking of reinforced concrete pipe to house the steel pipe, valves and fittings. (c) Construction of concrete vaults and thrust blocks. (d) Placing approximately 250 tons of pre-mix re- surfacing. The following pipe was installed : 20 inch O.D., 10-gauge 10,000 feet 16 inch CD., 12-gauge 1,034 feet 14 inch O.D., 12-gauge 1.064 feet (31 ) 32 SEA WATER INTRUSION IN CALIFORNIA Because the National Procluctiou Authority's re- strictions upon the use of steel were in effect at the time the transmission line was being planned, the Flood Control District found it necessary to purchase certain materials before the contract was let for the construction of the pipeline. Consequently, the Dis- trict furnished all pipe, fittings, valves and reinforcing steel. The contractor, E. W. Cannell of Gardena, Cali- fornia, furnished all other materials, equipment and labor. C. WELL INSTALLATIONS Wells were drilled for the project for two general purposes, recharge and observation. Most of the wells were drilled by "cable tool" methods in order to ob- tain tlie best well logs possible. Drilling of the in- itially-planned group of wells was begun in January, 1952 and completed in January, 1953. This group in- cluded nine 12-inch wells and thirty-six 8-inch wells. Later in 1953, one 12-ineh well (cable tool and ro- tary), replacing one of the original recharge wells, and eighteen 2-ineh wells (combination rotary, driven and jetted) were added to the program. In 1954, four 4-inch wells were drilled by the "rotary" method to provide further control data for the recharge opera- tions. The arrangement of the well pattern is shown on Plate 3. Drilling operations are pictured in Photo 1. well G-13, drilled near the ocean, is shown on Photo 4. From the results obtained in the 1950 well recharge test and because of economic limitations, it was felt that 12 inches would be an optimum diameter for re- charge wells. Of the original nine wells drilled, eight were standard 12-inch cased wells with perforations cut, after drilling was completed, in zones as indicated by drilling logs. The gravel-packed recharge well, E, of the original group was begun by drilling with cable tools in a 20-inch diameter casing and shoe down to the clay cap. A special reaming tool was used to taper the diameter of the hole in the clay cap to approxi- mately 36 inches. The enlarged area below the 20-inch casing was filled with cement grout and the 20-inch casing was then pushed into the grout. Drilling for a 12-inch easing was begun through the cement grout utilizing a 12"xl5"x3" gravel envelope well drive shoe and continuing on down to the required well depth. The annulus between the 12-inch casing and the 18- inch drilled hole was maintained full of gravel as drilling progressed. When it became necessary to replace recharge well I in the fall of 1953, the experience gained during re- charging operations to that date was considered in deciding what type of well to drill. It had been found that a gravel-packed well was far superior in recharge acceptance rate than the other wells. Also, it had been found that the "clay cap" in the vicinity of the re- charge line was easily erodiblo and that it was diffi- cult to form a seal between the clay cap and the cas- ing. To meet these difficulties, the replacement well, lA, was drilled in somewhat the same manner as well E, with modifications as indicated in the following : (a) A 2-inch hole was drilled well into the clay cap to obtain information on location and extent of the clay cap at that location. (b) A 12-inch pilot hole was drilled about 5 feet into the clay cap and reamed out to a 30-inch hole throughout its entire length. (c) The bottom 2.5 feet of the 30-inch hole was un- derreamed to a 44-inch diameter. (d) A 20-inch conductor pipe was lowered into the well to the bottom of the 30-inch hole. (e) While the hole still contained drilling mud and with the conductor pipe sealed at the top so as to be airtight and prevent grout from rising in the 20-inch conductor, the annulus between the conductor and the rotary hole was grouted through a 2-inch pipe tremie placed in- side the conductor, and extending to near the bottom of the hole. Grouting was continued until grout encased the conductor to the ground surface. (f) After grouting, a rotary rig was used to re- move grout remaining inside the conductor pipe. (g) Cable tool equipment was then used to drill and gravel-pack an 18-inch well with a 12-inch casing, utilizing a 12" x 15" x 3" gravel enve- lope well drive shoe to the required depth into the aquifer. Perforations in all the recharge wells, except well E, were made with a Moss Knife. Tlie dimension of cuts ranged from i" x If" to /j" x If". Well E was perforated with a Mills Knife, the cuts being fV" X 2f ". The Mills Knife makes a vertical cut in the casing. (See Photo 2) In contrast, the Moss Knife makes a horizontal slit in the casing and through hydraulic pressure forces the casing above the slit outward into the formation forming a protective louver over the perforation in the casing. This type of perforation is valuable in a predominantly sandy formation because it inhibits the movement of sand from the formation into the well. Following tlie perforation, wells were surged and bailed until sufficiently stabilized to allow satisfactory pump operation. Normal bailing and surging extended over a period of 16 to 18 hours. Pump development was done by installing an 8- iuch well pump of 1500 gpm capacity and continuing pumping and surging until discharge carried onlj' slight traces of sand or silt immediately following three consecutive surgings. Pun)p develoi)nient ex- tended over a period of some 24 hours. Upon com- pletion of pump development each recharge well was continuously pumped for periods of 6 to 12 hours to SEA WATER INTRUSION IN CALIFORNIA 33 conduct transmissibility tests. Pliotos 3 and 4 slunv the development pump, surface spreadintr for dis- posal of pumped water into the dune sands, and the measurement of pumped water during transmissibility tests. During: pumping development, measurement of dis- charge was made by utilizing orifice plates, and dur- ing transmissibility tests a 4' x 4' x 16' weir box, utilizing a V-noteh weir, was installed to provide more accurate discharge data as well as a continuous automatic water stage record. No attempt was made to determine the pumping capacity ol' any of the recharge w-ells, development procedures being limited to removing only sufficient sauds and silts from the gravels of the ground water aquifer to provide au effective filter around the per- forated portions of the casing. All of the original group of observation wells were 8 inches in diameter. This size was chosen primarily due to the fact that it was desirable from the geologi- cal logging aspect to have a moderate sized cable tool drilled well. Also, wells of this size were more ade- quate for use in the sampling program (described hereafter) because the 4-iueh submersible pump could be readily used inside the 8-inch casing, and permitted a conductivity cell to be raised past the pump in the well in order to collect data on con- ductivity under pumping conditions. Perforations in all the 8-inch observation wells were cut with a Mills Knife. The common perforation was lY' X If", six cuts to the round, staggered on 1-foot centers, variations from this standard being determined by the size and depth of gravels en- countered. Bailing and surging for a period of 4 to 8 hours proved to be adequate development for observation and sampling. In the one instance where the develop- ment was inadequate, air-jet development utilizing a f" airline with a 2-ineh return pipe was successfully completed. Extensive formation sampling was done, as drilling operations progressed, to determine the geologic and hydraulic characteristics of the underlying sediments and to select the most permeable zones for casing per- forations. Twenty-five pound bulk samples were taken from the bailer, and drivecore samples were taken in selected zones with a 1^" diameter Dames and Moore core barrel (See Photo 2) or a 5-inch District vacuum type core barrel. The samples were forwarded to the District laboratory and the Division of Water Re- sources for screen analysis and permeability deter- minations. Ground water samples were also taken from the bailer at approximately ten-foot intervals during drilling except in the more permeable zones, where sampling frequency was increased to four or five-foot intervals. The bailed water samples were analyzed in the ficlil for chloride salinity, and i'nv carbonate and bicarbonate constituents. Significant water samples were forwarded to the District laboratory for com- plete cliemical analysis. Bacterial samples were also taken and forwarded to the District's consultant, Dr. Carl Wilson. After recharge operations were begun, eighteen 2- inch obsei'vation wells were drilled adjacent to the recharge wells for purposes of measuring water sur- face above and below the clay cap, and in February, 19.54, four 4-incli wells, financed with District funds, were drilled for both water surface level observation and pumping samples at the internodal points, mid- way between recharge wells. Where it was not important to log the hole, the 2-inch wells were drilled by the rotary method with mud until the zone of interest was approached. These wells were then completed by driving and jetting pipe so that core samples might be obtained. The driven and jetted wells could not be pre-perfo- rated. Comnunieation with the water table was estab- lished by drilling ahead of the 2-inch casing and then back-filling the open hole and the bottom 2 feet of pipe casing with pea gravel, or by drilling ahead, placing a 1-inch pipe with a screened well point within the 2-inch casing and gravel-packing around the well point. Both types proved to be satisfactory; however, both required frequent flushing to maintain reliable communication. The open bottom pipe with gravel proved the most practical, as initial installa- tions were more economical, and also because the wells were better suited for flushing operations. The 4-inch wells were rotary drilled using an 8-inch rotary bit and cased with a 4-inch pipe casing. Per- foration was accomplished by including the 4-inch pipe casing sections of pre-perforated 4-inch pipe with 6 staggered rows of machine-cut slots. 125 mesh, 3- inch long, on 6-inch centers. These wells were pro- vided with a gravel envelope merely by placing gravel through the annular space outside the pipe casing prior to development. Sufficient gravel was added through the drilling mud to fill the annular space up to the "clay cap" and cemented off at this point. The grout was placed through a pipe tremie. Development was accomplished by injecting air under 85 pounds pressure to the bottom of the well through ^-inch pipe. The maximum pumping rate dur- ing development of the 4-inch wells ranged from 60 to 90 gpm. Development period was approximately 16 hours for each well. D. WELL HEADER ASSEMBLIES Recharge well connections to the feeder line were made with standard 6-inch pipe and fittings in order to provide ample flow to any particular well, if needed. Included within the connection assemblies were short 2— 5256S 34 SEA WATER INTRUSION IN CALIFORNIA lengths of 6-iucli pipe, elbows and flanges as required, a flexible bolted coupling, a 6-inch gate valve, and either a 6-inch meter or 6-inch flow rate controlling valve which includes a meter. Plate 4 is a section through a typical recharge well assembly and Photos 5 and 6 show the field installation of this equipment. Inasmuch as all recharge wells were adjacent to the pipeline, only minimum lengths of pipe were needed. Flow rate controllers were installed in the assem- blies of five of the recharge wells. They are designed to provide a constant flow under var.ying pressure conditions in the supplying line. Activation of the valve is dependent upon the flow indicated by an adjacent standard propeller-type meter. This meter can be used for totalizing, indicating and recording just as a standard meter. It was found that flow ad- justment mechanism on the controller was not depend- able enough for the requirements of recharging a well and, thus, that they were not useful for the project. Since it is desirable to maintain a constant flow at the injection wells, it is recommended that a pressure regulator be installed in the supply pipe- line if it is subject to any more than minor variations in pressure. x\. 6-inch flow conductor pipe was suspended in each recharge well to a position below the minimum water level in the well easing. For details of the suspension of this pipe, see Plate 4. Back-pressure valves were installed in all these suspended pipes in order to keep the 6-inch pipe full of water and thereby exclude air from the recharge flow. This valve is essentially a piece of 4-inch pipe sliding up and down in a larger piece of 5-inch pipe which is slotted on the lower half, sealed on the bot- tom and suspended from the bottom of the 6-ineh conductor pipe by means of flanges. (See Plate 4) The valve is raised and lowered in the well and actu- ated by means of a one-half inch galvanized pipe or cable extending from the valve to a control assembly at the surface. By raising and lowering the 4-inch pipe sleeve from a control device at the surface, the desired flow rate is obtained. Importance of the back- pressure valves, as related to recharge acceptance rate, was established by the District injection test in the City of Manhattan Beach Well No. 7, wlien water was permitted to plunge downward in the recharge well, thus inducing air entrainment. The water surface in the recharge well rose considerably in that test, indi- cating a build-up of the air-water mixture which resulted in excessive well injection head. This fact was reaffirmed in a test at well E during the barrier test, as described in Chapter V. To collect information upon the detailed changes of the rate of flow into the recharge wells, a totalizing, indicating and recording instrument was attached to the meter of each well. In an operating barrier proj- ect it would be unnecessary to have more than a total- izing meter at each well as adequate control can be maintained through a recorder on a mainline meter. E. FIELD OFFICE AND CHLORINATING EQUIPMENT A wooden frame structure, to house a 20' x 23' field office and a 14' x 11' ehlorination equipment room, was constructed by District force account along the Santa Fe right of way near the corner of Ardmore Street and Manhattan Beach Boulevard in Manhattan Beach. The field office included space for a small lab- oratory and office equipment. A roofed enclosure was provided at the north end of the structure for chlorine tank storage. (See Photo 8) Chlorinating equipment, furnished to the Los An- geles County Flood Control District by Wallace and Tiernan Company on a rental basis, consisted of: (a) Automatic control solution feed master chlorin- ator designed to meter the chlorine gas under a vacuum and automatically control the rate of feed. The chlorinator was capable of deliver- ing from 50 to 2500 pounds of chlorine per 24 hours. (b) Automatic chlorine evaporator, of the hot water type, to convert liquid chlorine into gas and so designed that excessively high pressures could not be developed under any condition of operation. The chlorine cylinder stood verti- cally in a hot water bath and was supported in such a manner as to provide free circulation of water. The evaporator was heated with immer- sion-type electric heating elements. (c) Automatic chlorine shut-off valve of the pilot- operated diaphragm t.ype using air as the pilot medium. The valve was designed to reduce the pressure of the gas, as delivered from the evap- orator, to that required for the best operation of the chlorinating equipment and to prevent reliquefaction occasioned by adverse tempera- ture conditions. The valve was also designed to automatically shut off the flow of gas, should the temperature of the evaporator bath fall below the operating range. (d) Residual chlorine recorder which measured the residual chlorine in a continuous sampling cell and recorded on a 24-hour chart graduated from zero to 20 ppm. (e) Air compressor complete witli motor, tank and appurtenances. (f) Chlorine solution pump complete with 20 II.P. 220/440, 3-phase, 60-cycle drip proof electric motor, magnetic starter and base plate. (g) Pipe, valves and fittings which were rubber- lined. The installation of the major portion of this equip- ment is shown in Photo 9. CHAPTER III GEOLOGY A. GENERAL PROGRAM A geologic investigation was couducted to pi-ovide subsurface geologic data iii the coastal test reach sea- ward of Sepulveda Boulevard where no prior data was available. Resultant data constituted a basis for determination of the component materials, extent and thickness, and other pertinent characteristics of the affected major aquifer which controlled recharging, and for conclusive interpretations of the hydraulic and water quality studies. The following constitutes the pertinent phases of a detailed report on the Geol- ogy of the West Basin Barrier Test included as Ap- pendix A, The program involved the drilling, coring, sampling and logging of nine injection and thirty-six observa- tion wells, located on Plate 1 of Appendix A. Some eighteen 2-inch and four 4-inch test and observation wells were added later during the test. Simplified graphic descriptions of subsurface deposits en- countered during drilling may be noted on geologic ■sections 1-1, 4-4, 14-14, C-C, G-G, and K-K. (Plates '.6, and 7 of Appendix A) In substance, drilling disclosed, at the base of the sand dunes, an extensive relatively impervious stra- tum called the "clay cap," which was underlain by a confined aquifer approximately 110 feet in thick- ness. The confined aquifer was found to be in con- tinuity with such important prolific inland aquifers as the "200-foot sand," the "400-foot gravel," and the Silverado zones. These important inland aquifers combine or merge as they approach the coast, and are then termed the "Merged Silverado Zone." Fine grained sediments underlying the Merged Silverado constitute the lower boundary of the aquifer. A zone of sharp flexuring or faulting noted seaward of the test site may act as a partial barrier to sea water intrusion. This is corroborated by the change of pie- zometric gradient indicated from measurements of water levels prior to recharge at observation wells located seaward of the recharge line. (See Plate 15) B. NATURE AND EFFECTS OF THE SAND DUNE DEPOSITS In general, three significant horizons were noted within the sand dime deposits overlying the clay cap : a localized snrficial "iron-bound" sand horizon, an intermediate horizon of relatively clean dune sands with occasional gravels in the basal portion, and a lower horizon of fine sands and silts with sandy strinirers which constituted a zone of transition to the "clay cap." Two-inch test wells adjacent to the injection wells were bottomed above the "clay cap" but within the lower horizon of the sand dunes to determine leakage from the aquifer. Subsequently, shallower test holes were drilled near the injection wells to check the levels observed in the original two-inch wells. These shallower wells confirmed the belief that recorded high water surface levels in the original test wells reflected semi-confined or partial pressure levels within the zone of transition to the "clay cap" rather than free water levels. Hence, leakage from the aquifer to sands lying above the clay cap was probably limited to that transmitted by the sandy stringers within the zone of transition to the clay cap. Inasmuch as the stringers were of minor thickness, flow through the stringers was doubtless equally limited and hence the leakage above the clay cap during the test is consid- ered to be of minor significance. C. EXTENT, NATURE, AND LIMITATIONS OF THE AQUIFER CAP Logging showed, as anticipated beneath the sand dune cover in the test reach, the presence of a rela- tively impervious stratum called the "clay cap," averaging 20 to 30 feet in thickness. (Plate 4 of Ap- pendix A) The upper surface of the cap varied in elevation from about 10 feet above sea level to 10 feet below sea level, (Plate 2 of Appendix A) and the lower surface from about 10 feet below sea level to 40 feet below sea level. (Plate 3 of Appendix A) On the basis of available data, a previous geologic study in the area by another agency assumed that the relatively impervious stratum extended some distance seaward. "Were this true, the underlying aquifer would have been exposed to sea water at some distance from shore. Test drilling, however, revealed the absence of the cap along the strand and at one point 800 feet inland, at test well K-fl. The stripping of the "clay cap" is of significance in that it reduced the distance of travel of ocean waters and hence hastened saline encroachment at these points. The stripping may have retarded, but did not prevent pressurization of the aquifer during recharge since a continuous effective clay cap exists along the recharge line and seaward of it for some distance. The aquifer cap was found not to be a true compact elaj% being composed of vari-colored silts, sandy (35) 36 SEA WATER INTRUSION IN CALIFORNIA clays, and clays containing silty fine sand stringers. These deposits, typical of many coastal reaches of the State, obviously are not as impermeable as a true clay body and are subject to some degree of erosion if water is permitted to move at excessive velocities along or through the cap. Hence, the annular space between casings and side walls of injection wells must be properly sealed throughout the cap to pre- vent rupturing induced by sudden changes in injec- tion rates, with consequent failure of repressurizing operations. The cap failures experienced at injection wells C, G and I were probably due to the creation of voids produced by overdevelopment and excessive leakage past the clay cap at the well easing. In con- trast, the continuous successful operation of wells E and I-A, which were gravel-packed and grouted, would indicate the benefits of this type of well con- struction. D. GENERAL PARAMETERS OF THE AQUIFER AS RELATED TO RECHARGE The Merged Silverado Zone is the major prolific aquifer which lies beneath the clay cap but above the fine, compact blue-gray portion of the lower San Pedro formation constituting the lower boundary of the aquifer. The zone is composed of an upper brown phase, an underlying gray phase, and the basal Silver- ado, all in hydraulic continuity. The uppermost por- tion of the brown phase consists primarily of yellow- ish brown sands and silts, and the lower portion of gravel stringers, with occasional clay bands. The up- permost portion of the underlj-ing gray phase consists of gravel, which grades progressively downward to very fine silt}- sands and clay bands. The basal Silver- ado is composed largely of sand and scattered gravel. The dimensions, pattern, and extent of well perfora- tions were determined by the changing character of the aquifer sediments. The presence of fine sands and silts within the aquifer, particularly within the upper brown phase beneath the aquifer cap, requires care in well development to obviate excess removal of the supporting fine sediments. The Merged Silverado Zone generally thickens ir- regularly Avith distance down coast (southerly) and inland from the ocean. (Plate 10 of Appendix A) This was confirmed by the original ground water grad- ient which also receded with distance down coast and inland in the vicinity of the test site, indicating greater transmissibility. Along geologic sections C-C, G-G and K-K, (Plates 5 and 6, Appendix A) extend- ing in a direction transverse to the recharge line, the depth is somewhat variable, being in general deeper inland than toward the south. Although the thickness of the aquifer varies greatly both seaward and land- ward along the recharge line (Section 1-1, Plate 5, Appendix A), the zone averages about 110 feet and the bottom elevation averages about 130 feet below sea level. The thickness along section C-C is about 90 feet, with the average elevation at the bottom beinu' about 1.10 feet below sea level. The thickness along G-G averages about 100 feet, and the bottom eleva- tion about 140 feet below sea level. The average thick- ness along K-K is about 120 feet, and the elevation of the bottom is about ICO feet below sea level. The thickness of the merged aquifer rapidlj^ increases along the coast southerly of injection well K towards Redondo Beach and generally decreases northerly of iujection Well C towards El Segundo. The necessity for stemming sea water intrusion within the Merged Silverado Zone along the coastline lies in the fact that this zone bifurcates inland into several important water-bearing members; as for ex- ample, tlie upper brown zone, a correlative of a pro- ductive inland aquifer designated the "200-foot sand" zone. A coincidence is evident in the marked thinning of Merged Silverado sediments at the coast (Sections C-C, G-G, K-K, Plates 5 and 6, Appendix A) and the remarkable linearity and parallelism of the topo- graphic features of provinces I and II with the present shore line, (Plate I. Appendix A). This coinci- dence suggests the posibility that deposition of sedi- ments was, at least in part, controlled by sharp flexur- iug or faulting parallel to the coast. Such a zone of sharp flexuring or faulting may act as a partial bar- rier to sea water intrusion, and is evidenced by the steep piezometric gradient, as indicated by oceanward observation well water surface levels noted prior to recharging (Plate l.^) and by dift'erentials in eleva- tions of stratigraphic surfaces zoned by foraminiferal assemblages, as noted in Appendix A (p. 83). This pattern of thinning of aquifer sediments with aj)- proach to a fault-controlled coastline is evident along other coastal reaches of the State and is of significance in that, where applicable, geologic delineation of the zone of thinning can define the most hydraulically effective and economical route for injection wells, pro- vided right of way acquisition is economically feasible. Effective injection within the thinnest portion of the aquifer, in continuity with affected inland welLs, as at the test site, obviously not only checks further major sea water intrusion M'ith the least possible injection rate but also permits recharging of the ground water basin with least possible waste oceanward. It is sig- nificant that such recharge can be accomplished in a local area of critically depressed ground water levels which is no longer replenished by a distant forebay area. Contours on both the top aiul bottom of the aquifer indicate several large irregularities, and reveal the bottom to be somewhat more variable than the top. (Plates 8 and 9, Appendix A) These surfaces indicate channeling transverse to the present shore line. An ancient channel transverse to the present shore line, I SEA WATER INTRUSION IN CALIFORNIA 37 extemliug iulaud from the recliarpie liue toward the abamloued Manhattan Beacli well field, is evident in the reach between injection wells G and I. Tiiis fea- ture, coupled with the absence of tiie day cap near this point at well K-9, 1000 feet west of the recharge line, has probably permitted a more direct contact of the Merged Zone with sea water at or near the strand rather than at some distance from the shore line as presumed in earlier investigations. Saline intrusion of the original Manhattan Beach well field probably was accelerated by existence of the channel and a steep- ened hydraulic gradient created by an accelerated drawdown of the nearby, now abandoned, well field. (See Plate 3) The Jlerged Zone is considered to be laid down largely under continental and .'^hallow marine condi- tions and hence it is to be expected that the nature of the deposits would vary greatly, both areally and with depth. This is evident from an inspection of the graphic well logs presented on Plates 5, 6 and 7 of Appendix A. This variation of course had a profound effect on the permeability and thickness of the re- charged aquifer. While the thickness of the Merged Silverado Zone is considerable, those portions having a high permeability were found to be considerably less in thickness. Obviously then, the aquifer was not ho- mogeneous and, despite the fact that the deposits were in hydraulic continuity, resultant variations both in vertical and lateral permeability doubtless constituted one control of the time lapse noted following changes in injection rate to effect a pressure mergence at the pressure mound internodal points. Recharging of the non-homogeneous aquifer was thus most effectively and expeditiously accomplished bj- the use of gravel- packed wells such as E and I-A. E. NATURE OF THE LOWER BOUNDARY OF THE AFFECTED AQUIFER Beneath the jMerged Silverado aquifer was an ex- tensive thickness of dark bluish-gray, very fine sands, silts and clays, which generally became more compact with depth. These relatively tight sediments consti- tuted the lower boundary of the aquifer inasmuch as they significantly restricted downward movement of waters from the overlying aquifer. Obviously, the finer grained sediments underlying the merged zone are not impermeable but relatively so ; hence, salinity intrusion within these sediments is of relatively minor concern. Transmission of pressure effects from the Merged Silverado zone to the underlving sediments eventually occurs within those portions having at least some degree of hydraulic continuity with the Merged Zone. F. SEISMIC EFFECTS Occasional earth shocks were noted on observation well charts. No damage to facilities was observed, how- ever, and noted effects on aquifer transmissibility were, as far as known, only temporary. G. RESUME Obviously the detailed geologic data collected was vital to the conclusive interpretations of the hj-draulic and water quality studies, and to the operation of the Ban-ier Test. More specifically the exploration deter- mined such vital factors as : (a) Effectiveness of the aquifer cap, so as to pro- vide background for such construction and maintenance as was necessary to permit con- tinued pre.ssurization. (b) Physical limits and homogeneity of the aquifer in relation to the validity of the hydraulic in- terpretations. (c) Extent and variation in thickness of the aquifer along the line of recharge, and con- tinuitj' of such deposits with correlative inland aquifers affected by sea water intrusion. (d) Geologic structural controls which affected the movement of ground water as related to the establishment of a fresh water barrier to sea water intrvision and as related to the intrusion via ancient coastal channels prior to recharge. The current detailed investigation should provide a valuable supplement to the basic geologic elements outlined during prior investigations of this area by the United States Geological Survey in 1948 and sub- sequently by the State Division of Water Resources for the Report of Referee on the West Coast Basin. This detail covers information on a reach more sea- ward than the area covered by the previous studies. In conclusion, it is significant to note that, during the exploration program, geologic conditions were encountered that were directly pertinent to the con- struction of the wells and are fundamental to the methods employed in the creation of the barrier. Hence, it may be predicated that the creation of a barrier along other coastal reaches of the state should be preceded by sufficient exploration to establish a sound basis for design and operation. CHAPTER IV CHRONOLOGY OF RECHARGE OPERATIONS A. OPERATIONAL PLANNING Initial plans concerning recharfcc procedures in- volved the use of five 12-inch diameter injection wells stationed at intervals of 1,000 feet along the A. T. & S. F. Railway right of way. The use of four inter- mediate 12-iueh wells was reserved for optional injec- tion should it later be desired to reduce the well spacing front 1,000 feet to 500 feet. The plans also provided for intial recharge to occur at the center- most well, well G to be followed in order by consecu- tive recharge starts at well pairs, (1) E and I, and [■2) C and K. The following table indicates the preliminary re- charge schedule with particular regard to proposed well rates and time intervals between well starts : Time from Start Proposed Well Rates (cfs) (in treeks 1 2 4 6 8 WellCr 0.50 Incr. to 0.75 Incr. to 1.00 1.00 1.00 WelU E, I 0.0 0.50 Incr. toO.75 Incr. to 1.00 1.00 WeUs C, K 0.0 0.0 O.r.o Incr. to 0.75 Incr. to 1.00 Through approximate adherence to the above indi- cated schedule, it was felt that lateral mergence of individual well cones would be more uniform and that gradual build-up of the fresh water mound would enable data of greater significance to be obtained. In conjunction with injection, considerable effort was expended upon the obtaining of data regarding piezometrie surface and water quality. In order to provide the former information, 10 continuous water level recorders, supplemented with 9, furnished by the District, were installed in key observation wells. Additional water level data at the remaining wells were obtained from frequent tape measurements made at regular intervals. Basic data on the subject investigation, as indexed in Appendix F, is filed at the District's office. Information on ground water quality was provided by a comprehensive sampling program. This consisted primarily of obtaining conductivity traverses and pumped water samples from observation wells. B. PERIOD OF MOUND BUILD-UP, 5 WELLS History of operations at each recharge well is shown on Plates 6 to 13. These include the water surface elevation in the recharge well, the ground water elevation in the 20-foot observation well, the injection rate and the chlorination rate. Profiles of the ground water elevation along the recharge line for several dates are drawn on Plate 14 and profiles normal to the recharge line through well G on Plate 15. February, 1953 Recharge was first begun with a three-hour trial injection run at well G on 2/14/53. Recharge and chlorination rates during this test period were 0.50 cfs aud 20 ppm respectively. Installation of the main- line Wallace & Tiernan Companj' chlorinator was not completed until 3/24/53 due to delays in shipment of necessary parts. Until this date, chlorination was effected through use of a small-capacity portable chlorinator located at well G. This delay postponed injection at wells other than G until 3/25/53. Continuous recharge of 0.5 cfs was initiated at the eentermost injection well G on 2/24/53. Subsequent increases in recharge rate were made at this well after intervals of operation of approximately one week each. Rate of movement of this fresh water was closely timed to the nearer observation wells. Evi- dence of arrival at adjacent observation well G-1-20 feet distant — was first detected some 5^ hours after injection began. In turn, the saline wave preceding the injected water was first observed at well G-2, located 250 feet landward 17 days later on 3/13/53. March- April-May, 1953 On 3/13/53, just two days after an injection rate of 1.00 cfs was reached at well G, a surface eave-in of considerable proportions occurred adjacent to this well. This resulted in a radical lowering of acceptance rate and it became necessary to reduce inflow in order to keep the injection head below ground surface. This occurrence is discussed in detail in Chapter VII. On 3/25/53 recharge of 0.5 cfs was initiated at each of wells E and I. In order to build up a more complete pressure barrier, injection at well E was increased to 0.75 cfs on 4/7/53 and to 1.00 cfs on 4/23/53. A definite decrease in chloride salinity was fir.st detected at observation m'cU E-4, located 500 feet landward of recharge well E, on 4/22/53, some 28 days after injection was initiated at the latter well. The average landward pressure gradient between wells E-1 and E-4 during this period was approxi- mately 0.010 feet per foot. The total water injected at well E by this date was 34 A.F. Freshening was estimated to have first begun at landward observation well 1-4, located 520 feet land- (39) 40 SEA WATER INTRUSION IN CALIFORNIA ward of well I, on about 7/7/53, 104 da.ys after iu- jection was initiated at well I. Average landward pressure gradient between wells T-1 and 1-4 during this period was approximately 0.01 2() feet per foot, while total water injected at well I was 91 A.F. Pressure mound build-up was extended laterally by initiating recharge at wells C and K on 4/30/53 and 4/28/53 respectively. Recharge at the former well was discontinued on 5/4/53 due to apparent excessive leakage upward past the clay cap as indicated by test holes drilled to the clay cap. On 5/10/53, mainline chlorination was reduced from 20 to 15 ppm. No adverse effect upon well ac- ceptance was noted. (See Plate 18) On 5/27/53, subsidence occurred at injection well I with the same general characteristics and results as previously observed at well G. In order to keep the injection head below ground surface, the injection rate was reduced from 0.5 to 0.2 cfs, and injection at a rate of 0.2 cfs or more was continued until October 14, 1953. No further subsidence has been noted at this well. June, 1953 Recharge at well G was again resumed on 6/9/53 at 0.2 cfs after being inoperative from 4/12/53 to 6/8/53 for purposes of rehabilitation. Subsequent in- creases in rate were made at this well until 6/16/53, when an injection rate of 0.75 cfs was reached and maintained. On 6/17/53 mainline chlorination was reduced from 15 to 12 ppm. No adverse effect upon well ac- ceptance was observed. (See Plate 18) C. PERIOD OF MOUND BUILD-UP, 8 WELLS In order to relieve the pressure against the clay cap at the four operating recharge wells, plans were made to iitilize the existing intermediate 12-incli re- charge wells (D, F, H and J). This would permit a reduced injection rate and injection head at each individual well while the total flow required to main- tain an adequate pressure mound could be sustained, but with a well spacing reduced from 1000 to 500 feet. Injection at well H was initiated on 6/20/53. The initial rate of 0.5 cfs was increased on 6/24/53 to 0.75 cfs. July- August, 1953 AVitli regard to movement of injected water land- ward from well K, decrease in chloride salinity is estimated to have first occurred at observation well K-4 on about 7/15/53, 78 days after injection was initiated at well K. Average landward pressure gradient between wells K-1 and K-4 during this period was 0.0122 feet per foot, while the volume of water injected at well K was 66 A.F. On 7/29/53 mainline chlorination was reduced from 12 ppm to 10 ppm. No adverse effect upon well ac- ceptance was noted. (See Plate 18) On 7/30/53 increases in injection rate were made at wells G, H and K from 0.75, 0.75, and 0.50 cfs to 1.0, 1.0, and 0.75 cfs, respectively. Sepf ember, 1953 On 9/2/53 injection at well H was reduced from 1.0 cfs to 0.6 cfs in order to decrease the injection head. Subsequently remedial work was undertaken at this well as discussed in Chapter VII. On 9/11/53 mainline chlorination was reduced from 10 to 5 ppm. Reference to Plate 18 shows a de- crease in acceptance rate indicating that clilorinatiou at the latter rate was insufficient to maintain pre- existing acceptance rates, although at the time the method of analyzing this data was not sensitive enough to indicate that fact. On 9/16/53 mainline chlorinator was placed on au- tomatic control. Through an interconnection with mainline flow meter, changes in chlorine addition (pounds/24 hours) thereafter were directly propor- tional to any changes in the water flow in the pipe- line. On 9/21/53 recharge of 0.3 cfs was initiated at intermediate well F. Further increases in flow were made in small increments on succeeding days until 9/23/53, at which time injection of 0.5 cfs was reached and maintained. Similarly, injection at well J was initiated at 0.3 cfs on 9/29/53, followed shortly thereafter by small increases in flow until injection of 0.5 cfs was reached on 9/30/53. October-November, 1953 Recharge at well H was discontinued on 10/5/53 in order to effect the remedial work at this well. By the time of shutdown, the indicated surface level of water above the clay cap, as indicated by well H-T-1, had risen to +48 feet M.S.L. Injection of 0.2 cfs was initiated on 10/6/53 at well D. Additional small increases were made until 10/7/53 when a recharge rate of 0.5 cfs was reached and main- tained. D. PERIOD OF MOUND MAINTENANCE, 8 WELLS Injection at well I was discontinued on 10/14/53 in order to complete drilling and development of gravel-packed replacement well I-A. Injection was initiated at the latter well on 11/10/53. The initial recharge rate of 0.25 cfs was increased in small in- crements to 0.5 cfs. Initial acceptance was quite satisfactory. During the period from 11/12/53 to 11/23/53, the elevation of water surface in well I-A I SEA WATER INTRUSION IN CALIFORNIA 41 rose from +7 feet to +11 foot M.S.L. while iiijeetion 15 ppm for this brief period caused no immediate approximated 0.5 cfs. (See Plate 11) The net rise in discernible effect. ■water surface was 4 feet, compared to a net rise of 55 feet at well I lor a like period and rate. February, 1954 Initial injection rate at all wells was the same, 0.5 On 2/25/54 mainline chlorination was reduced from cfs. At this rate, an unbalanced barrier became ev- 5 to 3 ppm in or.ler to determine longr-term effects ident as stability was approached. Individual well of lowered chlorination rate. The trend of lowering flows were adjusted to more nearly balance the re- acceptance at well J prompted a reduction in recharge charge barrier. On 11/24/53 the foUowing adjust- from 0.63 to 0.52 cfs at this well on 2/26/54 in order ments in well Hows were made : to keep the injection head below ground surface. Previous Injection New Injection The average length of ridge extending to an ele- Weii Rate— cfs Rate— cfs vation equal to or above -f 2.5 feet M.S.L. during the D 0.5 0.2 month approximated 3,600 feet and above sea level F IIIIIIIIIIIIIIIIIII o!5 OA for 4,800 feet. G 0.5 0.4 j^ order to provide more complete data on the I-A -IIIIIIIIZIIZZIIIII 0.5 1.0 shape of the pressure ridge, additional 4-inch di- J 0.5 0.7 ameter test wells D-E, E-F, H-I, and I-J were drilled This change created a minimum pressure mound during this month. equal to or higher than -|-2.5 feet elevation for 1550 lo'i/f feet along the line of recharge and above sea level March, for 4100 feet. Theoretically a fresh water head of 2.7 Acceptance rate remained nearly constant at all feet above sea level is required to balance the more injection wells except J. Due to the apparent con- dense sea water, assuming an aquifer extending to tinuation of lessening acceptance at the latter well, 110 feet below sea level. it was again necessary to reduce the flow at this well. The reduction from 0.52 to 0.45 cfs was made on December, 1953 3/16 54. Mainline chlorination was maintained at 3.0 oi- w , 1 J • -n. u -in^o ppm throughout the month, blight changes were made during December, 19o3 t f o in order to further modify extremes in level and raise The average length of pressure ridge maintained the average height of the pressure ridge. Average during the month at an elevation equal to or above length of the pressure ridge maintained equal to or +2.5 feet M.S.L. approximated 3,100 feet and above higher than elevation +2.5 feet M.S.L., during the sea level for 4,500 feet. period, approximated 2,450 feet and above sea level for 4,150 feet. The changes consisted of the following: April, T954 Previous Injection New Injection Mainline chlorination was reduced to 1.5 ppm on Dale Well Rate — cfs Rate— cfs 4/1 /,54 and maintained at this rate during the month. 12/17/53 I-A 1.00 0.75 _^t; all injection wells other than I-A, consistently de- 12/30/53— _ D 0.25 0.50 creasing acceptance was observed. At wells D and J , igcj temporary chlorination in excess of 1.5 ppm was °""°'^^' achieved "throudi the use of a portable chlorinator Recharge line pressure levels were further increased located at each well site. Chlorination of approxi- on 1/4/54 by adjustments of flow rates at the follow- ^^^^^^^^ ^^ pp^^^ ^.^^ maintained at well J from 4/1/54 mg wells: D ■ , ■ ,. >r r • <• to 4/7/54 and chlorination of 20 ppm from 4/14/54 Previous Injection New Injection '^^ t/i/.j-r ciiiii -. i t-r Weii Rate— cfs Rate— cfs to 4/21/54. Chlorination at this latter rate was main- E 0.4 0.5 tained at well D from 4/7/54 to 4/13/54. A tempor- G IIIIIIIIIIIIIIIIZIIII o!4 0^(5 ary improvement in acceptance was noted at wells D H 0.4 O.G anji J during these periods of relatively higher chlor- The average length of ridge equal to or higher ination. Although no sustained benefit was noted, it than elevation +2.5 feet M.S.L. during the month clid indicate that chlorination at 1.5 ppm was in- approximated 3,550 feet and above sea level for 4.700 sufficient to maintain pre-existing acceptance rates at feet. Chlorination rate was temporarily increased these wells. (See Plate -.0) from 5 to 15 ppm on 1/26/54 in order to observe pos- Due to a lowering acceptance rate, recharge at well sible effect upon acceptance rates of recharge wells. D was reduced from 0.50 to 0.43 cfs on 4/6/54, while Pre-existing rate of 5 ppm was again resumed on that at well K was reduced from 0.69 to 0.61 cfs on 1/29/54. As evidenced by Plate 18, chlorination of 4/11/54. 42 SEA WATER INTRUSION IN CALIFORNIA Additional changes were made at the following wells on 4/21/54 in order to obtain a more uniform barrier: Previous Injection New Injection Well Rate-cfs Rate-cfs D 0.43 0.34 E 0.50 0.45 F 0.58 0.53 I-A 0.80 0.90 J 0.42 0.45 The average length of pressure ridge extending to an elevation to or greater than -|-2.5 feet M.S.L. dur- ing the month approximated 2,600 feet and above sea level for 3,700 feet. May, 1954 Injection at well J was discontinued on 5/5/54 in order to carry out plans to redevelop this well as dis- cussed in Chapter VII. The injection rate at gravel-packed well E was raised in small increments in order to determine the maximum possible acceptance rate into this type of well. A maximum flow rate of 1.86 cfs was reached at this well on 5/12/54. An increase of mainline chlorination was made on 5/8/54 from 1.5 to 5.0 ppm in order to effect an im- provement in well acceptance. Reference to Plates 18 and 19 shows that this rate was not completely effective. In an effort to determine what eft'ect injecting an aerated water supply might have upon well accept- ance, aeration at gravel-packed well E was initiated on 5/14/54 and continued until 5/21/54. Further dis- cussion appears in Chapter V. The following changes in recharge rate were made on 5/25/54 in order to offset reduction in total re- charge caused by well J becoming inactive on 5/5/54 : Previous Injei ction New Injection Veil rate - cfs Rate- cfs D 0.4 0.55 H 0.65 0.75 I-A 0.85 1.10 The average length of pressure ridge maintained dur- ing the month extending to an elevation equal to or greater than +2.5 feet M.S.L. approximated 2,200 feet and above sea level for 3,400 feet. June, 1954 Injection at well J was resumed on 6/7/54 at 0.2 cfs. Subsequent increases in flow were made at this well until 6/8/54 when an injection rate of 0.4 cfs was reached and maintained. In order to create a pressure ridge of more uniform height, the following adjustments to individual well flows were made during the month : Date Well 6/2 /54 D 6/18/54 D 6/18/54 E 6/18/54 F 0/18/54 G 6/18/54 J Previous Injection Rate - cfs 0.45 0.3 0.3 0.5 0.5 0.4 Xcir Injection Rule — cfs 0.3 0.4 0.45 0.6 0.65 0.55 The addition of 2 ppm sodium metaphosphate glass solution to well I-A influent was discontinued on 6/15/54. In order to make more exacting determinations of the rate of movement of injected water in the aquifer, granulated sugar was added to injection well influent as a tracer mechanism on two separate occasions. Three hundred pounds of sugar were added to well I-A on 6/17/54, within a period of 3 minutes, while a flow of approximately 0.45 cfs was being main- tained. Similarly, 300 pounds of sugar were added to well G on 6/29/54 within a period of 12 minutes with an injection rate of 0.6 cfs. The results obtained in either instance were largely inconclusive because of the large amount of dilution which occurs and the resultant low concentration of sugar present in the samples. The average length of pressure ridge maintained during the month extending to an elevation equal to or greater than +2.5 feet M.S.L. approximated 2,450 feet and above sea level for 3,700 feet. A summary of the effect of the recharge operations can be seen on Plates 16 and 17 showing the ground water contours in the project area for February 20, 1953 and June 24, 1954, respectively, and Plate 5 which shows the build-up with time of the ground water elevations at various points in the barrier mound. Comparison of the ground water contours on the two dates clearly show the creation of a pressure mound. The points of measurement of the ground water elevations are : a well on the axis of the recharge line, 20 feet from a recharge well, well G-1 ; a well ap- proximately 500 feet oeeanward of the recharge line, well G-5; a well 1180 feet landward of the recharge line, well G-8 ; a lateral well on the axis of the recharge line to the south, well L-1 ; and a well 11,000 feet landward, a distance at which the landward effect of recharge is undiseernible, well MB-11. Also shown on Plate 5 are the total recharge rate and the total well production of the fields closest to the line of recharge. Total fresh water injection along the recharge line, and the resultant recharge to the basin, amounted to 3520 acre feet from the beginning of the test until June 30, 1954. Producing wells were those of the City of Manhattan Beach and the California Water Serv- ice Company located approximately 6500 feet land- ward of the recharge line. Discussion of the effect of inland pumping occurs in Chapter V, B. CHAPTER V HYDRAULIC ASPECTS OF THE RECHARGE LINE Tho hydraulic jiroinTtii's of a line of reoliarge wells, the parameters involved, and the application of known data relative to the design of a line of recharge wells were the most important factors to be deter- mined in the test. An attempt lias been made in the following discussion to compare observations from this field test to theoretical derivations and model studies concerning the phenomenon of recharfiinp: fresh (or lighter) water into a pressure aquifer in the vicinity of a body of saline (or denser) water. The technical factors to be investigated, which were specified in the agreement concerning the performance of the recharge test, are discussed below. A. THEORETICAL BACKGROUND In order to show the relationship that a single re- charge well has to its position in a line of recharge wells, it is of interest to follow the physical transition which occurs between the beginning of injection and the creation of a stable mound. The following descrip- tion presents the changing conditions as if there were a time lag between occurrences. In a true pressure aquifer this is not the ease, since pressure changes, theoretically, are transferred instantaneously. How- ever, practically all pressure aquifers have a storage factor which has been variously attributed to com- pressibility of the aquifer materials and/or the con- fining strata. Actually no pressure aquifer is perfect in that leakage to a greater or lesser degree occurs through the confining strata or ground water moves to adjacent or contiguous free zones. When injection begins at a recharge well, all the flow from the well is radial, although when the re- charge cone of a single well is superimposed upon an existing ground water slope, the larger portion of the injected water flows down gradient. Radial flow emanating from the well continues along streamline trajectories until the interference effect of adjacent recharge wells begins at a point near the line of re- charge about midway between the wells. Actual mergence, that is, the intersection of the streamline boundaries of two adjacent wells, is gov- erned b}' the ratio of the well recharge rate to the unit rate of flow occurring downstream from the line of recharge, which ratio determines the maximum width which the stream of injected water from any well can approach in a given aquifer. Hence, unless the spacing of the wells is smaller than this width, no intersection of streamline boundaries occurs and, therefore, no mergence takes place. Saline water would then continue to flow between the streamline bound- aries of adjacent wells. Conversely, when the afore- mentioned ratio greatly exceeds the well spacing, more than the necessary fresh water head will be developed at the point of mergence than is required to balance the intruding sea water. This, in course of time and displacement of the saline wedge, will result in an additional waste to the ocean. As the area of pressure due to recharge continues to expand toward stability, the pressure, at a given point, is the resultant of the effect of each well. The integrating effect of this mergence continues until a continuous mound is formed, this condition being characterized by the establishment, within a short dis- tance from the line of recharge, of two-dimensional flow, i.e., no lateral movement of recharge water. After establishment of the recharge mound, the most important factors involved in a line of recharge wells which is acting as a complete stable barrier to sea water intrusion are : (assuming a uniform, homo- geneous, completely confined aquifer). (a) The quantity of fresh water recharge in a given section parallel to the coast line must be suf- ficient to replace the previous quantity of sea water which was intruding. (b) To stabilize an intruding sea water wedge oceanward of the recliarge line, there will have to be some movement of fresh water toward the ocean. In reference to (a) a cursory analysis of the Darcy equation of ground water flow, Q^KiA (\STiere Q = rate of flow, K = permeability coefficient, i = hydraulic gradient, and A = cross sectional area through which ground water is moving) shows that the rate of flow through a given reach of an aquifer is proportional to the hydraulic gradient. In a large pressure aquifer, the gradient is dependent upon the location of supply source and the amount and pattern of pumping w-hich, in general, do not change rapidly. Hence, if sea water is the source of supply, conditions can only be changed by : changing the pumping amount and pattern; and/or providing a new source of supply equivalent to the existing supply. Factor (b) has been reduced to a quantitative equa- tion • • For derivation see Appendix E. (43) 44 SEA WATER INTRUSION IN CALIFORNIA q = y^ (Ss-Sy) ^ T where q = seaward fresh water flow per foot of ocean front Ss = specific gravity of sea water Sf — specific gravity of fresh water M = thickness of aquifer, down to lowest depth which must be protected T = aquifer transmissibility for 100% hy- draulic gradient L = length of sea water wedge, from ocean outlet to the inland toe. The reason seaward flow occurs is, basically, because sea water is more dense than fresh water. This fact requires that a higher fresh water head be maintained in order to balance the pressure in the intruding sea water wedge. A complete barrier can be attained only if a pressure balance in the lowermost part of the aquifer is obtained. At all other higher elevations there will be an oceanward gradient, which fact is responsible for the movement of flow toward the ocean. However, the test establishes the fact that such movement is of no practical or economic significance in tliat the rate of movement of injected water ocean- ward, in relation to its movement inland, is minor and relatively slow under gradients observed. Theoretically, the amount of recharge water re- quired at a given location may be estimated by com- bining the requirements of items (a) and (b) above. The discharge rate of the intruding sea water, which must be replaced, plus the quantity necessary to waste to the ocean to stabilize the sea water wedge eqi;als the necessary stabilized recharge rate. Since, in a com- pletely confined aquifer there is not storage available, the rate of ground water flow depends upon the in- land rate of pumping, which during the period of the test did not change appreciably. The ground water gradient depends upon the rate of flow, and therefore the gradient landward of the recharge wells should be approximately the same prior to and during recharge operations. However, at the test site, it is found that as a result of recharge, the landward gradient was increased some 58% by recharge, indicating that the required recharge rate was greater than the prior rate of sea water intrusion. It is concluded that one or more of the following factors is responsible for this gradient increase, and the corresponding indicated increase of flow landward : (a) Significant storage exists within the local aquifer and/or in some portion of the aquifer contiguous to that being recharged. (b) A significant quantity of leakage is occurring upward through tlie confining layers inland of the recharge line. (c) Lateral flow of the injected water is occurring to such extent that a partial barrier extends over an area much larger than is indicated, as yet, by the arrival of injected fresh water. It is feasible that a complete barrier can be formed within the reach of recharge wells and that a partial barrier is formed off the ends of the recharge line, the degree of effectiveness of the latter decreasing with lateral distance from the exterior recharge well. It may be noted that the effect of the increased gradient is sig- nificantly measurable only about 7000 feet landward. (Compare ground water elevations on Plates 16 and 17) In order to evaluate the effect of the increased gradient, an additional ground water flow factor must be considered. To demonstrate the use of this factor and to give an example of the magnitude of the amount of water which will move toward the ocean, the conditions at the test site can be used. The total recharge water needed, per foot, Q is: M Q = 14 {Ss-Sf) -J-- T Oceanward flow from the recharge line -\-i T Initial ground water flow landward -\-i' T Increased ground water flow landward due to recharge, where i = ground water gradient before recharge and i' = increase in ground water gradient over prerecharge gradient due to recharge operations. The ratio q/Q gives the proportion of injected water which will waste to the ocean if the recharge barrier is operated at the minimum effective elevation. q Q M(.025) 110 2000 (.025) 110 = .050 .0041-1-. 0024 2000 Sg = 1.027 Approximate specific gravity of ocean water obtained off Manhattan Beach pier Sf = 1.002 Average specific gravity of M.W.D. fresh water il/= 110 feet Approximate value assumed for the effective aquifer L = 2,000 feet Distance to ocean I = .0041 Average initial gradient along a normal to the recharge line through well G i'' = • . 0024 Average gradient increase along the well G normal. Although tlie average elevation along the recharge lini- was soniewliat above the minimum elevation due SKA WATKR INTRUSION IN CALIFOKXIA 45 til tlio necessity of iiiaiuta'miiif:' tlu' interuodal point at the minimuni, this fact would not have a Uirge eft'eet upon the quantitative result. The numerical values used in the above equation were estimated as follows: This ratio applied to the average recharge rate gives the approximate amount of recharge water flowing toward the ocean : .05 X 4..") el's = .'22 cfs. 5% of the total recharge flow. Since the inception of recharge, the average rate of movement oceanward has been less than 150 feet per year. At this rate, no actual waste will occur to the ocean within the next 10 years. B. CHARACTERISTICS OF THE OBSERVED PRESSURE MOUND In general, the pressure mound formed by the recharge wells has fulfilled the expectations of the planning and calciilations for such a barrier. It was originally estimated that 1000-foot spacing of the recharge wells would be adequate at the test site. However, it was necessary to reduce the spacing to 500 feet when it was found that the acceptance rate at the 12-inch recharge wells was too low under the vestricted injection head made necessary by the failure of the "clay cap" adjacent to the well casings. On the other hand, this disadvantage was balanced by the fact that a lower than estimated total recharge rate has been found adequate for the reach investigated. Because of the difSculties of creating a seal be- tween the recharge well casings and the claj^ cap, there has been a suspicion that part of the injected water leaked upward into the overlying sand dune layers. Test holes drilled to the vicinitj' of the top of the claj' cap indicate that there may have been some leakage, but it was of such limited magnitude as to have only a minor effect upon the formation of the barrier mound. It is necessary to define certain terminology, at this point, in order to clarify a basic concept which has been established in this test. Injection liead, as used in this report, is the head required in a recharge well to cause the injected water to pass through the well perforations and the face of the aquifer. It is the difference at a given time between the water sur- face elevation in the recharge well casing and the maximum piezometric ground water surface near the well resulting from the existent pressure mound. The injection head is dependent upon the iiuantity of water being injected, the type and size of recharge well, the number of perforations and the local trans- missibility characteristics of the aquifer in the im- mediate vicinity of the recharge well. The injection head is independent of the height or shape of the pressure mound. It is assumed that this head is in- fluenced by I lie oceuri'cnce of turbulent flow through the perforations and in the aquifer inunediatcly ad- jacent to the well. Normal pressure mound elevations are attained when the flow becomes laminar. Hence, the advantage of a large diameter well or a gravel- packed well to reduce the required injection head becomes obvious. Mound elevation is the piezometric surface (or pressure elevation) of ground water at a given point of the mound. The mound's size and shape are de- pendent upon the amount of water being injected in a given reach, the spacing of the recharge wells, the transniissibility of the aquifer, and the ground water gradient as it would be if unafFected by recharge. Observations of the build-up of the recharge mound during the initial period of recharge (See Plate 15) indicated a considerable lag in the development of a stable pressure gradient oceanward of the recharge line. About two months of recharge operations were necessary before the ground water elevation ocean- ward of the recharge line rose above sea level and about six months of recharge were required to de- velop a relatively stable gradient. It is concluded that this lag was the result of a continued by-passing of flow through the internodal pressure valle.ys prior to the development of the entire barrier to sea level and/or the storage effect within the oceanward aqui- fer. Well logs indicated that the confining clay mem- brane oceanward of the recharge line was not con- tinuous and would permit ground water storage. Due to the changes necessitated in recharge rate and to the interval of time between the initiation of re- charge at the individual wells, it is not possible to accurately evaluate the time required to establish a stable landward gradient. However, as recharge was commenced at a single well (G), pressurization effects were noted 1180 feet landwai'd almost instantaneously and, under a constant recharge rate, stabilization was attained in approximately five days. Visual inspection of the piezometric gradient indi- cates that the individual cones occurring around each recharge well merge into a relatively uniform barrier within 250 to 500 feet of the recharge line both land- ward and oceanward. This fact is indicated on Plate 15 by the relatively uniform ground water gradient between wells G-4 and M-B-4 and between wells G-5 and G-13. After the oceanward pressure lag had been over- come, there was no difficulty involved in maintaining the desired mound elevation by maintaining a con- stant recharge rate into the recharge wells. However, the question as to whether a complete barrier to sea water intrusion has been formed is not as readily ap- parent. The most critical spot is, for obvious reasons, assumed as the point midway between the recharge wells, called hereafter the internodal point. At thi.s location the mound elevation is the least, so that if f 46 SEA WATER INTRUSION IN CALIFORNIA ciioug-li fresli water head is not maintained at this lo- cation to overcome the higher density of sea water, then intrusion will continue near the bottom, although at a reduced rate. On the basis of an average elevation of the bottom of the aquifer of 110 feet, it is necessary to maintain a fresh water head of approximateh^ 2.7 feet above sea level at the internodal point. Although geologic interpretation suggests that the bottom of the Merged Silverado may be 10 or 20 feet lower, inspec- tion of well logs indicates relatively impervious sedi- ments in these lower limits of the aquifer. Since the aquifer is not uniform and homogeneous and since the effective bottom of the aquifer is not known, it is of interest to know- whether the sea water is completely blocked or not. An indicator of this factor is the chloride salinity at the internodal wells if they are perforated deep enough into the aquifer to indicate salinity conditions at the "bottom" of the aquifer. The gradient from the recharge wells through the in- ternodal wells is very slight as compared to that in a landward direction from the recharge wells and, con- sequently, the movement of injected fresh water to- ward the internodal points is very slow. The chloride concentration in the internodal wells penetrating the entire depth of the aquifer have continued to diminish slowly during the test. Hence actual freshening of the internodal points may not occur until after several years of recharge. In this connection, it may be noted that if mergence of the flow from adjacent wells takes place oceauward of the recharge line, the internodal wells will eventually freshen. Conversely, if it takes place landward, a pocket of salinity or point of stag- nation may occur, thereby precluding complete fresh- ening at the internodal well. As of June 30, 1954, none of the observation wells at the internodal points had become fresh, although all of them had a decreas- ing trend. (To date, well G-H has decreased from 16,700 to approximately 2000 ppm chlorides.) These indicator wells are : D-E, E-F, F-G, G-H and I-J. The chloride salinity history of these internodal wells is shown on Plate 23. It should be noted that H-I pene- trated only the top portion of the aquifer and may not indicate the true condition of the entire aquifer. Freshening could be hastened by increasing the injec- tion rates and the resultant pressure levels ; however, such a procedure will also increase the rates of move- ment oeeanward. In connection with the chloride salinity history of internodal wells P-G and G-H, the following facts should be noted. During the period from June to Sep- tember 21, 1953 well G was recharged but well F was not. Thus, well F-G was not an internodal well and was largely influenced by injection at well G. From September 21 to October 29, 1953 the injection rates were unbalanced in favor of well G, which means that well F-G was still not exactly the internodal well. But after October 29, 1953, rates of recharge at welLs F and G were sucli that well F-G was the internodal point, and the drop in chlorides which began in March, 1954 represents chloride reduction at the inter- nodal point. A similar analysis of the location at the internodal well applies to well G-H. Injection began at well H on June 20, 1953 and continued at a rate equal to that of well G until September 3, 1953, when it was neces- sary to reduce the injection rate at well H because of an impaired acceptance rate. The injection was stopped at well H on October 5, 1953. After repairs, injection was again started on October 24, 1953 at a rate equivalent to that of well G. The drop in chloride salinity at well G-H which occurred in January, 1954, probably indicated the reestablishment of this well as an internodal point indicator. Pumping rates at the wells of the City of Manhat- tan Beach and the California Water Service are indi- cated on Plate 5. These wells, located about 7000 feet inland from the line of recharge, had no immediate effect upon the piezometric surface between the line of recharge and the pumping wells. However, pvimp- ing of ground water inland obviously contributed to the depletion of the supplies of the basin and results in an increase of the inland gradient, thereby increas- ing the required water supply at the recharge line. Obviouslj^ this effect is slow and cumulative unless the proximity of the pumping well results in an immedi- ate drawdown at the recharge line. The location of observation wells for the test was based on the experience with a single recharge well. Observation wells for the project were drilled: at a distance of 20 feet from recharge wells along the axis of the line of recharge; at the internodal points of the iutially-planned 5 recharge wells ; and at intervals along three separate lines normal to the line of re- charge and through a recharge well. Later, when recharge was reduced to 500-foot spacing by re- charging the former internodal wells, it was found necessary to drill new internodal wells. It is concluded that adequate control for barrier recharge operations can be maintained with observa- tion wells at internodal points, occasional observation wells near the coastline located on lines normal to the recharge line or along probable flow lines and passing through a recharge well, and occasional ob- servation wells inland of the recharge line, preferably on a normal line or along probable flow lines and through the internodal point. Under uniform conditions, it may be expected that the flow line (after mergence) from the wells would be normal to the recharge line. However, in the event the recharge line is not parallel to the ground water contours, the flow lines may deviate somewhat from a line normal to the recharge line. With control of the recharge rate possible by an internodal well and with the recharge rate limited by the available head in the recharge well, there is no necessity, in a purely SEA ^YATER INTRUSION IN CALIFORNIA 47 operational line of wells, for an intermediate observa- tion well along the axis of the line of recharge (such as the 20-foot observation wells used in this test). However, an occasional intermediate test hole to de- termine the masiunim height of pressure mound, as compared to the required injection liead within the recharge well, would be desirable to evaluate the efficiency of the recharge well. C. RECHARGE WELL ACCEPTANCE RATE The recharge well acceptance rate, or rate at which the well will accept flow, is a fundamental limit to maximum rate of 1.86 cfs was obtained in gravel- packed well E for a few hours. Long-time rates at gravel-packed wells E and 1-A indicated the ease with which the acceptance rate can be maintained in these wells. The maximum sustained average rate at well E was l.OC cfs with an elevation in the well of 67 feet, and at well I-A .74 cfs with 30-foot head. Average injection rates at the non-gravel-packed wells were maintained with somewhat greater relative injection heads. The following table gives the maxi- mum average injection rate and its accompanying in- jection head at the test recharge wells: Maximum Average Injection Well Date Rate, cfs D 10/22/53 .48 K S/20/53 1.06 F 2/18/54 .61 G 10/22/53 1.03 H 8/20/53 1.01 I 5/14/53 .50 I-A 2/25/54 .74 J 11/26/53 .70 K 12/17/53 .74 ^\'ater Service dictation in Recharge Well Ft. Estimated Mound Elev., Ft. Indicated Injection Head. Ft. Indicated Unit Injec- tion Head, ft./cfs 55 14 41 85 67 14 .-.3 ."lO 48 9 3!) 04 69 12 57 55 68 10 58 57 70 — 2 72 144 30 10 20 27 67 11 56 80 54 5 49 66 determining well spacing as well as an important factor in determining the economics of building and maintaining a barrier mound. The acceptance rate of a recharge well depends upon: the type and size of well in relation to the characteristics of the aquifer; the amount of free chlorine available in the recharge water to prevent bacterial slime formation; and the degree and type of redevelopment utilized when the acceptance rate falls too low. Redevelopment is dis- cussed later in Chapter A'll. It was found that the time plot of a factor called "unit injection head" or specific injection head for a unit injection rate is a very useful guide to analyze and control the items affecting acceptance rate. Its use, however, is limited to the analysis of items affect- ing each well separately and cannot be used to make comparisons among the wells. Unit injection head is the injection head in a recharge well, as defined in Section B of this chapter, divided by the injection rate at the well. To simplify the detailed analysis, the elevation of the water surface elevation in the recharge well was used in the place of the actual in- jection head. Although this causes a small error in the magnitude of the unit injection head, as long as there are no significant changes in injection rate along the line of recharge, the modified value of unit injection head yields a factor which can be used to depict changes in the ability of the well to accept water. These values are plotted for the recharge wells on Plates 18, 19 and 20. Injection rates used in the test varied widely during different portions of the investigation. A In order to determine the injection head in the above table, it was necessary to estimate mound eleva- tions by use of the composite distance-rise equation discussed in the section on transmissibility. Section D of this chapter. For estimating these mound eleva- tions, values of "T" and "d" used were those derived in the transmissibility section by the use of the equation. In regard to the type of well, gravel-packed wells, represented in this test b.y wells E and I-A, definitely had better recharging characteristics. It should be noted, however, that the aquifer encountered was pri- marily sand with scattered gravel layers. Under these conditions the gravel-packed well was able to directly recharge all parts of the aquifer through which it passed, in contrast to the non-gravel-packed wells. In a predominantly sandy aquifer, the advantage of in- creased acceptance rate and the consequent lower in- jection head requirement is sufficient to justify the use of gravel-packed wells for recharge purposes. Although most aquifers along the coast, where re- charge mounds of this tji^e might be needed, would most likely be similar to that encountered at the test site, it is probable that in a coarse gravel aquifer, re- charge could occur through a non-gravel-packed well without excessive injection head requirements. A comparison of the acceptance rate can be made from the data collected for gravel-packed well E and non-gravel-packed well I at which injection began at the same time and at the same rate. As indicated on Plates 7 and 11, the injection head at well E rose 5 feet, while that at well I rose 55 feet. Well I was perforated for 35 feet, while the gravel-packed well 48 SEA WATER INTRUSION IN CALIFORNIA E was perforated throughout the entire depth of the aquifer penetrated. Hence the injected water was permitted to leave the casing through a greater area of perforations at a reduced velocity, thereby reduc- ing the required head. Further, the 20" ± diameter hole for the gravel-packed well, as compared to the 12-inch hole for the non-gravel-packed wells, exposes approximately 175% more area at the face of the aquifer, thereby reducing the initial radial velocity from the well and the resulting ground water mound at the well. The injection velocity and head also de- pend upon the transmissibility. Reference to Plate 22 indicates that, in this ease, the transmissibility is higher at the location of well I. Considering this fact, the difference between acceptance rate of the wells is even greater. Specific studies of the optimum well size, by drill- ing various sizes for the recharge line, were not made because it was felt that the success of the overall pro- gram should not be jeopardized by the possibility of drilling too small a well. However, based on experi- ence in the 1950 recharge well test and with the two types of wells drilled for the recharge line, it is felt that the best well for recharging purposes, in rela- tion to the aquifer material at this location, is one with a relatively large gravel-packed envelope and a relatively small casing. Consideration of the eco- nomic factors and of the necessity of providing suf- ficient work space for well tools, sample pumping equipment, conductor pipes and valves within the casing indicates that a 24-inch gravel-packed well with an 8-inch casing may prove to be best adapted for recharging purposes. Effect of Chlorinafion It was concluded in the 1950 recharge well test that it was necessary to maintain a free chlorine residual in recharge water in order to prevent the rapid de- cline of acceptance rate through the formation of bac- terial slimes, which apparently tend to clog easing perforations and local aquifer interstices. Thus, chlorination was considered in the earliest phases of planning for this test. The chlorination rate required to forestall the bac- terial slime formation will probably vary with differ- ent waters used. The use of the unit injection head curves to determine what the minimum necessary chlorine residual rate may be for a given water is demonstrated by the experience gained from this test. The chlorine residual was reduced in steps from 20.0 ppm to 1.5 ppm over a period of about 13 months. The experience with wells E and G, Plate 18, indi- cated that with the treated Colorado River water being used, a chlorine residual down to 10 ppm is capable of preventing clogging of the aquifer or per- forations. This subject is discussed more fully in the subsequent chapter on Water Quality and Treatment. Effect of Aeration An aeration test was conducted at well E during the period May 14 to 21. 1954. Air was allowed to enter the conductor pipe through the top of the well liead "T" (See Plate 4), while the back-pressure valve was open, and the water v.-as allowed to fall freely in the conductor pipe. Although the results of til is test were obscured by the necessity of changing injection rates along the recharge line occasioned by rehabilitation work at well J and also by a required change in the chlorination rate, Plate 18 indicates that the trend of the well's acceptance rate prior to and following the test showed no significant change. During aeration, an increased head in the well was apparent, which was due, at least in part, to a foam- ing action within the well, thereby reducing the density of the water head in the well. This action made measurement of actual head extremely difficult. The results approximately duplicate those originally observed in 1950 during the District's test at the abandoned Manhattan Beach Well No. 7. (7)* Indications of some degree of increase in well ac- ceptance were noted following shutdowns for well repairs. The increased acceptance may have resulted from the release of progressive air-binding and/or from the probable mild surging effected by closure and subsequent reactivation of the recharge wells. D. TRANSMISSIBILITY Transmissibility and gradient are the prime fact- tors in determining the total rate of injection, over a given reach, required to create an effective fresh water barrier to sea water intrusion; while Avell spacing must be determined on the basis of the accept- ance rate of the well in relation to the required total rate of injection. A descriptive definition of transmissibility is : The rate at which percolating waters pass through a unit width of a given aquifer under a unit hy- draulic gradient. Transmissibility may be expressed at r = PM where 2' = transmissibility P = permeability M =: aquifer thickness. In view of the fact that the effective depth of the aquifer is not definitely known and varies consider- ably over any given reach between two wells, the actual determination of permeability would be diffi- cult and probably inconsistent. Transmissibility, being determined directly from the observed data, has there- fore been used throughout in these analysis and is expressed in cfs/ft. per unit gradient. In order to find relationships or disparities between transmissibility as it is determined by pumping draw- down and transmissibilitv as measured during re- • See Bibliography (7). SEA WATER INTRUSION IN CALIFORNIA 49 charge, it was necessary to derive au equation relating the effect of transniissibility with the pressure eleva- tion at any p:iven location during recharj;e. The de- rivation of this equation and its application are dis- cussed below, along with other stautlard methods of determining transniissibility. Pumping tests were made upon the recharge wells shortly after comple- tion of drilling and development. Data from these tests were used to evaluate aquifer characteristics, and particularly transniissibility. Transniissibility values were also determined from the data related to recharge. Theoretical Basis of Transmissibilify Deferminafions Established methods are available to determine the value of transmissibility in the field by means of data collected from pumping wells and nearby observation wells. In general, they are divided into two classes, i.e., equilibrium and non-equilibrium conditions. Equilibrium conditions are defined as those where very little change occurs over a period of time in the values of drawdown in both pumping and observation wells, with a constant rate of discharge. Non-equi- librium conditions are those where drawdown is in- creasing, in relation to time, at magnitudes readily measurable. Pumping Transmissibility The characteristics of the aquifer at tlie test site, the rate of pumping discharge used, and the length of pumping at the test wells precluded tlie use of equilibrium formulae for the pumping test data. However, non-equilibrium equations were well fitted for use of the data collected. A simplified application of the non-equilibrium equation has been developed by Cooper & Jacob. (11) * Thus, Q -iirT In (2.2.5 Tt) (r'-S) where s = drawdown of the ground water surface in feet, (s = rise of ground water surface in later com- putations relative to recharge), Q = discharge of the well, cfs, T = transmissibility of the aquifer, cfs/ft., In denotes logarithm to the base e, t = time elapsed since start of discharge in seconds, r = distance from the recharging well in feet, and S = coefficient of storage, dimensionless (volume of water that a unit decUne of head releases from storage in a vertical prism of the aquifer of unit cross section). According to Cooper and Jacob, this approximation of the more general equation is probablj- sufficient if the quantitv, u, is less r'S than . 02, where u = 4T< By plotting r, /, or (r /<) against s on a semi-logarithm paper (with s on the arithmetic scale) and finding the slope of a line best fitting the plotted points, the transmissibility may be cominited in the following equations, respectively: (As, difference in drawdown over one logarithmic cycle, is the aforementioned slope.) Transmissi- Coefficient Method bilily of Storage Distance — Drawdown Time — Drawdown Composite — Drawdown T 2.303Q 2Asjr 2.303Q 4As7r 2.303Q S = s = 2.25Tt 2.25Tto 2.257 4As7r (rVO. The coefficient of storage, as calculated from the above equations, is a measure of the degree of confine- ment of the aquifer in the vicinity of a tested well. The values of To, t„, and (r^/i)o are those at s = on the semi-logarithmic plot. The evaluation of the stor- age factor is inherently subject to a much larger error than that of transmissibility. In view of the extreme variability of the aquifer in the test area, it is felt that the use of values of S (coefficient of storage) for anything more than an indication of a trend would not be justified. It has been shown by Theis (12) * that the time drawdown equation for transmissibility holds for re- covery from pumping if the quantity of t/t' replaces t in the semi-logarithmic plotting, where f = time since pumping ceased, and t is as before. The use of this factor permits a second, independent calculation in observation wells and, also, the evaluation of trans- missibility at the pumped well itself. Theoretically, the eqiiations arc based on an infi- nitely extending artesian aquifer of uniform thick- ness and permeability. Although the aquifer is by no means infinite, being bounded by a seacoast at an effective distance of approximately 2000 feet, analysis of observation wells in the vicinity indicate that no serious effect from the presence of the ocean is felt during the relative short pumping times used (8 hours). No measurable drawdown evidence of pump- ing was felt at distances from the pumping well greater than 2,000 feet. To minimize the effects of variable thickness and permeability of the aquifer, values were determined based only on the nearest observation wells, where possible. Recharge Transmissibility Transmissibility in the vicinity of a stabilized bar- rier mound which is just balanced so as to prevent saline intrusion can be determined from the Darc.v equation of ground water flow, Q ^ K a i or, as used in this report, Q =: T i L, where L is the length of reach considered. Since, theoretically, the rate of ground water flow landward from the barrier mound • See Blbllo^aphy (11), page 526. * See Bibliography (12), pages 519-524. 50 SEA WATER INTRUSION IN CALIFORNIA is known, and the frround water gradient can be measured in the area landward where the flow lias become linear, it is possible to compute values of transmissibility from this equation. Along the recharge line itself it is possible to derive a theoretical equation for the shape and size of the barrier mound. Equilibrium conditions of a recharg- ing well may be represented by the equation : s = Q AttT In (2rf+t/)-+x- where s = rise of piezometric surface from initial conditions at a given observation well or point, where d = distance of recharge well from the seacoast, X = distance to observation well from the re- charge well measured in a direction parallel to the coastline, and y = distance to observation well from the re- charge well measured normal to the coast- hne. This equation is derived as follows: If we assume two-dimensional flow, then the flow due to a recharge well near a straight coastline may be represented by that due to a source at the well and a sink located at the mirror image of the source beyond the coastline. "Whereas a source may be represented by a well being recharged, a sink may be thought of as a pumping well. In this type of analysis the coastline, a line of equipotential, is termed a "line source." The piezometric surface rise at any point is simply the algebraic sum of the rises due to the source and sink. In other words, a single well (point source) in a semi-infinite aquifer (bounded by the coastline or line source) is replaced by a source and sink in an infinite aquifer (13)*. The rise due to a recharge well in an infinite aquifer is: Q 4:TrT In (2.25 rt) where s = rise at a point r feet from well. The rise due to the source and sink is : s = Q ■iirT In 2.25 Tt In 2.25 Tt riS J where ri = distance from source, and )'2 = distance from sink, therefore s = Q 4:tT s = Q s = Q 47rr In 2.25 Tt , , 1 , 2.25 Tt In in ^ In 1 ''2- In 2 ^^(2d+jr±x^ ,; By the use of Equation II, it is possible to deter- mine a value of transmissibility measured at a giveu point with a single recharging well. Likewise it is possible to determine a value of transmissibility measured at a given point with multiple recharging wells as derived below : At any point on the x-axis where y — Q In 4d.'+x' III 4TrT X' As the effect of recharge at any point is the sum of the individual effects of all recharging wells in the line (13)*, then at any point on the axis of the recharge wells, i.e., the x-axis: 1 4irT Qi In 4d'+Xi 2 Xi -Qi In M'+Xi' + Qi In 4d'+x,' IV x\ where Xi, X2, . . . x, = distance to recharge wells from point of measurement of s, and Qi, Q2, . . . Qi = rate of injection at recharge wells. The use of this equation has been called, in this report, the com- posite distance-rise method. After equilibrium has been reached, the values of s measured at an observation well and Q at the re- charge wells can be used with the constants d and x to obtain T for each point of measurement of s. The values thus obtained from the field data were not con- sistent with values obtained by other methods and. therefore, it is indicated that the equation does not fit conditions at this test site. The transmissibility values obtained by use of the equation are a little more than one half of those computed bj- the other methods. The fact that this equation does not yield values consistent with those determined by other methods is probably due to the assumption of a per- fect pressure aquifer and because it is based upon the reaction of stabilized pressures at a considerable dis- tance from a "line source." Hydrologic data indi- cates, and the geology shows, that portions of the aquifer oceanward of the line of recharge are not under pressure. In a more completely confined aqui- fer this equation may prove more adaptable. However, since the values of T and also d, indicated by the solution of Equation IV, were consistent with each other, they were considered to be constants which include the effect of the nonpressure characteristics of the oceanward aquifer. These constants were used to estimate recharge well mound heights in Chapter V, B. It is felt that the value of mound height, de- rived in this manner, is a reasonable approximation to an otherwise difficult value to estimate. • See Bibliography (13), page 175. ' See Bibliography (13), page 509. SEA WATER INTRUSION IX CALIFORNIA 51 Deierminafion of Field Transmissibility The accuracy of usini; values of transmissibility (Ictcrmined in field tests as averages representing the entire aquifer is limited by the effects of the natural variations which exist in the aquifer materials. Local- ized areas of high or low transmissibility in the vicin- ity of the pumping well and, to some extent, near the observation wells can have a large influence upon the value derived. Hence, individual values can vary considerably from those which might represent the average aquifer. As discussed below, several methods were used to compute transmissibility and the values thus derived are summarized in a table at the end of this section. To graphically illustrate the variation in transmissibility, these values for the various wells are plotted on a profile along the recharge line on Plate 22. During Pumping Determinations of field transmissibility from pump- ing tests were obtained from data collected in the following manner: A pump was installed at the well to be tested with its discharge line connected to a weir box to provide accurate measurements of the pumping rate. (See Photo 3) In each pumped well, an airline (calibrated by measurements with a steel tape before and after pumping) measured the pump- ing drawdown. Taped and airline measurements of water surface elevation, after pumping ceased, con- tinued until recovery was virtually completed — usu- ally within a 12-hour period. Observation wells were equipped Avith recorders which gave a continuous record of drawdown and recovery during the pumping tests. In general, obser- vations at weUs beyond 500 feet from the pumped well were not depictive of applicable results. It was found that the distribution of observation wells about the recharge wells, together with the limited pumping time, was such as to prohibit an accurate evaluation of transmissibility by the distance- drawdown method. Results by the other two non- equilibrium methods were relatively consistent, and are felt to be representative of the test area. A typ- ical computation by the time-drawdown method is presented on Plate 21. The value of transmissibility at well E for the pumped-well time-recovery method appears to be in error but is included on Plate 22 to provide continuity between wells D and F. Data for this test was too in- complete for proper evaluation. The value of trans- missibility at well I, using either pumped-well time- recovery or initial recharge indicates that a local area of low transmissibilitj- exists near the well. Deter- minations based on the 20-foot observation well are more consistent with the determinations for the bal- ance of the recharge area. During Recharge When recharge is initiated at a specific well, a non- e(|uilibrium condition, comparable to that of a pump- ing well, occurs for a limited period — usually less than a i)eriod of two days. Provided no other disturb- ing elements exist or occur during the period, another set of transmissibility data may be obtained. It was found that results were comparable in magnitude to those determined by pumping tests. In order to compute transmissibility values during stabilized recharge conditions from the Darcy equa- tion, it is necessary to make assumptions as to the length of reach of the recharge line which is un- affected by the lateral gateway at the ends, and the rate of ground water flow which is applicable to this reach. The assumption made herein was that the por- tion of the mound from the centerline of well E to the centerline of well J was relatively unaffected by end conditions. The validity of this assumption can be studied by reference to Plate 14, showing the profile of the barrier mound. The rate of ground water flow applicable to the above-chosen reach was then assumed to be the sum of the injection rate of the wells between wells E and J and one half of the rate at the latter two wells. However, since all the flow from the mound is not landward, it is necessary to modify this quantity somewhat. It would be possible to determine the average gradient from the recharge mound to the ocean and correct it for the head differential due to the differ- ence in density between intruding water and injected water. However, it is more advantageous, in the pres- ence of a mound stabilized at the minimum barrier elevation, and probably sufficiently accurate, to use the theoretical equation concerning waste to the ocean discussed above. Thus, about 95% of the water in- jected between wells E and J was assumed to be mov- ing landward. The landward, linear gradients used to determine transmissibility values by this method were those indicated on the ground water profile nor- mal to the line of recharge through well 6 (Plate 15). Spacing of Recharge Wells As discussed in Section A of this chapter, having determined the rate of intrusion of sea water through- out a given reach of coastline and estimated the ac- ceptance rate of the recharge wells as related to the transmissibility of the aquifer, the spacing of such wells can be determined merely by dividing the total required flow by the estimated acceptance rate. Prac- tically, the above criteria might be applied to the de- sign of a recharge line as described below. Assuming that the geology of the general area has been thoroughly investigated, the conditions of overdraft in the basins having been established and 52 SEA WATER INTRUSION IN CALIFORNIA UJ Z o < X u O z o < z g I— < z Of >- to to Z < Of >- Of < oca 1 g"g, •^ jS "3 a-o .2 Time Re- very umpe WeUs o 2 ix; « c-i CI CO in CM a S « "• o o (N c^ o CD 2 8^ E- 1 1 0) me .e- •ery 20' ells 2 o ■^ o o CD -* a (N o C^ C^l O 'a B H*^ o^'^ 3 ex. 1 Time Draw- down at 20' Wells to "* »n o Ol (N IN 1 T) Dist. from umpe WeU Ft. r- t^ in o o o in O O O »c o I—I M (N CO (M W C) S CS -H 1 Uili ' « "O 03 " 00 rf rf w ^ ^ in CO >n CO m «5 O CO o IN CO T3 ~ tec « — . Q T3 3 E^ m O in in 6 CD o CO in CO o o in Cft CD d. lO CO t- (N 00 IN lO CO M a Time of ampin ra.-Mi: TJH -^t* • t^W O 00 t* CO >n CO M (N in o in ^o-* ^ Oi O ^ 00 cr. »•! CJ CJ ^ t" M "* c^ •»)* o o in b- eo CO o) in W -^f CI in IN in w a» ' ' ^ M " ' 6 s (N CO (N CO N CO CO CO N CO eo CO CO CO eo CO CO eo CIJ lO tn m m in m in »n in o in in >n uo in in in lo 1 n ^^ -^ ~^ \. ^v. -v.. ■^•^ \ w -^■^ ■^\ \ w "C in o -H (O in m t^ ^ Oi OJ O (N O •«j< m 'Ji* C4 00 OJ — « eo w IN N C) ^ ^ (N (N IN ^ IN B •\ \ "^ "^ ^ "^ ^■\ ■^'^•^ "^\ \"^ \ w N "f N O N CO O* Oi (N CO W CO f-1 -^J* Z Pumped or Re- charge Wells C d H Cb 6 te ^ )£ 1 o 5 UJ 05 .- a» : " o 3a" f" tl fe O & C3 •" C3 g »*-l ^ o ago! U QJ (U 0) S C8 P. "■" ja '5 t; m P 1= o 5.S 1-2 iS 4-1 O ^ a 2 * © m s a c 5P O u-i m o C.2 P -ate tJD +J ^ « C I a S g S 3

-• a o *^ •- r- - §2 "I s -*i ^ a •" —■ c a ■- fM ^' -^ - -'-' c3 tn " §".22 a «3z: ii i; "S '■3 ago SEA WATER INTRUSION IN CALIFORNIA 53 having t'ouiul suitable right of way, the procedure would be to : (a) Based on the best available geologie informa- tion, divide tlie eoastliiie to be protected into reaches of estimated uniform transmissibility. (b) Drill a tost hole near the center of each reach throughout tlie entire depth of aquifer mate- rials to definitely confirm the geology and es- tablish the dejith of recharge well required. This test hole later could be used either as an adjacent observation well or a pilot well for the recharge well. (c) Based on the established lithology and esti- mated permissible injection head in relation to the transmissibility of the aquifer, a recharge well would be designed and constructed at or near this location. (d) A pum]iing test would be performed on the re- charge well after the completion of drilling to establish the aquifer's transmissibility. Such a pumping test should be made with moderate rates of pumping to prevent over-development of the well. (e) The well drilling and pumping test will provide a check on the initial assumed conditions. Ob- serving the principles as outlined in this report, a more accurate calculation could then be made to determine the required spacing. (f ) With the spacing thus determined, the adjacent internodal observation wells could be drilled providing a further check on the geology. This would be followed by the drilling of the adja- cent recharge wells at each side and the above check of geology and transmissibility dupli- cated. Such a procedure coidd then be contin- ued until all recharge wells were completed. This procedure would, of course, also applj- to an extension of any existing recharge line. Obviouslj-, since the many variable factors can be determined only with approximate accuracy, a corresponding safety factor must be considered in determining the actual spacing and the estimated acceptance rate of the recharge vrells. The change in spacing ultimately adopted at the test site, i.e., from 1000 to 500 feet, was due to the lim- ited acceptance rate of the nongravel-packed wells which, at higher rates, induced excessive injection pressures on the clay cap. This reduction in spacing permitted maintenance of the barrier at an average injection rate of approximately 0.50 cfs per well. An inspection of Plates 7, 9 and 11 shows that injection rates from 0.75 cfs to 1.0 cfs could easil.y be main- tained at the gravel-packed wells. (Following the re- medial work done at well G, it could be considered as a quasi-gravel-packed well.) This data would indicate that an injection rate of 0.75 cfs at 6 gravel-packed wells with a spacing of 750 feet would have main- tained an equivalent barrier at the test site. A higher acceptance rate would, of course, permit even greater spacing. E. MOVEMENT OF INJECTED WATER Variation in chloride salinity constituted the basic indicator for tracing ground water movement land- ward and seaward as well as along the line of injection wells. The variations were detected by chemical deter- minations made of ground water samples collected by pumping, by use of a "thief sampler," and by con- ductivity traverses made in observation wells. The chloride concentration history of the key project ob- servation wells are shown on Plates 23 to 28. These plates show comparisons of the chloride history of the internodal wells along the recharge line ; of the line of wells normal to the recharge line through well G; of the wells lying approximately 500 feet landward of the recharge line ; of the wells lying approximately 1000 feet landward of the recharge line; of the wells laterally north and south of the recharge line ; and of a group of wells used as special indicators of the chloride concentration on the perimeter of the imme- diate test area. Comparison of the isochlor maps for Februar.y, 1953, and June, 1954, Plates 33 and 34, show the size, shape and mergence of the fresh water bulbs resulting from injection. Of particular interest is the fact that the isochlors indicate that the movement of water is influenced by a southerly component of the ground water gradient which has caused the flow to veer southerly from the expected route of travel normal to the recharge line. This deviation to the south places most of the landward project observation wells in the zones of mergence rather than in the main areas of fresh water movement. Sampling Techniques and Field Procedure Previous District expericTiec with water well sam- pling has shown that samples taken from a non- pumped well, particularly from one not pumped for a considerable period of time are not necessarily rep- resentative of the quality of the ground water body. The desirability of obtaining pumped samples without the disadvantage of disturbing ground water flow with a large capacity pump led to the use of a small diam- eter, 10 gpm submersible pump for sampling pur- poses. Most of the project's ground water samples were taken with this type of pump. The principal ground water sampling equipment utilized consisted of: (1) A li HP submersible pump about 4 inches in diameter and 5 feet long, pumping about 10 gpm; (2) Pickup truck for transporting tlie pump and miscellaneous equipment, and upon which was mounted a winch and cable reel for lowering the 54 SEA WATER INTRUSION IN CALIFORNIA pump into the -wells; (3) Generator, powered by a gasoline engine, mounted on a trailer and pulled by the pickup truck. (See Photo 6) To provide comparatire data, each sampled well was initially pumped until the concentration of chlorides stabilized. Stabilization time varied from 1 hour to 17 hours. Numerous analyses of the pumped effluent were made during the initial pumping to determine the point of stabilization. Thereafter, upon pumping a given well, the stabilization period previously deter- mined was used and cheeked by spot analyses. Generally, the pump was placed between two sets of perforations to facilitate interpretation of conduc- tivity data (described below). The small capacity of the pump produced such a small amount of mixing in the well that very little change in well water salinity was noted below the bottom set of perforations. When salinity declined in the aquifer due to recharging operations, it was noted that the water below the bot- tom perforations remained highly saline. In conjunction with the water sampling program, numerous "thief" samples were taken. A "thief" •sampler is a device by which water samples may be taken at anj^ desired depth within a well. The sam- plers used consisted of a small "thieving" head unit which coiild be fitted to sample containers of various sizes. The "thieving" head unit used is designed so that when a brass shim serving as a sealing diaphragm is pierced by a plunger, the attached container is filled. The sampler is lowered into a well by a %2-iiich stainless steel cable utilizing a small reel. "When at the desired depth, the sample is taken by releasing a cylindrical weight, drilled and fitted around the sus- pension cable. The weight runs down the cable and activates the plunger which breaks the brass sealing shim and allows the container to fill. As the hole punched through the shim is small in relation to the capacity of the container, little mixing of consequence occurs during the removal of the container from the well. Examples of two types of samplers used in the test are shown on Photo 7. Partial chemical analyses for chloi'ide, carbonate and bicarbonate were made at the field laboratory, and complete analyses for significant constituents of well samples were made at the District's testing lab- oratory. Detailed results are available in this Dis- trict's files. (See Appendix F) Electrical conductivity equipment was used, in con- junction with the sampling pump, to obtain informa- tion on chloride concentrations at various horizons of the merged aquifer — -those corresponding with various sets of well perforations. In many cases the interface between fre.sh and salt water could be iden- tified by means of this technique. Conductivity equipment consisted of a conductivity cell, a reel of conductor cable, and an alternating cur- rent Wheatstone Bridge. Power was supplied either by the pump generator at 110 volts AC or by a 6 volt DC storage battery converted to 110 volt AC with a vibrator converter. The conductivity cell consisted of a pair of electrodes rigidly mounted, protected by an insulated shield, but exposed to the water when sub- merged. The resistance between the electrodes was measured by means of the Wheatstone Bridge. In order to convert resistance to eonducti\'ity and to occasionally check the conductivity cell constant, a one-tenth normal solution of potassium chloride was used as a standard. The effect of temperature variations upon conduc- tivity values was significant. In pumping tests, the temperature of the pumped effluent was measured with a thermometer. In non-pumping tests a thermis- tor was lowered into the well. Resistance of this in- strument measured on a Wheatstone Bridge, when converted, indicated the water temperature. A set of conductivity traverses was generally taken in conjunction with the pumping of well water sam- ples as follows : (a) Static Traverse The conductivity cell was lowered to a posi- tion below the lowest well perforation before pumping in order to avoid the necessity of lowering the cell after the pump was in posi- tion which had proved to be difficult. Readings were taken at predetermined depths (usually at 5 or 10-foot intervals) as the cell was low- ered. Without removing the cell, the pump was then lowered into the well and the well pumped until the chloride concentration was stabilized. (b) Stabilized Traverse With the pump running and stabilization reached, the conductivity cell was then with- drawn from the well with conductivity read- ings being taken during the process. Additional static traverses, made in a man- ner similar to that described above, were often made at non-pumped observation wells to de- termine static trends. Effects of Fresh Water Injection The progressive effects of recharging on the ground water body were reflected in the salinity time histories of observed project wells. The initial effect of con- tinued injection was the creation of an inland-moving temporary saline wave. The sustained effects of con- tinued injection were: (1) Formation of an interface between waters of different salinities and develop- ment of an overriding fresh water wedge; and (2) progressive mergence of the individual fresh water bulbs into a continuous fresh water front. The temporary saline wave must not be confused with the continual increase in chloride salinity result- ing from sea water intrusion which was occurring prior to the commencement of recharge and which continued subsequently as the trapped saline waters continued to move landward under the influence of SEA WATER INTRUSION IN CALIFORNIA 55 oxistiug ground water gradients, even thougli re- permeability to transmissibility, the following rela- eharge had cut off and replaced the source of supply. tionship evolves : The magnitude of the temporary wave, as reflected T^st by chloride coneentrations, is indieated on Plate 24 P ~Md*~ and by the curves for observation wells G-2, G-4. and ^^^^^ ^ effective porosity G-8. The wave at G-2. 250 feet inland increased tlie ^^^ following table lists the actual velocity of chloride concentration about 3.7% dunng March ^^^^^j ^^ ^^^ ^^^^ ^^^^^^ ^/ ^^^ ^^.^^^„^ ^^^^i^^t 19o3; at G-4 500 feet niland. it increased the concen- ^^^^^ ^^.^j^ ^,^.^ ^^^^ occurred, and the calculated ^™J'?, o^' I''- ^ """^,,^^'*>' ^"^^ '^""" 3 :■ ;■ ; values of effective porosity, based on a transmissibility G-S, 1 ISO feet inland, the w.ive was not distinguish- ^^ jg (discussed in the previous section) and able. In general, the area affected by the emporary ^^^^ ^^^^^^^^^ ^^ ^^^^ observation wells, saline wave IS approximately the same as the radius ,^^^^^^^^ ^^^^^^^^ j,^^^^.^^ of influence noted during pumping drawdown tests. Reach ffrafUiiit velocitij porosity An indication of a saline wave was detected within G-G-2 O.OKk; O.S ft./day 24.0% specified horizons in the aquifer at each well bv con- D-C-4 .OlHO fi.O 25.8 ductivity traverses, but the magnitude was masked by K-K-4 Ofi 60 29.6 interference of waters of different quality in the well. — It may be noted that injection commenced in an Average 6.9 ft./day 27.07/s/54 to 6/30/54 5 to sudden pressure which is associated with well op- The lowest concentration that would satisfy the eration, to be most serious. Hence, future wells should first two criteria was 1.5 ppm. However, the required be gravel-packed to eliminate large voids adjacent to injection head (at constant recharge rate) started to the casing, and properly grouted to seal the "clay increase at this low dosage as shown on Plate 19. The cap" to the well casing. dosage was then increased to 5 ppm, which still re- sulted in an appreciable continued increase in the re- D. EFFECT OF MICROBIOLOGICAL GROWTHS quired recharge well head. A low bacterial count and AND CHLORINATION ON AQUIFER ^ detectable residual chlorine in the "20-foot" well are therefore not reliable criteria for controlling PERMEABILIIY chlorine dosage. The third criterion, that of adding The effect of microbiological growths natively pres- sufficient chlorine to permit maximum acceptance, ent in the aquifer and in the recharge water is of would appear to be the logical one to use. The one great concern. Dr. Carl Wilson, the District's Con- difficulty inherent in this method is that even an opti- sultant, states: (See Appendix D) mum dosage may permit an insiduous bacterial or "In connection with recharge operations at Man- o^?^"^*^ "J^^tli to collect in the perforations without hattan Beach, it was assumed that the steadily ^^^^^ ^■<'^*^°*^*^ ^° ^" immediate reduction ,n ac- diminishing rate at which water could be intro- ceptance rate. By the time reduction is noticed, a duced into an aquifer was principally due to the sufficient quantity of slime may have accumulated growth of microorganisms in the sands and gravels ^^^^ ^°"1^^ ^'"^ ^^!<^<^"1* ^° ^^"^''^^ ^^ increasing the of the aquifer. It is a well established fact that ^''^^^'' "^ «™;'ll '"cfmef «. However, by giving the virile bacteria, and some other micro-organisms, are recharge well a shock treatment of 20 ppm or always present in soils and sands and travels, where °iO''e. it may be possible to remove most of the they persist indefinitely in small members in static accumulated slime. equilibrium with the environment. Should condi- Dr. Carl Wilson, District Consultant, states that^ tions within the environment become more favorable, ' ' The best criterion of chlorine sufficiency is un- these dormant organisms would burst into activity. doubtedly found in a constant acceptance rate, but The introduction of an imported surface water, like in this connection, he who fixes the chlorine dosage the Colorado River water used to recharge the must remember that physical conditions within the underground basin, which is relatively high in aquifer will always limit the rate at which water organic matter, would provide a powerful stimulant may be injected into it, and when this limit is to the growth and multiplication of the native or- reached, no increase in chlorine dosage can further ganisms at the same time that it brought with it a augment the acceptance rate." (The constant ac- foreign flora to complicate the situation." ceptance rate, as far as it is affected by growths. 60 SEA WATER INTRUSION IN CALIFORNIA can be accomplished only by heavy chlorine dos- age.) "Experience at Manhattan Beach seems to point strongly toward twenty parts per million as a proper initial dosage to prepare the aquifer vesti- bule to a sufficient distance to permit injection at the desired rate. . . . After stability has been at- tained and maintained for a period of perhaps a week, then the dosage may be reduced twenty-five per cent. Experience leads to the belief that after stabilization has been obtained, a dosage of eight to ten parts per million will be required to maintain a constant injection rate." To determine the potential corrosive properties of Colorado River water when heavily chlorinated (12 ppm) prior to injection into the ground, Dr. Wilson analyzed a series of water samples. The least benign of the waters had a pH of 7.45 and a free carbon dioxide content of 10 ppm which, in his opinion, was unlikely to be actively corrosive. Hence, chlorination at a dosage of 12 ppm appeared to impose no special hazard of corrosion. E. ADDITIONAL EFFECTS ON AQUIFER PERMEABILITY 7. The Effect of Iron in Recharge Wafer Although chemical analyses indicated an iron pickup from the water transmission line, it was too small to adversely affect the recharge water. How- ever, there was a sufficiently large amount of iron in the recharge observation wells to constitute a prob- lem. The formation at the perforations of a gelatinous iron hydroxide which is encouraged by the presence of oxidizing substances, chlorine and oxj-gen, would tend to reduce the acceptance of the recharge water. Examination of water samples i-evealed presence of iron oxide floe embedded in bacterial slime. Iron present in water in insoluble forms, usually as hy- droxides but often as sulphides, is found in most alluvial fills. Where iron is taken into solution, it occurs as ferrous bicarbonate due to the carbon diox- ide released by bacterial activity. When such water comes into contact with the air, the iron is oxidized and is finally precipitated as red-brown gelatinous ferric hydroxide. The effect of hydroxides of iron is not only to reduce aquifer permeability but also to cause accumulations of corrosion products in back- pressure valves, as well as to affect the operation of other equipment. 2. The Effecf of Calcium Carbonate (Sludge) and Suspended Solids The presence of suspended solids and rust particles have been of no apparent consequence during the test to date. However, on one occasion during the test, a temporary Metropolitan Water District plant opera- tional difficulty released an appreciable amount of sludge into the Metropolitan Water District feeder line which proved to be primarily calcium carbonate. Although this sludge had a temporarj- etfect on the operation of the chlorinator, appearance of such sludge would not be normally expected. 3. Effect of Dissolved Oxygen Although aualj'ses of ground water taken during development tests indicated a dissolved oxygen eon- lent of fi-oni 0.6 ppm to 2.6 ppm, it is believed that a leaking pump column made these results question- able and that the actual dissolved oxygen in ground water was well below 1.0 ppm. Little dissolved oxy- gen released by sulphate reduction would be antici- pated, largely because of the rapid utilization of such small increments of the gas by bacteria present within the aquifer. The injection of recharge water with a high dis- solved oxygen content would not only significantly contribute to the growth of slime-forming organisms, but also probably tend to affect the aqiaifer by air binding. The average temperature of the imported Colorado River water was found to be about 18 °C which would permit a maximum dissolved oxygen content of 9.6 ppm. The concentration at times at Manhattan Beach exceeded the maximum solubility of oxygen and the excess was made apparent by the appearance of trapped air bubbles. Further tests are required to conclusivelv establish the effect of air binding. I CHAPTER VII MAINTENANCE AND OPERATIONAL PROBLEMS A. WELL G-REHABILITATION The most (.Titioal mainteuauce problems, causes, effects aud remedial measures undertaken were all related to the subsidence that occurred at well G early duriiiir the test. Subsequent well rehabilitation work, preventative measures and operational iiroccdures adopted were directly eouueeted to this experience. The success of these measures and procedures in bringing the test to an ultimate successful conclu- sion substantiate tlie validity of the analj-sis relating to the causes of this failure. It is therefore felt that a chronological account of this occurrence and the subsequent repair provide a valuable background to the problems that might occur in connection with re- charge. Specifically, this failure established criteria for the: (1) design, (2) development, (3) rehabili- tation, (4) operation, and (5) redevelopment of re- charge wells. Recharge at the test site was initiated at well G on February 14, 1953. The recharge rate was in- creased in stages and by March 11 the input was one second-foot. During this period the constant rate flow controller valve gave considerable difficulty and constantly required adjustment. This resulted in fre- quent rapid increases and decreases of input rate. On March 13 during the period of such an adjust- ment and with several people present, a subsidence occurred on the south side of the well. The subsidence occurred almost instantaneously. The residting cav- ern was roughly circular in area, and exposed a hole at the surface some 10 feet in diameter. This area increased below the surface to some 15 or 16 feet in diameter and some 25 to 30 feet in depth, exposing about 15 feet of the well casing. The inflow to the well was immediately reduced to approximately 0.4 second-feet and the cavern filled with native material. During the original drilling of the well, two sub- stantial horizons of water-bearing gravels were en- countered in the aquifer and the well was perforated in these two zones. At the request of the State Engi- neer's office, the lower zone was perforated first, and then developed by bailing, surging and pumping, and subsequently pump-tested for transmissibility. Following this, the upper gravel horizon was per- forated and likewise bailed, surged and pumped, and a second pumping transmissibility test performed. During the second stage of development, materials removed from the well definitely showed that the intervening sediments between the two gravel horizons had brokeu down and that the upper yellow-colored gravels had combined or merged with the lower blue- gray-colored gravels. Hence it may be assumed that the well had been overdeveloped and large voids had probably formed below the clay cap. Exploration was next undertaken in an attempt to determine if the failure had occurred in the clay cap. A 2-iuch test hole was jetted and driven to a depth of 145 feet in the area of subsidence, about two feet south from the original well. No evidence of a clay cap was encountered. A similar test hole vras drilled about three feet north of the recharge well and defi- nitely indicated that the clay cap in this area was still in place and apparently had not been disturbed. The well was grouted through this hole near the lower por- tion of the clay cap, employing a pressure grout pump and a grout mix of 27 sacks of cement and three parts of water. Maximum pressure employed was about 50 pounds per square inch at the ground surface. To pre- vent the grout mix from sealing the casing perfora- tions, injection at the recharge well was continued during grouting operations. Following the grouting, the 2-inch pipe casing at the test hole was flushed clean and pulled clear of the upper level of the clay cap to serve as a test well to jnensure water levels in the overlying sand dune materials. Upon testing re- charge at well G, water levels in the test well indi- cated that this grouting was not effective. The re- charge well was then dismantled and filled with sand above all perforations to prevent grout from moving through the perforations into the well, and the area was again grouted through the northerly 2-ineh test hole by stages (i.e., raising the pipe at intervals dur- ing grout placement), using 34 sacks of cement with a 3 : 1 water/cement mix and a maximum of 50 pounds pressure. The recharge well was then cleared of sand. A third test hole was drilled to check water levels above the clay cap. The recharge well was again tested by injection, and water levels above the clay cap im- mediatelj' responded, indicating the ineffectiveness of the second grouting procedure. The recharge well was then filled with sand to a depth of 105 feet below ground surface, the approximate elevation of the bot- tom clay cap ; and grout was placed in stages by using the recharge well itself as the conductor and by cut- ting two sets of additional perforations within the clay zone. Grout was pumped at 50 pounds pressure, utiliz- ing 270 and 41 sacks of cement, respectively. Clearing of well casing was again commenced ; however, run- ning sand and mud prevented complete cleanout. The (CI) 62 SEA WATER INTRUSION IN CALIFORNIA well was again filled to approximately 110 feet with sand and regrouted above the previous gronting zone with a 42S-sack grout mix. On commencing cleanout, running sand and mud indicated that the grouting had not completely filled the grouting perforations and the well was again grouted with an additional 49- sack mix. During the operations noted above, additional sub- sidence had occurred at the location of the original cave-in, particularly during periods of cleanout opera- tions at the recharge well. After the final grouting mentioned above, and commencement of bailing opera- tions, this subsidence continued. In order to stabilize this movement, it was decided to gravel-pack the well through two 6-inch conductor pipes which were placed in rotary-drilled holes on either side of the recharge well. These two 6-inch casings were placed to a depth of 106 feet and both conductor pipes were thoroughly grouted with a 52-sack mix and a 35-sack mix, re- spectively. After setting, the grout was removed with a rotary drill and the hole continued on into the aquifer. Subsequent bailing operations at well G and attempted gravel-packing through these conductor pipes proved unsuccessful. During the bailing opera- tions, the original subsidence area continued to move. This indicated that the well might be gravel-packed through the area of subsidence ; hence additional per- forations were added so that tlie well was completely perforated within the aquifer from 121 feet to 212 feet below ground surface and bailing was continued as gravel was added to the subsidence area. A total of 76 tons of gravel was added in this manner. As this gravel continued to move, decomposed granite was added in an attempt to form a seal over the gravel. Bailing continued until the decomposed granite had moved approximately to the vicinity of the clay cap. That this gravel actually moved into the aquifer and successfully gravel-packed the well was indicated dur- ing bailing by the pick-up of small particles of im- ported gravel. Injection was reinitiated at well G on June 9 and has continued to date with no indications of further subsidence or failure in well acceptance rate. B. WELL C ABANDONMENT Following the experience at well G, test holes were installed immediately adjacent to the remaining re- charge wells in order to indicate water levels in the sand dune materials overlying the clay cap. Recharge at well C resulted in immediate response in its test hole indicating excessive leakage through the clay cap. The well log obtained during drilling at well C indicated a relatively thin clay cap some 8 feet in thicknes.s. Dur- ing development, the pump discharge showed that ex- cessive quantities of fine sands and silts were being removed from the aquifer. In view of the above and based on the experience at well G, it was not considered advisable to continue re- charge at this location. Therefore, operations were sus- pended and the well was retained as an observation well. C. WELL I, FAILURE AND ABANDONMENT Late in May, 1953 a subsidence similar to that ex- perienced at well G occurred at well I. However, due to the reliabilitation costs encountered at well G and the uncertainty of the permanency and effectiveness of such rehabilitation, no such measures were deemed advisable until the work at well G could be properly evaluated. The subsidence area was back filled with gravel and injection was continued at a reduced rate. In accordance with a decision from the State Engi- neer's Office, a new gravel-packed well (I-A) was planned and installed near this site, as described previously in Chapter II. During the interim period, injection at well I was continued until October 14, 1953. D. WELLS D, F, H, AND J-GROUTING In order to prevent excessive leakage tlirough the clay cap adjacent to the well casings and po.ssible subsequent subsidence, remedial grouting work was undertaken. Prior to the initiation of recharge at these intermediate recharge wells, the wells were all filled with sand to approximately the bottom of the clay cap and stage grouted through added perfora- tions in the easing. Wells D, F and J were grouted with high pressure grouting equipment uuder inde- pendent contract between the State Department of Public Works, Division of Water Resources, and BJ Service, Inc. (Well H was grouted by District forces). Grouting at these four wells indicated that voids of significant capacity were present, in that a consid- erable quantity of grout was used at each of the wells. As an example, over 400 sacks of cement were used in grouting well H. The existence of such voids around the casing was also indicated by the occurence of casing misalignment at each M'ell during grouting procedures. Casing misalignment was severe at wells D and H and required swaging to realign the casing before well tools could be employed to clean out the wells. At both wells D and II swaging operations ap- parently ruptured the casings. These ruptures were immediately apparent at well D, which was repaired bj' filling the casing with grout through the damaged area and drilling a smaller 8-ineh hole through the grout, thereby effectively cement lining the casing. The rupture at well H was not apparent until later when the acceptance rate of the well declined and soundings showed that sand had run into the well covering a portion of the aquifer jierf oral ions. Dur- ing bailing cleanout operations, the rupture became obvious. The repair of this rupture was accomplished by groTiting an 8-inch steel liner within the r2-inch casing. I SEA WATER INTRUSION IX CALIFORNIA 63 All four wells are still operating successfully to ilate; however, after seven months of recharixe, the acceptance rate at well J decreased radically and soundinfTS indicated that the well had commenced to run sand. On coiumeneinir bailing operations, it be- came obvious that the upper aquifer perforations were running mud and sand, thereby indicating that tlie original grouting had not been elTectivc. This well was rehabilitated by grouting off the upper 26 feet of aquifer perforations and cementing in an 8- inch steel liner to cover the upper perforated area. During these operations approximately 125 sacks of cement grout were forced into the formation. Sub- sequent recharge operations indicate tliat this stabili- zation procedure was successful. E. LEAKAGE THROUGH THE CLAY CAP TO THE OVERLYING SAND DUNE MATERIALS Although evidence of some leakage at the well casings, through the clay cap, has been noted at the ad.i'acent test holes which were drilled to clieck such leakage, no subsequent well failures have occurred. Such leakage is considered insignificant in that, al- though relatively high levels have occurred in some of the test holes, additional adjacent test borings have failed to disclose any free water levels in the .ud dune materials. As previously mentioned in the > liapter on Geology, the test holes probably tap semi- pressure stringers near the top of the clay cap and do not reflect free water levels in the sand dune material. F. CONCLUSIONS RELATIVE TO MAINTENANCE PROBLEMS (1) From the above, it is apparent that a well drilled into coastal deposits, consisting of fine to coarse sands with limited stringers of gravel, should be a gravel-packed well in order that the materials removed from the aquifer dur- ing well development procedures can be re- placed by gravel and prevent the formation of excessive voids aiul subsequent serious well subsidence problems. Further, it is obvious that in a pressure aquifer where the confining clay layer is perforated by the well casing, an area of weakness may develop and result in excessive leakage; hence a properly con- structed cement seal in this zone will always be desirable. (2) It is of particular interest to note that in re- lation to the grouting operations mentioned, the major grout movement appeared to be lateral aiul upward, while there was very little tendency for the grout to move downward. This wa,s substantiated by the absence of any evidence of grout within the perforations be- low the grouting zones and particularly by the rotary drilled holes at well G which passed through previously grouted areas. (3) It is also obvious that recharge operations will place the clay cap under additional pressure, particularly at the zone of weakness along the well casing; therefore, care should be taken during operations to avoid rapid pressure changes residting from rapid increases or de- creases in injection rate. Recharge should be commenced and stopped with small increments of change in the flow, allowing sufficient time for pressures to stabilize between such changes. In this connection, it is recommended that a water supply to the recharge line be controlled with a pressure regulator and that changes at individ- ual wells be made with manually controlled valves. (4) As a result of the above-mentioned experiences, operation procedures were adopted approxi- mately as follows : Well injection was started by increasing flows slowly until the 6-inch con- ductor became filled. After reaching this point, at which the well was under complete control, continued increases were made at 2-hour inter- vals in increments of 0.10 cfs or less. Adjust- ments to well Q's requiring reductions of flows were made in decrements of 0.05 cfs at 4-hour intervals, the same procedure being followed in complete well shut-downs. (5) Redevolpment of recharge wells as distinguished from rehabilitation required by failure of the clay cap, was not necessary until after the period of this test report. Although not discussed in the foregoing, following the test period, well K was successfully redeveloped by moderate surging and bailing. A study of the unit acceptance rate for the well during this later period shows a marked improvement in acceptance rate follow- ing the redevolpment. Due to the aquifer condi- tions encountered at the test site, it is not expected that pumping will be used as a rede- velopment procedure because it removes excess quantities of sand which would probably be conducive to additional failure of the elaj' cap. However, in a gravel-packed well, pumping could probably be safely used so long as gravel was added to the gravel envelope to replace the re- moved aquifer sediments. CHAPTER VIM ANALYSIS OF PROJECT COSTS In aocordaiico witli aforenientioiKHl li'^islative au- tliorizatidii, the State Water Kesourees lioard was authorized to cooperate and contract with otlier agencies in order to make this investigation. The Board contracted with tlie Los Angeh'S County Flood Control District for the purpose of eoiulucting the field experimental project, including preliminary in- vestigation, planning, construction of facilities, and operation. Originally $450,000.00 was allocated for this work. Before the completion of the investigation, supplenu'ntal sums of $187,000.00 and $5,126.30 were allocated, bringing the total State funds to $642,- 126.30. A sum of $9,000.00 of the above total was reserved for payment to the District upon completion of a final report. From a detailed analysis of total District expendi- tures reimbursable by the State (October 1, 1951 to December 31, 1953), a cost breakdown was made. The expense items were segregated as follows : A. Caiiitiil Costs .$.S34.13r).72 R. Operiition and Maintenance Costs 98,833.53 C. EnisinecrinK, Investigation and Testing lOOJO."."."! I). SiipiTvisor.v Costs S4.L'01.IM» E. Misot'llaneous Costs 14,.S(!0.."i2 Mrnsiiring Kquipmi'nt, .Mi-tcis, Recorders and InstruniiMit Shelters 12.340.75 Total $641,730.10 • * Final accounting is not yet complete. A detailed cost analysis follows: A. CAPITAL COSTS The following breakdown of capital cost expendi- tures includes all related costs such as labor, materials and supi)lies, rented equipment. District equipment, private auto mileage, utilities, services, photos and blueprints, contract expenses, damages, and appli- cable District overheads. COSTS 1. rreliminary Study, Planning and Design 2. Rigiit of Way Acquisition, Kiglit of Way En- gineering 3. Pijieline' a. Design and Survey $17,747.8" 1). Installation 107.(l!)."i.!K) Well Cost.s a. It— 12" Recharge Wells' $44,030.91 b. 3<>- c. 14- S" OKservation Wells' 2" Test Holes' 98,r.20..V) 6.5C3..50 5. Tran.sniissil)ility I'nnip Tests (9 wells) 6. Recharge Well Appurtenances, Materials & Installation $5,262.67 1,829.58 $125,443.77 $] 49,214.91 4.455.17 20,.'>21.7S I'"ie!d Ollioe and Chlorinator Housing Costs a. Design $1,121.90 1). Construction 10,269.78 9. Design and Installation of Chlorine Equip- ment. Storage Facilities Total Capital Costs. $11.:;91.74 3.066.35 $334,135.72 ipipcllne consists of 10,000 feet of 10 ga., 20" O.D.: 1.034 feet of 12 ga., IG" O.D.: 1,064 feet of 12 ga., 14" O.D. welded steel pipe anil appurtenances. * Includes drilling and development of 8-cable tool standard-cased wells and 1 cable tool, gravel-packed well. ' Includes drilling and development, All wells Installed liy cable tool drill rigs with standard casing. * All wells driven and jetted Into place by di^flrlct personnel. The capital costs represent over 50% of the total expenditures. It may be noted that right of way ac- quisition costs were a minimum due to the coopera- tion of local agencies and property owners. Pipeline, design and survey costs were relatively high ; how- ever, it must be borne in mind that survey costs were significantly increased because of the extensive under- ground exploration required to locate existing city water mains, sewer and gas lines. B. OPERATION AND MAINTENANCE COSTS This breakdown includes those costs for the period February 1953 tlirough December 31, 1953, related to labor, materials and supplies, rented equipment. Dis- trict equipment, private auto mileage, utilities, photos and blueprints, services, right of way access and ap- plicable District overheads. Such costs are also in- cluded in C and E below. 1. Recharge Water $32,967.80 2. Chlorine 2.8.53.91 3. Recharge Line Operation 36,589..58 4. Recharge Well Rehabilitation • 20.734.17 5. Routine Project Facility Maintenance — 5,688.07 Total operation and maintenance costs $98,833.53 • Does not include B/J Service, Inc., charge of $3,426.59 for grouting wells D. F. and J. This was billed directly to State. It is believed that the operation and maintenance costs are in proportion to the magnitude of the pro- ject and may be used as a basis for estimating opera- tion and maintenance costs involved in constructing a similar barrier project. The use of these costs on such a basis are discu.ssed under Section G below. The water used for this jiroject was treated Colo- rado River water imported by the .Metropolitan Water District in that no other supply was available. Addi- tional water not included in the above costs was pur- chased for the test bv the West Basin Water Associa- 3—52568 (65) 66 SEA WATER INTRUSION IN CALIFORNIA tiou. All water was obtained at a cost of $20.00 per acre foot. C. ENGINEERING, INVESTIGATIONAL AND TESTING COSTS Engineeriiifr investigation and testing costs for the test project were, of course, much more detailed, comprehensive and costly than similar engineering work required for operating a project; this also is discussed in more detail in Section 6 below. The items of work included in this category are as follows : 1. Well Sampling Operation (IncIudinR "waiver wells") $33,540.16 2. Preliminary and Operational Test Data : Collection, Compilation, In- terpretation, and Computations 40,596.05 3. Geological Analysis 13,280.76 4. Water Analysis and Soils Testing 20,788.36 5. Water Consultant 1,500.00 Total $109,705.33 Attention should be called to the fact that the geologic analysis costs noted above do not include field work in connection with logging and coring of test holes. These costs are included in capital outlay for well construction. D. SUPERVISORY COSTS 1. Immediate Supervision $33,604.18 2. Divisional Supervision 50,596.82 $84,201.00 E. MISCELLANEOUS COSTS 1. Well I-A, Drilling and Development » Partial $1,870.02 2. Final District Report to State (reserved) 9,000.00 3. District Accounting 3,990.50 Total Miscellaneous Costs $14,860.52 ♦ Does not include contractual cost of drilling and developing gravel-packed replacement well I-A. Tliis contract cost of $7,156.15 was expended independently by the State. F. UNIT COSTS 1. Pipeline a. 10,000 feet of 10 ga., 20" welded steel pipe ; 1034 feet of 12 ga., 16" pipe and 1064 feet of 12 ga., 14" pipe — Total $125,443.77 b. Average Unit Cost $125,443.77 -^ 12,098 feet = $10.37 per foot. 2. 12" Recharge Well — Standard Casing a. Eight wells, 2,134 lin. ft. of casing— Total $37,021.70 b. Average Unit Cost $37,021.70 -^ 2,134 = $17.35 per lineal foot. c. Average Cost per Well of 266' average depth .$37,021.70 --- 8 = $4,627.71 3. 12" Oravel-packed recharge well (B) a. 240" depth of casing— Total Cost $7,009.20 b. Unit Cost $7,009.20 ^ 240 = $29.20 per lineal foot. 4. 12" (Jravel-packed recharge well (I-A) a. 258' depth of casing— Total Cost ♦ $9,026.17 b. Unit Cost $9,026.17 -7- 258 = $34.99 per lineal foot. • Includes state contractual cost of drilling and developing well I-A. 5. 8" Observation AVell — 2-ply casing a. 33 wells, 10.803 lin. ft. of casing— Total Cost $94,447.90 li. .\vorage Unit Cost .$!I4,447.90 H- 10,803 = $8.74 per lineal foot. c. Average cost per well of 328' average depth $94,447.90 H- 33 = $2,862.06 6. 2" Test holes, driven by District a. 14 wells, 1,497 lin. ft. of casing— Total Cost $6,.j63.50 b. Average Unit Cost $6,.563..50 -^ 1,497 = $4.38 per lineal foot. c. Average cost per well of 107' average depth $6,563.50 -H 14 = $468.82 7. Recharge Well Appurtenances a. Nine Wells (incl. well C)— Total cost $20,521.78 b. Average cost per well $20,.521.78 -- 9 = $2,280.20 8. Recharge Line Operation (excluding water costs, well sampling and engineering) a. Period Slarch through December, 1953. b. Total cost $36,589.58 c. Average cost per month $36,589.58 h- 10 = $3,658.96 The unit well costs are of particular interest to this analysis and included all costs related to logging and coring, and the collection of samples in the field. It also includes one field geologist's time in relation to such duties. The costs of the two gravel-packed wells were $29.20 per lineal foot and $35 per lineal foot, as compared to a non-gravel-packed well at a cost of $17.35 per lineal foot. However, in view of the increase in acceptance rate, the minimizing of maintenance difficulties and the overall efficiency of the well, a well designed gravel-packed well is considered economi- cally more advantageous. The increase in acceptance rate will permit increased spacing dLstance, thereby reducing the required number of wells. The costs of remedial work where a well fails due to the collapse of the clay cap, as was experienced at well G, will probably in itself more than offset the ditferential costs. Experience with the observation wells at the test site woidd indicate tliat an 8-ineh diameter ob.serva- tion well would be the minimum acceptable size if pumped samples and conductivity traverses are de- sired. This would also be the minimum size in the event a continuous recorder was to be installed to record ground water levels, in that a smaller well could not be used to contain adequate size floats, counterweights and clock weights. A 2-inch test hole to observe water surface elevation only is not recom- mended in aquifers containing considerable fine sand and silt in that the perforations of such a small well are subject to clogging and the reliability of water levels mea.sured therein are therefore subject to ques- tion. The 2-inch M'ells at the test site required frequent flushing and other rehabilitation measures. A 4-inch rotary-drilled hole is believed to be desirable and most economical for observation wells where water surface measurements only are required. Such wells were drilled as internodal wells at four locations and have, to date, given no difficulty. ■ SEA WATER INTRUSION IN CALIFORNIA 67 G. TEST COSTS AS RELATED TO OPERATING hh'uI would suflicc, tlien-by ivauciii- this cost to ap- PROIECT COSTS proximately $5,000.00 per mile. Field office, chlorinatinpr housiiifr, and chlorinating In order to adapt the test cost experience to a rou- equipment would also probably suffice for an entire tine operational barrier project, analyses of the perti- five-mile reach, and was proportioned accordin-rly. uent test costs were made, as outlined hereafter. Total Based on the above, the estimated capital cost per capital outlay and annual operation and maintenance ^^^g foj. gn operational recharge line is indicated as costs per mile of recharp:e line were $186,000 and follows: $32,000. respectively. CAPITAL COSTS Capital Costs l- P'-«-'i™i"'''-y pi''"">"k "■"' i>'-ig" •^^•"'^ " 2. Geologic luvp.stiKatioii b.OOO Experienced preliminary planning and design costs 3. Distribution ripeiine ^ ., , .^ . • »i r 11 5280 feet X $10.50/ft. 56,000 are considered to be approximately applicable. 4 Recharge Wells (gravel-packed, 250' deep) Right of wav costs were not considered in that this 7 x $7,. 500 each 52,500 factor would vary widely, in accordance with local 5- ^';2';"*jj2"2.^reach^I-!"Al^:I'^-^^^^^^^^^ 27.(M)0 conditions. 6. TransmiKsihUty Pumping Testa Geologic investigation and analysis for the test were ^ ^"^ ,^^'''"%^. *;•"'!' ''"'"'j ^'^^ . , ", , ^, ., , ., , 1 , • J J! 7. Recharge Well .Appurtenances considerably more detailed than would be required tor 7 weiis x ,$i,0(X) 7,000 normal operation. It is estimated that less than one 8. Jieasuring Equipmont 5,000 half of that expended for the test project would be ^ J^Z^^tS:^'':-"^!-:::: sffi sufficient for one mile of reach. Pipeline costs indicated in the following table as- To'al ^^^al^ T . -1. .• I- 1 1 4.1, 4. 4 ^A u 10% Contingency 16,900 sume a distribution line only and that water would be — . available at the distribution line. Although a perma- Total Capital Cost $185,600 , 1- , •. ,• ■ ^■ n 4.- • 4. Say, $186,000 per mile nent distribution pipeline for an operating project •" ' would be more substantial than that used for this test, . ,_ .. jii-x /-i . . ^ ,, ,, ^ ,, . ,. 1 c 4. c -41 Annua Operating ana Maintenance Costs It is felt that the cost per Imeal foot of approximately f^ » $10.50 is applicable. The pipeline sizes, anchorage and The operations personnel required for the short difficult underground problems, increased the cost of reach of this test would be more than sufficient for the the test project line to a point where its unit cost is operation of approximately five miles of recharge line, commensurate with a more permanent type of instal- in that the collection and analysis of the detailed data lation. collected for the test would not be required in the Under the assumption of aquifer characteristics sim- practical operation of a recharge barrier project. An ilar to those experienced at the test site, and for operating crew of some five trained operators to pro- reasons discussed in Chapter Y, Section D, it is felt vide for three shifts a day, seven days a week, would that seven recharge wells and twelve observation wells probably suffice. In addition, one Hydrographer would would suffice for each mile of reach. This would per- be required for such a reach to measure observation mit a recharge well spacing of 750 feet, observation wells and maintain recording equipment. The salaries wells at the internodal points, and one line of observa- for such a group, including an engineer for immediate tion wells normal to each mile of recharge line, as- supervision of operations and engineering control com- suming two observation wells oceanward and three putations, and including general supervision costs landward. A unit cost of $30.00 per lineal foot for totals some $45,000 a year, or $0,000 per mile of gravel-packed wells and $9.00 per lineal foot for ob- recharge line. Including power, telephone and other servation wells were used. related costs, the operational costs, as indicated below, The experienced costs of transmissibility pump tests have been estimated at $10,000 annually per mile of for tliis project are considered too favorable in that recharge line. bids for this work from other well contractors were Routine project maintenance costs as experienced approximately double those experienced at Manhattan at the test site could probably be reduced slightly on Beach, and it is doubtful that a similar contract with a larger scale and have been estimated accordingly. a driller could be repeated. Hence, a value of $750.00 It is estimated that recharge well development for per well has been assumed. a gravel-packed well would be required about one each Test experience indicated that a simpler well con- two years, at a cost of $1,000 for each operation, nection can be designed and that flow controllers are Water costs have not been considered in that such not desirable. Therefore a unit cost for well appurte- costs would vary widely with the availability of sup- nances of $1,000.00 each could probably be realized. plies and must be evaluated in relation to the benefit It is also estimated, in contrast to the test, that less of such a supply as a source of replenishment to the than one half of the measuring and recording equip- basin. 68 SEA WATER INTRUSION IN CALIFORNIA Control data required for operations could be mini- In addition to the real value of 30,000 acre feet, a mized as compared with those experienced during the value of considerable proportions must be considered test, as indicated in the following cost analysis : in the loss of perhaps some million acre feet of basin ....V..... »n.-r,.-r.v.^ -^r-r/. storage capacity. The present import facilities of the ANNUAL OPERATING COSTS ,, , '" ,./ ,,; + r,- . ■ ^ i, • «; • * , ,, Metropolitan Water District have insufficient capacity Operation and Mninfcnance , . , „ . , , n •, i i , , ,, . . . /^ .• 1. , ffinnnn to provide lor either peak daily demand or peak sea- 1. Suiiervision ami OperatiDiis Personnel $10,000 ^ n ,t 2. Routine Project Maintenance and sonal demand; hence surface Storage would have to Miseellanei>us Materials 4,000 be provided in the event ground water supplies were '• "^ ;.t w^nsTi^SKTen"! 3,500 destroyed. Although sufficient data are not at hand 4. Chlorine (5 cfs, 8 ppm) 4,500 to accurately evaluate this required storage, it is obvi- Engiueeriiig Control Data ous that an economic analysis of the costs involved in 1. Well Sampling Operations 5,000 developing such surface storage considered in rela- 2. Water Analysis and Related Labora- tJon to the COStS of a pressure barrier would result tory or- ' in a definite economic benefit in favor of the barrier. Total $29,000 The costs for such storage have been estimated by the 10% Contingency 2,000 ^Tp^t Basin Water Association to range from $-4,500 Total Annual Operating Cost ____ $31,000 tO $18,000 an acre foot. Say, $32,000 per mile rp^g ^^^^^ ^j ^ ^^^^^^ Supply could be Considered Econom/c Jusf;fica//on on ^he basis of its value as recharge to the basin. Application of the costs itemized above to a pro- thereby permitting a greater safe pumping yield from posed sea water barrier, and the economic iustification ., , • tt ■ ..■ ^ n V ,-. ij • nil- the basin. Hence, assuming a very conservative value or such expenditures, would require a carerul analysis of each item to fit conditions existing in tlie area con- ^^ ^^^ per acre foot of ground water, any cost of sidered. If it were assumed that the aquifer conditions water less than this wonld provide a definite addi- along the entire coast were similar to those encountered tional economic return. For example, it has been esti- at the test site, the application of these costs and the mated that some 52,000 acre feet per year is pres- economic justification thereof in relation to a complete ^^^^j^, intruding as ocean water. In order to control barrier in the West Coa.st Basin might be analvzed, , . ". . : . , . , ,. , , ,, , ,-- „/^r. as indicated in the following. *!"« intrusion, it might be estimated that some /.xOOO Such a barrier would ultimately insure the annual ^^^'^ ^'^^ ^ ^^^^^ '"'^ ^« required. If such a water safe pumping yield from the basin. This yield has supply could be purchased at $10 an acre foot, then been established, at present as approximatelj- 30,000 an economic benefit to the basin of $5 per acre foot, acre feet per year (West Coast Basin Reference by or a total of $375,000 per year, eould be realized. State Engineer, February 1952). It is obvious that if some protection is not provided, the entire resources Considering these costs in relation to the extension of the basin will eventually be destroyed. of the existing recharge line for the entire eleven As indicated below, the costs of water have not been miles of affected coast line in the West Basin, it is considered, in that the protection of 30,000 acre feet presently indicated that due to the limited thickness of annual production alone would, under the assump- of the aquifer to the north, the cost would be con- tions made above approximately justify the capital .iderablv lower than those presented for an average outlay and annual operating costs. , ,4,, ., , , , , .„ , reach. While to the south where the aquifer becomes Annual Costs vs. Annual Benefits deep, quite permeable and uneonfined. it is antici- Anuuai Operating Cost patcd that the costs would be considerably liigher. $32,000 X 11 miles of reach $352,000 1,, , , , , , , ' , , Annual Costs of Capital Recovery Throughout the Southerly reach, recharge through .$i4,!KK)*x 11 miles of reach _: ^64,000 seepage pits aiid/or surface basins would become a Total Cost $516,000 consideration. This would dictate a need for further Annual value of 30,000 A.F. at geologic exploration and a study of the applicability $21/A.P. less pumping costs (!t;4..'-.0)t $495,000 „ ., . . j * * *i, ui * . An analysis of the capital outlay items, assuming lO-year MN for recharge «,Is "^ ^^^ experiences tO date tO the problem of creating and apiiurtenances and measuring equipment; 20-year life for the halance of the „,, pffp,.tivp hni-riiM- ill tlnp niipnnfinpfl nniiifpr readips replaeeahle items: and with Interest at 4 percent per annum. Indicates an annual "" lUttTUP DailKI ill ine UUCOnnnea aqUlier rCdCneS capital recovery cost of $14.fHI0. tj j virlr» "R I t Based on an average pumping lift of 150 feet at a co.st of $0.n3/A,F./fl. of lift. near I\eaonuO IseaCU. CHAPTER IX FINDINGS, CONCLUSIONS AND RECOMMENDATIONS It may he oonrhuled that tlu' subject iiivestiga- tioual work for the prevention and control of sea water intrusion has established that for areas with similar f^eolojrie. hydrologie and topojrraphie condi- tions as those found at the test site in Manhattan Beaeh in the "West Coast Basin: (a) Sueh prevention and control can be success- fully realized in a confined coastal aquifer by recharge through wells. (b) Such recharge can pressurize a confined aqui- fer continually through a given reach, thereby reversing any pre-existing landward gradient and preventing further sea water intrusion. (c) Such recharge will provide significant re- plenishment to the inland ground water basin with only an inconsequential loss of fresh wa- ter oceanward in relation to the total quantity of injection. (d) Such recharge can be performed in an aquifer previously degraded by sea water intrusion and, within the physical and hydrologic limita- tions as established at the test site, will not cause any consequential deleterious effect on inland pumped supplies. In fact, all evidence collected to date indicates that the degraded portion of the aquifer can be reclaimed by re- charge through wells. The project was located in the cities of Manhattan and Hermosa Beaches, Los Angeles County, Califor- nia. A line of recharge wells covering a reach of some 4500 feet in length was installed approximately parallel to and some 2000 feet inland from the Pacific Ocean coast line. Nine recharge wells, spaced at aOO feet along the recharge line, and thirty-six observa- tion wells, located along, inland and seaward of the line, were originally drilled. Eighteen additional observation wells were added as the test progressed. The wells penetrated a pressurized aquifer confined at the top by an impervious clay cap slightly below sea level and bounded at the lower limits by rela- tively impervious sediments some 110 feet below sea level. The ground waters at the test site were com- pletely degraded by sea water intrusion while de- creasing effects of intrusion extended approximately 5000 feet inland. Treated Colorado River water, im- ported by the Metropolitan Water District, was used for injection purposes and was piped from a con- nection with the El Segundo branch of the "West Basin Feeder located some 7300 feet inland of the recharge line. Prior to recharge, sea water was flowing landward under the influence of an inland ground water gradi- ent. As injection was initiated at the first well, a pressure cone formed, centered at the well. The pressure effect extended almost instantaneously in all directions and within a few hours was noted 1000 feet distant from the well. As injection was initiated at adjacent wells, the pressure cones overlapped, forming a pressure ridge along the recharge line with alternate peaks and valleys. As stability along the entire line was reached, the cones merged into a stable, constant landward and oceanward gradient at a distance of some 250 to 500 feet from the re- charge line. At this distance the individual pressure cones lost their identity and the shape of the mound was essentially uniform. This ridge reversed the pre- existing landward gradient between the recharge line and the ocean, and prevented further sea water in- trusion. Initially, fresh water moved radially from the re- charge well and finally merged with flow from the ad- jacent wells. After moving some 250 to 500 feet from the recharge line, all radial movement was trans- formed into essentially landward and oceanward flow. As the gradient oceanward was relatively flat, move- ment oceanward was extremely slow, while the steeper landward gradient maintained relatively higher ground water velocities landward. As the fresh water advanced, it overrode and displaced the pre-existing saline ground water. Although the pressurizing effect of injection was relatively rapid and further sea water intrusion was thereby prevented, the movement and actual mergence of the injected fresh water was rela- tively slow. Detail findings, conclusions, and recommendations are enumerated as follows : Project Faciiifies 1. Experience with the well types used during the test indicated that a gravel-packed recharge weU with a minimum diameter for the gravel en- velope of 20 inches is desirable in aquifers simi- lar to those found at JIanhattan Beach, where fine to medium sands may be encountered with limited gravel stringers. In aquifers composed of such non-homogeneous materials, all critical por- tions should be penetrated to insure an effective and rapid pressurization. (69) 70 SEA WATER INTRUSION IN CALIFORNIA 2. In addition to gravel-packing, the recharge well casing should be of sufficient diameter to permit the entry of standard sized well tools. In this connection the 12-ineh casings used at the test project proved adequate. In the event no rehabil- tation work was anticipated, a somewhat smaller casing could be used. 3. Test holes through the uneonfined surface sands and bottomed near the top of the ' ' clay cap ' ' im- mediately adjacent to the recharge wells, gave evidence of some leakage from the confined por- tions of the aquifer. This leakage apparently occurs along the well casing through the confin- ing clays as a result of the casing having per- forated the clay and having left a zone of weak- ness along the easing. The test established that the construction of a proper cement seal is re- quired through this zone of weakness. 4. Observation well diameters should be in relation to their anticipated use. If water stage recorders are to be installed and/or pumped water samples with conductivity traverses are to be obtained, then an 8-inch casing is a minium required di- ameter. If water level elevation measurements only are to be obtained, a 4-inch observation well is recommended as a minimum size. 5. All observation well perforations should be care- fully located and cut in relation to the position and extent of gravel stringers encountered in the aquifer. The need for this care is obviated in a gravel-packed well, which should be perforated to permit a maximum discharge with a minimum velocity in the perforations. 6. Development by pumping and/or sand bailing should be limited in order to avoid the creation of voids in the aquifer by the removal of excess quantities of sand. Such limitation is not as important in a gravel-packed well in that the re- moved sand will be replaced by gravel. 7. Operations during the test showed the need for accurate control of the recharge flow. Hence, if the supply is subject to pressure variations, the installation of a pressure regulator on the main distribution line would be recommended. A man- ual control valve with a reliable flow meter at each well would probably provide the simplest and most economical flow control mechanism. Each well would require independent control of recharge in relation to its acceptance rate and the transmissibility characteristics of the local aquifer. In addition, a well conductor pipe from the surface of the ground to the ground water level should be provided, with means of prevent- ing aeration. 8. Although the chlorine equipment used at the test was provided with automatic chlorine feed con- trol and a chlorine residual recorder to facilitate adjustment to the frequent changes made in both recharge and chlorination rate, in a practical operation, after initial stabilization, both re- charge and chlorination rates would be expected to remain fairly constant. Hence, manual feed control and occasional field tests to check chlorine residual would suffice to operate a stabilized pressure barrier. Geology 9. The aquifer at the recharge site is extensive, con- fined, heterogeneous, and pressurized (the latter imperfectly). It is probably similar to some aquifers occurring along other seaward margins of coastal ground water basins of California. At Manhattan Beach the geologic sections defi- nitely showed correlation between the Merged aquifer at the coast with important inland water- producing aquifers. 10. The confining "clay cap" has been eroded near the sea coast some 2000 feet oceanward of the recharge line, which may have provided a limit to pressurizing and resulted in a delay in the stabilization of the seaward gradient by permit- ting storage in the uneonfined portion of the aquifer. 11. A zone of sharp flexuring or faulting appears to exist oceanward of the recharge line. Its effect upon the test results is probably limited to a lo- calized reduction in transmissibility. This zone has probably somewhat restricted sea water in- trusion in the past, and caused a reduction of normal oceanward flow during initiation of re- charge. 12. All wells drilled in connection with recharge should be carefully logged and analyzed in re- lation to the general geology of the area in order to: (a) determine the structural effectiveness of the confining cap of the confined aquifer; (b) determine tlie physical limits and homogeneity of the aquifer in relation to the construction of recharge wells and to its hydraulic parameters; (c) identify the correlative formations of the im- portant inland water-bearing aquifers; and (d) detect special conditions within the aquifer which may accelerate, limit, or prevent ground water flow in a given area or direction. Hydraulic Aspects of fhe Recharge Line V,i. The total required recharge rate per foot in a long line of recharge wells is the sum of: (a) oceanward flow from the recharge line; (b) ini- tial ground water flow landward, i.e., prior to I SEA WATER INTRUSION IN CALIFORNIA 71 rei'liai-fre; and (c) increased jrround water flow liiMihvard due to reeliarp-e. Wliereas items (a) and (b) depenil on nieasnrable gR)nnd water ele- vations, afiuifer constants and the physical di- mensions of the a(|Mifer, item (c) evidently de- pends upon a (luantity unmeasurable before re- eharjre beprins. At the test site the increased gra- dient and therefore the increased recharge rate required was about 60% greater than the rate of sea water intrusion under the existing landward gradient prior to recharge. This additional re- quirement may be due to an inland storage de- mand or to leakage through imperfectly confined portions of the aquifer. 14. The total head measured in a recharge well con- sists of two ]iarts — tlie mound elevation and the injection head. Mound elevation is that elevation of the ground water piezometric surface which is created adjacent to the recharge well due to recharge operations. Injection head is that head reciuired in the well to overcome the energy loss occurring when the injected water passes through the casing perforations and enters the aquifer face. (The aquifer face may be defined as the vertical undisturbed surface of the aquifer ex- posed to flow from the recharge well.) 15. The mound elevation depends upon the amount of water being injected in a given reach, the spacing of the recharge wells, the transmissi- bility of the aquifer, the ground water gradient as it would have been if unaffected by recharge, and the degree to which this later gradient may be increased by recharge. 16. Injection head is dependent upon the quantity of water being injected, the type and size of re- charge well, the number of perforations, and the local transmissibility characteristics of the aquifer face immediately adjacent to the re- charge well. 17. At the test site at Manhattan Beach it was found that the average transmissibility determined by pumping tests prior to recharge was about 0.165 cfs ft. under a unit gradient; that during re- charge it was about 0.18 efs/ft. per unit gradi- ent ; that the average landward gradient before recharge commenced was about 0.0041 feet/foot; that the apparent porosity landward of the re- charge line averaged 27% ; ajid that based on a stabilized landward gradient of 0.0065 feet/foot during recharge, the rate of landward movement of injected fresh water was about 1250 feet per year. 18. Maximum acceptance rate tests at well E indi- cated rates up to 1.86 cfs. Although the maxi- mum acceptance rate of well I-A was not de- termined, it indicated even more efficient accept- ance characteristics at the lower ilows experi- enced. Rates of 0.5 to 0.75 cfs were maintained for long periods at the non-gravel-packcd wells. Hence, with the aquifer characteristics as ex- perienced at the test site and a well designed and constructed gravel-packed recharge well, it may be expected tiiat a well acceptance rate of 1.5 cfs could be maintained. 19. The spacing of recharge wells can be determined by estimating the unit recharge requirement and the expected acceptance rate of the wells. As a result of the findings of this test, it is believed that the installation of recharge wells should be based on a sequence of well installations. As each well in an assumed uniform transmissibility reach is completed, a pumping test should be made to determine anj' error in assumption of the value of transmissibility. If the transmissi- bility has changed to any extent, the adequacy of the original estimated spacing should be reevalu- ated and, if necessary, a new spacing adopted. 20. Test experience indicates that the number of ob- servation wells required to control the operation of a barrier mound is limited to internodal wells and occasional sets of wells (perhaps one set each mile) landward and oceanward of the recharge line. An occasional observation well adjacent to a recharge well is recommended to measure maxi- mum pressure against the confining strata and to determine the actual injection head in relation to the well's efficiency in acceptance. 21. Pumping at nearby well fields, unless the cone of depression approaches the line of recharge, does not affect the required rate of recharge except as it contributes to the depletion of the supplies of the basin and results in an increase of the land- ward gradient in the basin, thereby creating a continuously increasing recharge rate to main- tain the sea water barrier. Pumping at nearby producing well fields, where the resulting cone of depression is suffi- ciently remote (6500 feet at the test site) no immediate effect is apparent at the recharge line. However, as such pumping contributes to the general depletion of the ground water supplies and increases the landward gradient in the basin, the required recharge rate necessary to maintain the barrier will increase. In the event pumping was sufficiently close and the cone of depression approached the recharge line, it would be ex- pected that the influence would be immediate and require a corresponding increase in recharge rates. 22. Results of conductivity traverses taken in the project observation wells corroborated previous studies performed by the University of Call- 72 SEA WATER INTRUSION IN CALIFORNIA fornia at Berkeley which showed that sea water intrudes as an uiidorriding wedge ; however, the traverses indicated that historically some mixing had taken place and that prior to recharge the existing waters at the recharge line and for over 4000 feet inland, were saline to a varying degree. The traverses also showed tliat injected fresh water moved across the more dense native water as an overriding wedge. The traverses further showed that there was a minimum of mixing be- tween the injected fresh and i^artially degraded native water. 23. In connection with the initiation of fresh water injection into an aquifer degraded by sea water intrusion, a minor "saline wave" is produced which moves landward from tlie recharge line. As it moves into areas of the aquifer which are not as highly degraded, the magnitude of its characteristically higher saline content is rapidly dissipated. At the test site, the saline wave was indistinquishable at well G-8, a distance of 1180 feet inland of the recharge line. Water Qualify and Mainienance of Aquifer Transmissibility 24. If fresh water recharge is initiated into an aqiiifer containing sea water, a cation exchange reaction between the sodium-saturated aquifer sediments and the calcium and magnesium ions of the injected water will occur. Thus, if an in- ferior water (inferior due to hardness) is used as a source for recharge supply, a softening effect may be expected so long as the sodium concen- trations will act to replace calcium and mag- nesium ions in the recharge water, therebj' mini- mizing any possible degradation of the native ground waters. 25. Ultimately, it may be expected that the injection of fresh water will increase the aquifer's trans- missibility slightly through cation exchange. Al- though the indication of this effect has not occurred during the limited period of the test, ultimately a corresponding slight increase in re- charge rate may be required. 26. Tests established the presence of many "slime- forming bacteria" in the aquifer. Chlorination of the recharge water is necessary to inhibit the growth of such slime-producing organisms which, if uncontrolled, will rapidly reduce the acceptance rate of the well. The necessary chlo- rination rate may vary with individual recharge waters and aijuifcrs. At the test site, it was found that initially a chlorine dosage of 20 ppm was sufficient to prepare the aquifer for con- tinued injection, and that the minimum neces- sary sustained chlorine dosage was more than 5 ppm, but less than 10 ppm. 27. The etfect of su.spended solids was not evaluated in that the recharge waters were almost devoid thereof. Other chemical effects, such as deteriora- tion of well casings and facilities due to rust, were also not evaluted due to the relatively lim- ited period of operations. The effect of entrained air in relation to air-binding and stimulation to the growth of slime-forming organisms also re- quires further research to evaluate. Reference is made to Chapter VI, E for a brief discussion of these factors. Maintenance and Operational Problems 28. The test demonstrated that overdevelojunent of a recharge well in coastal acpiifer materials could be serious in that voids could be created within the aquifer and that during recharge op- erations a structural failure of the confining clays adjacent to the casing could occur with a subsequent major subsidence of the overlying materials. Test experience indicates that the danger of subsidence is minimized in a gravel- paeked well. 29. The cement seal at the clay cap, mentioned in Conclusion No. 3, will probably prevent or at least minimize leakage along the well casing and subsequent structural failure of the clay cap adjacent to the well casing. Although the danger of subsidence in a gravel-packed well is more remote than in a non-gravel-packcd well, the proper sealing of the claj' cap at the well casing to prevent leakage is considered essential re- gardless of the type of well installed. 30. Experience with the project wells showed that rehabilitation of a well, where failure of the clay cap and later subsidence has occurred, can be difficult and costly with no assurance of success. 31. Rapid changes in recharge rate subject the con- fining clay cap to excessive pressures and/or erosion adjacent to the well casing Avhere a zone of weakness may permit excessive leakage. Hence, it is recommended that all changes in injection rates be made in small increments at a sufficient time interval to permit local equaliza- tion of pressures created by injection. 32. Althovigh the necessity of redevelopment as dis- tinguished from rehabilitation did not occur during the test period, following the test, well K was successfully redeveloped by moderate surging and bailing. Due to the preponderance of fine to medium sand in the aquifer, redevel- opment by pumping is not recommended in a non-gravel-packed well. SEA ^YATER INTRUSION IN CALIFORNIA 73 Costs :>3. The test's costs cited in Chapter VIII are con- sidered to be typical of those required for a siinihir research project that might be consid- ered for an area where it is desired to evaluate detailed controls and conditions relative to re- charijre for pressurizing and/or replenishing a ground water aquifer. Omitting right of way and water expenditures, the costs have been analyzed in relation to those required for developing, constructing and op- erating a practical barrier operation (i.e., omitting the colleetion of the detailed control test data required in the research project). 77) IS analysis ivas iased on a minimum reach of five miles, assiiminri aquifer charaderisfics and ground water overdraft conditions similar to those at Manhattan Beach. On this basis, the capital outlay has been estimated to be about $186,000 per mile and the annual operation about $32,000 per mile. Considerations of right of way expenditures and water costs were omitted in that these would vary w-idely with local conditions, e.g., right of way costs for the test project were negligible; water costs were $20 per acre foot. In ground water basins where significant over- draft exists and imported supplies are limited to rates of delivery less than peak consumer de- mands, the benefits to be realized from a pres- sure mound barrier to sea water instrusion are : (a) the protection of the existing safe yield of potable water from the basin; (b) the value of replenishment provided through recharge; and (c) the prevention of sea water intrusion and the resultant preservation of fresh water stor- age. Such benefits can economically ju.stify the expenditures entailed in constructing and op- erating the barrier facilities. The latter benefit must be evaluated by considering the costs of surface storage which would be necessary if pumping from the aquifer to meet peak de- mand were not available as well as the intangible value in relation to a safe emergency water supply. 52568 ACKNOWLEDGMENTS State Water Resources Board and Division of Water Resources The Los Angeles County Flood Control District hereby expresses its appreciation to the State Water Resoiirees Board for the eontraet offered tlie District to perform the investigational work for the prevention and control of sea water intrusion and pratefully acknowledgres the excellent cooperation of the Members of the State Water Resources Board and the technical assistance of the followinfr personnel of the State Divi- sion of Water Resources who, through an exchange of pertinent engineering and geological data as the test progressed, aided inuneasureably in the success of the experimental field project. State Supervisory Personuel A. D. Edmonston State Engineer Secretary, State Water Resources Board Harvey 0. Banks Assistant State Engineer Max Bookman Principal Hydraulic Engineer Jack J. Coe Associate Hydraulic Engineer Elmer C. Marliave Super\'isiug Engineering Geologist State Geological Staff Raymond C. Richter_Senior Engineering Geologist Lawrezice B. James Senior Engineering Geologist Donald L. McCann Assistant Engineering Geologist Philip J. Loreus Assistant Engineering Geologist United States Geological Survey The cooperation of the United States Geological Sur\-ey and the loan of conductivity apparatus by that agency, during initial stages of the test merits special acknowledgement, particularlj- to : Joseph F. Poland District Geologist Arthur A. Garrett Engineer District Engineering and Operations Personnel In addition to the authors of the report and report appendices, the following District personnel contrib- uted to the final successful completion of the test and development of this report through their field and office work on the investigation. Arrigo, Joseph H. Bauer. Frank R. Castro, Rose M. IConnell, Garland Jr. Davis, Everett Dysart, Ben W. Falconer. John C. Farmer, Thomas C. Gillam, Harold R. Greenberg, Harry Groff, John Hershkowitz, Paul N. llirschkind, Richard W. Ilylen, Walter W. Johnson, Mildred B. La Bahn, Edward Lutz, Lawrence J. ]\IcLeod, Herbert J. Meyerhofer, Steven J. Osborne, LeRoy R. Remington, Staidey Roth, John Scribncr, Robert K. Smith, Elva Spencer, Edwin N. Tanaka, Hajime Volpc, Thomas J. Watkins, Philip S. Wood, Donald Wood, Verne M. In addition to the foregoing, personnel from the instrument repair shop of the Hydraulic Division and from the Design, Survey, Right of Way, Right of Way Acquisition, Construction, Operation and Maintenance, Communication and Testing Divisions all cooperated in the final design and installation of the project facil- ities. Other Public and Civic Agencies The coordinated efforts and splendid cooperation of the following public agencies were vital to initiation, progress and continued operation of the barrier test : West Basin Water Association Metropolitan Water District of Southern California City of JIanhattan Beach City of Hermosa Beach City of El Segundo Other Cooperative Organizations The cooperation of the Atchison, Topeka and Santa Fe Railway Company, in granting, without payment, a permit for installation of the test facilities along the railroad right of way is gratefully acknowledged. Special acknowledgement is due the following or- ganizations who gave freely of their time, experience and advice : Roscoe-^Ioss Company Peek and Sons, well drillers C. A. Rock, well drillers Wallace and Tiernan, Inc., who additionally fur- ni.shed the loan of expensive .special chloriuation equipment at no cost Sparling Meter Company The interest and cooperation of the following organ- izations, in furnishing water production records, water surface elevations, and water quality tests, is also gratefully acknowledged : California Water Service Company Standard Oil Company of California General Chemical Company (75) BIBLIOGRAPHY 1. Moiuloiiliiin. \V. C, "Dcvolopmeiit of Underground AVaters iu the Western Coastal Plain Region of Southern Cali- fornia," U.S.G.S. AVater Supply Paper No. 18!), 1!K)5. 2. Dockweiler. .T. H., '"Comprehensive Plan — Flood Control and Conservation — Nigger Slough Project," Los Angeles County Flood Control District unpublished report, 1932. 3. Conkling, Harold, "Report to West Basin Water Associa- tion, an Imported AA'ater Supply for West Basin, Los Angeles County, California," July, l!t46. 4. Poland. .1. F., Garrett, A. A., and Sinnott, Allen, "Geology, Hydrology, and Chemical Chai-acter of the Ground AVaters in the Torrance-Santa Monica Area, Los Angeles County, California," V.S.G.S., 1948. 5. California Department of Public WorUs. Division of Water Resources. "Sea AA'ater Intrusion into (iround AA'ater Basins Bordering the California Coast and Inland Bays," AVater Quality Investigation Report No. 1, December, 1050. C). Arnold, C. E., Hedger, H. E., and Rawn, A. M., "Report upon the Reclamation of AA'ater from Sewage and Industrial AA'astes in Los Angeles County, California," 1949. 7. Laverty, F. B., Jordan, L. AA'., and van der Goot, H. A.. "Tests for the Creation of Fresh AA'ater Barriers to Pre- vent Salinily Intrusion.' District, 1951. Los Angeles County Flood Control 8. California State Lcgi-slative, Chapter 1500, Statutes of 1951. 9. California Department of Public AA'orks Division of AA'ater Resources, "Proposed Investigational AVork for Control and Prevention of Sea AA'ater Intrusion into Ground AA'ater Basins," Report to State AA'ater Resources Board, August, 1951. 10. Harder, James A., "Final Report on Sea Water Intrusion," Sanitary Engineering Research Laboratory, L'niversity of California, 1953. 11. Cooper, H. H., Jr., and Jacob, C. E., "A Generalized Graphical llethod for Evaluating Formation Constants and Summarizing AVell-Field History," Transactions, American Geophysical Union, A'ol. 27, No. lA', August, 1946. 12. Theis, C. A'., "The Relation Between Lowering of Piezome- tric Surface and Rate and Duration of Discharge of a AA'ell Using Groimd AA'ater Storage." American Geophysical Union, August, 1935. 13. Muskat. JI., "Flow of Homogenous Fluids," llcGraw-Hill Book Company, Inc., 1937. (76) PHOTO PLATE 9 WEST BASIN BARRIER TEST Chlorinator Room I AMafiis ceuHrr Flood Comtbol Dtaratc 10-7-53 1 2i45 m 176 R 12.2-B WEST BASIN BARRIER TEST Chlorinator Room PHOTO PLATE 1 M-> *^ >!►• Loa AnoiLia Couwtt r^ooo Coht«d(. Distsict 'I, '3-21-52 1 905 AM 1 "° 176 N 1.1-B ,r WEST BASIN BARRIER TEST Drilling Operations, Well G-2 ■■•'"•«« LOB AHOILI* COuHt* fLOOO C0«.»»01. D>»>«rCt " 3-a-52 1 lOiOO Am| "° 176 H 1.6-B WEST BASIN BARRIER TEST View of Boilef, Well G-4 PHOTO PLATE 2 Loa ANacLC* CouM-rr Flood Comtmol O'stmict 3-21-52 I lOilO *H 176 H 1.9-B WEST BASIN BARRIER TEST Mills-type Knife Perforator 176 II X.U-1 WEST BASIN BARRIER TEST Core Sampling Barrel Assembly PHOTO PLATE 3 **!: >* ,r^ Lo* Ambilc* Counrv rt.ooo Cohtnol OiarMier 10-23-53 ^lOilJ IM 176 II U-l-B WEST BASIN BARRIER TEST Well lA, Well Development Spreading Areas With 30" Seepage Holes -il^.-^^ ■*T' t;"- J^ • 1 .Si. xJ3l ^'^^v.x:;r'\-i^^/^ ■1 ^.^•- •-^•.l^ t.Oa AMQCLI* COUHtT rLOOO CDH*«Oi. 0>«t*lCT 12-5-52 ISiiQ FM 176 N 6.?-B WEST BASIN BARRIER TEST Transmissibility Pumping Test, Showing Weir Box PHOTO PLATE 4 I I KB^H'-^*^ I.US AHDTLCk CuuNTi rt.oao CONTaoc Oi9Ta>cr i-A-52 1 1:35 PM I 176 N 2.2-B WEST BASIN BARRIER TEST Development, Pumping Well G fF >A- :-i*!!«6*?*'¥ .■i'-''V*r w^' •=Jpf^' iH "va-M 1 uioo at \ "'iTe ■ i.i5-» WEST BASIN BARRIER TEST Observation Well G13 PHOTO PLATE 5 2-2i-i3 10:40 «M 176 II 8.1-B WEST BASIN BARRIER TEST Well C, Showing Meter, Metro Valv and Recorder 176 I 13.tr* WEST BASIN BARRIER TEST Well G, Rochorge Location, Showing Typicol Installation PHOTO PLATE 6 i/ r =;; pr=: -^T-is 2-2A-53 I UUO *M ' 176 II B.6-B WEST BASIN BARRIER TEST Showing Backpressure Valve Header, Well G riHi smmm (.O* AMVCLia COUMTY FlOOO COHTSOl. O'VTVIOT 'lO-2>}3 I "llltf ill I " 176 » U.VB WEST BASIN BARRIER TEST Somple Pump Operation ai Observation Well K'12 PHOTO PLATE 7 f'"''^irwig ''^''"WBff '"* *''ii' iii'>i' 176 N 13. 2-] WEST BASIN BARRIER TEST "Thief" Type Woter Sampler PHOTO PLATE 8 L. A COUNTY FJ-OOD CONTROL DIST. 10-21-5J "2:35 m Mo N 13.1-B WEST BASIN BARRIER TEST Main OfRce, Manhattan Beach Storage Shed for Chlorine Tanks La« AMOCkIS COUMTV rLOOe COMTM^ OlSTBiCT '"a-2 S-53 I 1120 W I "°176 H 9.1-B WEST BASIN BARRIER TEST New Office Building, I 1th and Ardmore Sts., Manhattan Beach F. C. D BOUMDARY i^ PLATE I ^ WE.ST BASIN SALINITY H£Y WELLS LOS AM6E.LtS COU WTY FLOOD COWTROL DISTRICT WEST BASm 5ARRIER TEST PROJECT LOCATION PLAN \ \ PLATE I L AC. F. C. D . eouMOAiiy \ • WtST BASIN SALINITY ICEY WELLS fa a 10 LOS AN6ELES COUMTY FLOOD COWTROL DISTRICT WEST BASIN BARRIER TEST PROJECT LOCATION PLAW PLATE 2 3200 1938 LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST CHLORIDE CONCENTRATION OF TYPICAL WELLS POLLUTED BY SALINE INTRUSION- WEST COAST BASIN FROM 1938 TO 1954 PLATE 3 31 1299 ® C309S* 1 J> "<"■> * ,309*1 7 1309 H « V I309F* 1309 L ® r 1309 Q 'y 1309 eO 9 *l309Ji^ 1309 M® e 1309A ^^ 9 „ I309< / I309N 1 \ MB-50 4)''' 1 \a\ y V\\ ^ •I \\^ -V^ ^ 1 \V\ x"'''''''*'''^ ° J \V\ ,,;^^''''^ MB-12* 1 \\N\ \S^r^ 1 \v\ >^ *-!<• a f WW ^ ^ \W / 3 \\V X % \\v 1 " \\^ \ , EL SEGUNDO Wvv^ \ I MW.D. FEEDER WW*. \ MANHATTAN BEACH BLVD. (i WyC* \ ^^^e "^"'^ MB-ll® \\\\-»- ^\\,^^^ & ,. •-^^ ■• Wvv ji^cr •» / ^\ > w ^ ••^6^ •'*' ^ ABANDONED MANHATTAN o \VA^ •, 6-* \>r\ BEACH WELL FIELD ^ \\w_ »^ • PROJECT OBSERVATION WELLS Ww^i* ^^^^ ^^ \ ^**^ 722 8* K M-i4a GOULD M-ie« lane © PRIVATELY OWNED WELLS \V\ \ \\ \ 7124 7I2B © CITY OF MANHATTAN BEACH WELLS UVk'^ \ A \ 712 "^'^l^ « STANDARD OIL COMPANY WELLS VvX-x \ \l TI2 « • CALIFORNIA WATER SERVICE CO. \\\* \ ft \ • LAC.FCD. TEST MOLES \\\° \ S '• * GENERAL CHEMICAL CO. WELLS \V\» \ * « ^ LOS ANGELES COUNTY %- \ \ \ FLOOD CONTROL DISTRICT W\n> \ ^ WEST BASIN BARRIER TEST — - PIPE LINE \\Vl* ft \\ ft*^ SCALE In FEET LOCATION MAP U'X 5000 4000 SOOO 2000 1000 PLATE 4 <*w«>«r/'*>'w««wjk< kt^Vi9^\^V^J.,WJ^WJJ.'i. W ^A^''^ FEEDER LINE LATERAL GATE VALVE DRESSER COUPLING METRO VALVE — MAIN LINE METER AND HYOROVALVE MERCURY GOVERNOR INDICATOR- TOTALIZER RECORDER WELL HEADER BACK-PRESSURE VALVE AIR AND VACUUM RELIEF VALVE AMD GAUGE LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST TYPICAL RECHARGE WELL ASSEMBLY llJ ill u. UJ o £ 3 > UJ -1 (n 1 rr UJ li 40 - < 7 * < UJ 1 ? o 1 z o 2 1- 1- UJ •10 ■15 -20 -25 -30 -35 PLATE 5 TOTAL WELL PRODUCTION CALIFORNIA WATER SERVICE WELLS 7l2-A,B,CaG 722-BaC 731 CITY OF MANHATTAN BEACH WELLS 9, II, 13 a 14 MEAN WEEKLY, CFS TOTAL WELL PRODUCTION FEB MAR APR 1953 MAY JUNE JULY AUG SEPT OCT NOV DEC JAN FEB 1954 RATE LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST HYDRO GRAPHS OF PROJECT WELLS, OF PUMPING IN IMMEDIATE VICINITY OF PROJECT, RATE OF PROJECT RECHARGE AND PLATE 6 100 f 80 LU U. UJ _J UJ S 60 3 _I U> < b^ 1- < 2 * < 40 UJ 11 2 o 1 S 7 D o t 5 < ° 20 > UJ 1.0 2" .6 UJ UJ .4 i^ .2 o -20 < z a: O _i I o 20 10 •a GROUND SURFACE ELEVATION 112 WELL D AVERAGE DAILY WATER SURFACE ELEVATION IN RECHARGE WELL Irv^VV^ GROUND WATER PIEZOMETRIC SURFACE 20' FROM RECHARGE WELL AVERAGE DAILY INJECTION RATE Zr- SHOCK CMLORINATION CHLORINATION RATE JAN 1953 FEB MAR APR MAY JUNE JULY AUG SEPT OCT ~ NOV DEC JAN FEB MAR APR ' MAY ' JUNE 1954 LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST HISTORY OF OPERATIONS AT RECHARGE WELL D PLATE 7 100 80 1- lll III u. u' _l o liJ -_ 60 u. u u -I llJ z «) ? a: z o E 3 -20 20 2 Q. a. ,.r 10 GROUND SURFACE ^ <\\W//X\\\\V GROUND WATER PIEZOMETRIC SURFACE- 20' FROM RECHARGE WELL WELL E-l- j-f- l -FAILURE OF FLOW-RATE CONTROLLER AVERAGE DAILY INJECTION RATE MAXIMUM INJECTION TEST AT WELL E SHOCK CHLORINATION JAN FEB CHLORINATION RATE - I9S3 MAR APR MAY JUNE JULY AUG SEPT OCT NOV T DEC JAN FEB MAR APR MAY JUNE 1954 LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST HISTORY OF OPERATIONS AT RECHARGE WELL E 100 80 H U bJ IL ^ UJ rr w 60 —} UJ If) ill < UJ < # Z 40 < o UJ 2 1 Z z 3 o 1- fef < O 20 > UJ 1 Ul 1.0 -20 p " .6 < 2 _ 20 Q. Q. Uj' 1£ 10 JAN 1953 PLATE 8 GROUND SURFACE NX///X\\\\\>'////' MEASUREMENTS QUESTIONABLE FROTH IN WELL AVERAGE DAILY WATER SURFACE ELEVATION IN RECHARGE WELL WELL F GROUND WATER PIEZOMETRIC SURFACE 20' FROM RECHARGE WELL MAXIMUM INJECTION TEST WELL e -AVERAGE DAILY INJECTION RATE CHLORINATION RATE - FEB MAR APR MAY JUNE JULY AUG T SHOCK CHLORINATION SEPT T OCT NOV DEC T 1954 JAN FEB MAR APR MAY JUNE LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST HISTORY OF OPERATIONS AT RECHARGE WELL F PLATE 9 1953 MAR APR MAY JUNE JULY AUG SEPT OCT NOV DEC JAN FEB MAR APR MAY JUNE 1954 LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST HISTORY OF OPERATIONS AT RECHARGE WELL G 100 PLATE 10 LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST HISTORY OF OPERATIONS AT WELL H PLATE 1953 FEB MAR ' APR MAY JUNE JULY AUG SEP OCT NOV DEC JAN FEB MAR APR I MAY I JUNE LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST HISTORY OF OPERATIONS AT RECHARGE WELLS I 8 1954 lA PLATE 12 100 80 60 U UJ UJ UJ U z Z 3 I ° 20 UJ -) LI 1.0 Z (0 .8 p o 6 z < 2 -■= § 2 -20 20 Q. Q. 10 GROUND SURFACE /A\\\>'//'/A\\S\X/ AVERAGE DAILY WATER SURFACE ELEVATION iN RECHARGE WELL INJECTION DISCONTINUED TO REHABILITATE WELL -WELL J-l -GROUND WATER PIEZOMETRIC SURFACE 20' FROM RECHARGE WELL AVERAGE DAILY INJECTION RATE JAN FEB ' MAR ' APR ' MAY ' JUNE 1953 JULY ' AUG ' SEPT LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST HISTORY OF OPERATIONS AT RECHARGE WELL J JUNE 1954 100 80 UJ UJ 60 3 < bJ 40 Z < UJ z z i O b 20 2 Q > UJ _) UJ O u. - (E 1.0 -20 B .6 .4 .2 z° 5 8: z cE UJ Q 20 PLATE 13 GROUND SURFACE //Cs\W'^//X"sSSW////' AVERAGE DAILY WATER SURFACE ELEVATION IN RECHARGE WELL INJECTION DISCONTINUED TO REHABILITATE WELL- GROUND WATER PIEZOMETRIC SURFACE 10' FROM RECHARGE WELL r WELL K-l AVERAGE DAILY INJECTION RATE -INFLOW REDUCED TO REPAIR BACK PRESSURE VALVE. 1 ■ SHOCK CHLORINATION CHLORINATION RATE- JAN 1953 FEB MAR APR MAY JUNE JULY AUG SEPT OCT ' NW ' DEC ' JAN FEB ' MAR APR MAY ' JUNE 1954 LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST HISTORY OF OPERATIONS AT RECHARGE WELL K KJ PLATE WELL D WELL DE WELL E WELL EF WELL F WELL FG WELL G WELL GH WELL H WELL HI WELL Ifl WELL IJ WELL J INJECTION RATES, C.F S. DATE E-12-53 3-10-53 6-15-53 10-11-53 IE-3-53 6-24-54 WELLS 0.50 0.24 0.41 1.05 1.04 39 0.45 0.63 0.37 0.59 075 0-54 1.05 0.39 0.64 H 0.47 0.70 I A O" o' 036 0.37 1.03 I 13 0.51 0.78 055 48 0.74 0.74 O30 TOTAL 75 2.43 484 4.41 4.77 « WELL I LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST GROUND WATER PROFILES PARALLEL TO COAST THROUGH LINE OF RECHARGE SOUTHEAST WELL G-13 WELL G-9 WELL G-5 WELL G-3 WELL G WELL G-2 WELL G-4 WELL G-8 U 111 cr o I in +15 + 10 + 5 z o < > UJ _i -5 -10 ■15 I 1750 1500 1250 1000 750 DISTANCE IN FEET FROM WELL G I I I INJECTION RATES, CF.S. DATE WELLS C D E F G H 1 J K TOTAL 2-12-53 3-10-53 0.75 0.75 6-'5-53 1.05 054 0.36 0.48 2.43 10-11-53 050 1.04 0.63 1.05 0.37 0.51 0.74 4.84 12-3-53 0.240.39 0.37 0.390.47 1.03 0.78 0.74 4.41 6-24-54 0.41 0.45 0.59 0.64 0.70 113 0.55 0.30 4.77 PLATE 15 WELL MB-4 500 250 250 500 LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST GROUND WATER PROFILES NORMAL TO LINE OF RECHARGE THROUGH WELL 750 2000 PLATE 16 PROJECT H£CH»BC£ WELLS PFWJECT OBSERVATION WELLS TTAN BEACH WELLS STANDARD OIL COMPANY WELLS CALIFORNIA WATER SERVICE CO WELLS GENERAL CHEMICAL CO WELLS LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST GROUND WATER CONTOURS FOR FEB 20 1953 PLATE 17 PLATE 18 DEC I JAN 1954 JUNE LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST UNIT RECHARGE WELL INJECTION HEAD FOR WELLS "E" S "S" PLATE 19 I40 WELL F 60 10 I 10 20 MARCH 301 19 29 1954 19 APRIL 1954 MAY LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST UNIT RECHARGE WELL INJECTION HEAD FOR WELLS F, H, I A, AND K 29 JUNE 28 PLATE 20 K> 20 MARCH APRIL UNIT LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST RECHARGE WELL INJECTION HEAD FOR WELLS "j' EFFECT OF SHOCK CHLORINATION AND "D" PLATE 2 u v> Ul C 55 »- UJ 3 2 1- H (/) Z _J UJ O Z < o q: 1- z Q. UJ 8 UJ S ffi o o or >- ^ o < 1- a. 2 _l 0. m -1 m 3 a. >- 2 (0 o CO 2 ^ < (/) o z o CD ^ v> (0 UJ 1- ^ UJ _J UJ e> z (O UJ -1 < < ^ o -I PLATE 22 3 O a z UJ (r lO in 1 10 lO £ tc lO CM < a 1- < (M u> (D 111 -1 » -1 -J » UJ o ■7 ■? o 2 _ -) -1 M n -1 *- Ul <9 z O o S CL 3- z . s 3 (1 a. (M s s .> 3 0. UJ Z «9 » >• >- K O K E «* Ui Ul » > > O o o Ul 4 <> n (X S! Ul a: -1 4 W UJ UJ X a 2 o < a X O lO d li/eiO *Aini8l86IM8NVUl O o CM — 6 6 o o o O o O "^ s - m < I- co o UJ < X o UJ o H co o _J o cc H Z o - I- z u o o UJ a: tt: < CD UJ UJ o a: < X o UJ IL UJ X C3 z o CO UJ < 3 m o o o o o ID o o o UJ UJ t^ _l CO UJ LU < o < > CO (7) z < O PLATE 23 lejDOO 15000 a: a: 4 .'12000 9000 UJ o K O -I I o 6000 3000 i9s; OAN 3 m^LL. WELL 0-t WELL H-n: 1 FEB" MAR I APR I may" JUNE I JULY \ AUG I SEPT I OCT | NOV I DEC" LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST CHLORIDE CONCENTRATION HISTORY OF PROJECT OBSERVATION WELLS INTERNODAL WELLS OF LINE OF RECHARGE 1954 JAN FEB "mar I APR I MAY I JUNE PLATE 24 DISTANCE FROM LINE OF RECHARGE 1953 .JAN FEB MAR APR MAY JUNE T JULY T AUG SEPT OCT NOV DEC ' JAN 1954 LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST CHLORIDE CONCENTRATION HISTORY OF PROJECT OBSERVATION WELLS NORMAL TO CENTER OF LINE OF RECHARGE 1 FEB T MAR APR MAY JUNE PLATE 25 18000 15000 ^12000 a. a. >' t z -19000 UJ 9 o: 3 5 6O00 3000 WELL E-4 JAN 1953 WELL DISTANCE FROM LINE OF RECHARGE C-4 580 E-4 560' 6-4 520' H-4 550' 1-4 520' K-4 470' FEB MAR APR MAY JUNE JULY LOS AUG SEPT T OCT NOV DEC JAN FEB ' MAR ' APR ' MAY ' JUNE 1954 ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST CHLORIDE CONCENTRATION HISTORY OF PROJECT OBSERVATION WELLS APPROXIMATELY 500 FEET LANDWARD OF LINE OF RECHARGE 18000 PLATE 26 a. 0. 15000 ^ >- 1- z < 12000 U) UJ o K O 9000 J 6000 3000 WELL C-8- WELL K-8 WELL C-e PUMPING DEPTH DECREASED FROM 276' TO 263' WELL C-8- WELL G-8- WELL C-8 G-e K-8 DISTANCE FROM LINE OF RECHARGE 940' 1180' 930" "" FEB I MAR ' APR ' MAY ^ JUNE ' JULY ^ AUG ' SEPT ' OCT ' NOV ' DEC LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST CHLORIDE CONCENTRATION HISTORY OF PROJECT OBSERVATION WELLS APPROXIMATELY 1000 FEET LANDWARD OF LINE OF RECHARGE JAN 1953 Tan 1 FEB i MAR I APR I Way ^ June 1954 PLATE 27 WELL B-l- 18000 15000 > I- 12000 VI U, 9000 o _l X 111 _I ■20 -40 -60 i-80 ■100 120 ■ ■140 CONDUCTIVITY (MHOS/CM. AT 25° C) .01 .02 .03 .04 .OS CONDUCTIVITY TRAVERSE STABLE PUMPING CONDITIONS WELL I- 1 o 3- 6-83 A 5- I -93 + 11-30-53 D 4-27-54 LOCATION OF PUMP 3-6-53 a 5-1-53 11-30- 53 a 4-27-54 LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST UNDERRIDING EFFECT OF SALINE WATER 6-8 PLATE 31 200 150 100 UJ 111 u. z g i UJ -J -150 -20( 2000 3000 4000 DISTANCE FROM OCEAN FRONT, FEET LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST IDEALIZED SECTION SHOWING EXTENT OF SALINE INTRUSION PRIOR TO RECHARGE JANUARY, 1953 G-8 200 ISO 100 UJ 15 > UJ _i UJ -50 ■100 -150 -200 PLATE 32 2000 3000 4000 DISTANCE FROM OCEAN FRONT, FEET LOS ANGELES COUNTY FLOOD CONTROL DISTRICT 5000 WEST BASIN BARRIER TEST IDEALIZED SECTION SHOWING DEVELOPMENT OF THE INJECTED FRESH WATER BODY LEGEND P'TTl SALINE WATER ^^ NJECTED FRESH WATER fV^ NATIVE RELATIVELY FRESH WATER 6000 7000 JUNE, 1954 PLATE 33 N PROJECT RECHAnce WELLS PflOJECT OBSERVATION WELLS CITY or MANHATTAN fiCACH we STANOAftO OIL COMPANY WELLS CALIFORNIA WATER SERVICE CO GENERAL CHEMICAL CO WELLS PIPE LINE LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST ISO-CHLORS FOR FEB 12 1953 PLATE 34 N SECUNDO D FEEOEfl LEGEND PROJECT necHARce wells PBOJECT OBSERVATION WELLS CITV or MANHATTAN HEACH WEL STAHOARO OIL COMPANY WELLS CALIFORNIA WATER SERVICE CO WELLS GENERAL CHEMICAL CO. WELLS - PIPE LINE LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRrER TEST ISO-CHLORS FOR JUNE 24 )9S4 LOS ANGELES COUNTY FLOOD CONTROL DISTRICT APPENDIX A GEOLOGIC STUDIES RELATIVE TO INVESTIGATIONAL WORK FOR PREVENTION AND CONTROL OF SEA WATER INTRUSION By EDWARD J. ZIEIBAUER and RICHARD S. DAVIS 5 — 52568 TABLE OF CONTENTS Page Abstract 79 Physiography 79 Soils 79 Description of Major Deposits and Characteristics Affecting the Recharge Test 80 Dune Sand and Coastal Deposits (Recent) 80 Palo Verdes Formation (Uppermost Pleistocene) 80 "Clay Cap" (Upper Pleistocene) 80 "200-Foot Sand" Correlative (Upper Pleistocene) 81 Merged Silverado Zone (Upper and Lower Pleistocene) 81 Upper Brown Phase 81 Lower Gray Phase 81 Silverado Phase 82 Lower Member of the San Pedro Formation (Lower Pleistocene) 82 General Parameters of the Aquifer as Related to Recharge 82 Paleontology 83 Acknowledgments 84 Bibliography 84 PLATES (Tip in between pages 84 and 85) Plate No. Title 1. Geology, Geologic Sections, and Physiographic Provinces 2. Contours on Top of Clay Cap 3. Contours on Bottom of Clay Cap 4. Isopach of Clay Cap 5. Geologic Sections 1-1 and G-G 6. Geologic Sections C-C and K-K 7. Geologic Sections 4-4 and 14-14 8. Contours on Top of Merged Silverado Zone 9. Contours on Bottom of Merged Silverado Zone 10. Isopach of Merged Silverado Zone TABLE Table No. Title 1. Pebble Count of Aquifer Gravels — West Basin Barrier Test 85 PHOTO Photo No. Title 1. Megafossils Noted During Test Drilling 87 (78) GEOLOGIC STUDIES RELATIVE TO INVESTIGATIONAL WORK FOR PREVENTION AND CONTROL OF SEA WATER INTRUSION ABSTRACT Geologic studies were undertaken to determine the nature, extent, thickness, and other pertinent char- acteristics of the affected Merged Silverado aquifer reehars-'od at the TVest Basin Barrier test site at Man- hattan Beach anil Ilerniosa Beach. California. Drilling disclosed at the base of dune sands and coastal deposits, a continuous, relatively impervious body of sediments averaging 20 to 30 feet in thick- ness. This body extended over the entire project area except for a discontinuous band adjacent to the strand, where it appears to have been removed by along-shore current action and transverse channeling. The clay horizon was underlain by the Merged Silverado Zone, the major prolific aquifer which bifurcates inland into several important water- bearing members. This zone, averaging about 110 feet in thickness along the recharge line, was found to be typically divided into two phases — an upper brown phase and a lower gray phase. The upper brown phase consisted primarily of continental and littoral deposits, portions of which were correlative with the "200-foot sand" zone, an important inland aquifer. The lower gray phase appears to have been deposited in a shallow marine environment. Beneath the Merged Silverado aquifer was an ex- tensive thickness of fine-grained sediments which gen- erally became more compact with depth, and consti- tuted the lower boundary of the aquifer. It was found that recharging of the non-homogene- ous aquifer was most effectively and expeditiously ac- complished by the use of gravel-packed wells. Strip- ping of the capping sediments may have retarded but did not prevent pressurization of the aquifer by re- charging. Preservation of the cap sediments to permit continued and effective pressurization required both grouting throughout the cap and care to prevent overdevelopment of injection wells. Stripping of the cap sediments along the strand and ancient channel- ing extending inland transverse to the present shore line, coupled witii a steepened hydraulic gradient, probably accelerated sea water intrusion at the nearby, now abandoned. City of JIanhattan Beach well field. The coincidence in the marked thinning of Merged Silverado sediments at the coast, the re- markable linearity and parallelism of topographic features with the pi-esent shore line, and differentials in elevations of stratigraphic surfaces zoned by for- aminiferal assemblages, suggests the possibility that deposition of sediments was controlled to some degree by sharp tiexuring or faulting parallel to the coast. This flexuring appears to have established a partial barrier to sea water intrusion, as evidenced by the steep piezometric gradient indicated by oceanward observation water surface levels noted prior to re- chai-ging. PHYSIOGRAPHY The project area lies within the limits of three of four minor physiographic provinces, I, II, III, and IV, outlined by Metzner ^ and delineated on plate (1). The project recharge line lies within province II. In brief: Province I, extends from the Pacific Ocean inland to the first valley paralleling the coast, a distance of from 1,500 to 2,000 feet. Rather steep, recent sand dune escarpments lie along the major portions of both boundaries. Province II, next inland, includes a series of sand dune ridges and valleys paralleling the coast line. At Manhattan Beach the province swings sharply coast- ward and then southerly flattens into a terrace. The remarkable parallelism and linearity of the topo- graphic features of provinces I and II, and of the shore line, suggests the possibility of fault control. Province III, further inland, includes a much wider band of sand dune hills and depressions having a general trend nearly perpendicular to that of province II and the coast line. Province IV, extends from the eastern boundary of province III to the Newport-Beverly Hills Uplift (luglewood Fault Zone) and includes the main por- tion of the "West Coast ground water basin. SOILS Soils within the area investigated are, as mapped by the V. S. Department of Agriculture Bureau of Soils,- of three principal types: (1) Coastal beach and dune sajid, (2) Oakley fine sand, and (3) Ramona sandy loam. Coastal beach and dune sands occur directly along the shore and are composed of pervious yellow-brown to buff", fine to very coarse sand, quite clean and poorly consolidated. The Oakley fine sand series is brown, with varia- tions to grayish brown and light brown to buff, and with the lower part of the section frequently lighter ' See Bibliography. = See Bibliography. ( 79) 80 SEA WATER INTRUSION IN CALIFORNIA in color. The soils generally contain considerable fine material, giving them a somewhat loamy appearance, and often contain irregular zones of clay-binding material which makes them relatively impervious. The Ramona series, presumably derived from altered old unconsolidated water-laid deposits, is brown in color with slightly reddish brown to grayish brown variations. It is underlain by heavier, more compact, brown, reddish brown, light brown or red soils. These sandy loams exist only in the most land- ward extension of the area and there seems to be some reason to believe that locally the series may be dune material rather than water-laid and should actually, despite fineness and compactness, be classified as the Oakley wind-blown sand. DESCRIPTION OF MAJOR DEPOSITS AND CHARACTERISTICS AFFECTING THE RECHARGE TEST Dune Sand and Coasfal Deposifs (Recent) The main portion of the surface of the area is oc- cupied by the El Segundo Sand Hills. These hills are largely dune sand and are composed of light yellowish brown to dark reddish brown, fine to coarse sand. (Plate 1). The sands show some iron staining and clay binding, and are locally silty, compacted and relatively impervious. The sorting is fair and the shape ranges from angular to subangular. These coarse sands, primarily of granitic origin, contain an appreciable amount of magnetite, pyroxenes, amphi- boles, and micas with lesser amounts of zircon, topaz, garnet, epidote, etc. Near shore, the older dune sands are directly overlain by active dune material which is loose and highly permeable. In general, three significant horizons were noted within the sand dune deposits above the "clay cap" — the relatively impervious stratum confining the underlying major aquifer: a localized surficial "iron- bound" sand horizon, an intermediate horizon of relatively clean dune sands with occasional gravels in the basal portion, and a lower horizon of fine sands and silts with sandy stringers which constituted a zone of transition to the "clay cap." Two-inch test wells adjacent to the injection wells were bottomed within the lower horizon of the sand dunes but above the "clay cap," to determine leak- age from the aqiiifer. High water surface levels were observed in the test wells. Subsequently, shallower test holes were drilled near the injection wells to cheek the levels observed in the original two-inch wells. These shallower wells confirmed the belief that recorded high water surface levels in the original test wells reflected semieonfined or partial pressure levels within the zone of transition to the "clay cap" rather than free water levels. Hence, leakage from the aquifer to sands lying above the clay cap was probably limited to that transmitted by the sandy stringers within the zone of transition to the "clay cap." Inasmuch as the stringers were of minor thick- ness, flow through the stringers was doubtless equally limited, and hence the leakage above the "clay cap" during the test is considered to be of minor signifi- cance. The sediments along the beach and underlying the dune sand are recent marine coastal deposits com- posed of sands and gravels, with included cobbles from three to eight inches in diameter. The two types of deposits are normally not differentiated. Palos Verdes Formafion (Uppermosf Pleistocene) Deposits constituting the Uppermost Pleistocene Palos Verdes formation are marine sands and gravels underlying the dune sands. They are very similar to the coastal deposits, except that they occur somewhat farther inland and include calcitic fragments not found in the coastal deposits. Diagnostic marine mega- fossils are often noted. In portions of the area the formation is absent, probably due to stripping by marine currents and/or fluviatile activity. To the east of the El Segundo Sand Hills and paral- lel to the coast, a very small area of Upper Pleistocene sediments is exposed within the area investigated. "Clay Cap" (Upper Pleistocene) Directly underneath the dune sand and coastal deposits and lying at or near sea level is a continuous horizon of relatively impervious deposits referred to, for simplicity in this investigation, as the "clay cap." This cap was not a true compact clay, but composed of varying colors of silts, silty fine sand stringers, sandy clays, and clays normally yellowish brown but occasionally gray, green, or mottled. In cases where sediments comprising the cap are gray in color, the upper brown member of the Merged Silverado Zone may be non-existent. In certain cases in the easterly portion of the area, thin discontinuous sand and gravel lenses are included as a part of this cap. These deposits, typical of many coastal reaches of the State, obviously are not as impermeable as a true clay body and are subject to some degree of erosion if water is permitted to move at excessive velocities along or through the cap. Hence, the annular space between casings and side walls of injection wells must be properly sealed throughout the cap to prevent rupturing induced by sudden changes in injection rates, with consequent failure of repressurizing oper- ations. The cap failures experienced at injection wells C, G, and I were probably due to the creation of voids produced by over-development and excessive leakage past the clay cap at the well casing. In contrast, the continuous successful operation of wells E and I-A, which were gravel-packed and grouted, would indi- cate the benefits of this type of well construction. SEA WATER INTRUSION IN CALIFORNIA 81 Test well data within the subject area revealed the upper surface of the "clay cap" to be unusually flat aud to vary from 10 feet above to 10 feet below sea level. (See plate 2). Existing: well data indicated larger irregularities inland due, at least in jjart, to differences in logging by various drillers. The relative flatness of the "clay cap" is indicative of probable planation aiul deposition of an older, more irregular surface. Contours on the bottom of the "clay cap" appear to reflect, to a degree, transverse drainage channels. (See plate 3.) Tliickness of the "clay cap" along the recharge line, as delineated by isopachs on plate 4, averages from 20 to 30 feet and varies from zero at the strand to 48 feet on the inland side of the re- charge line. This is also shown on geologic sections, plates 5. 6, and 7. The presence of this "clay cap" over the entire area definitely identified the aquifer as a pressure or confined aquifer. This was of considerable importance in the establishment of a pressure mound or ridge along the recharge line. On the basis of data available prior to the test, an investigation by another agency indicated the aquifer as being exposed to saline water at some distance seaward of the coast. Wells drilled along the strand, however, showed the absence of clays as well as at one point some 800 feet inland from the strand, at test well K-9. This apparently resulted from stripping and channeling by geologically recent erosioual activity. Local stripping of the "claj- cap" along the shore margin is of significance in that it reduced the dis- tance of travel of ocean waters aud hence hastened saline encroachment at these points. The stripping may have retarded but did not prevent pressuriza- tion of the aquifer during recharge, since a continuous effective "clay cap" exists along the recharge line and seaward of it for some distance. "200-foot Sand" Correlafive (Upper Pleisfocene) Beneath the "clay cap" and basinward of a line about one mile east of the coast is a zone of inter- bedded sands and clays overhang the San Pedro for- mation. These sediments are normally barren of mega- fossils and, on the basis of lithologic studies to date, are distinguished only in a general way from the San Pedro formation. This zone appears to be, at least in part, correlative with the "200-foot sand," an im- portant inland aquifer mapped by the United States Geological Survey in this general area in 1948.* Its existence is inferred due to the absence of fossils, which usuallj- occur in the lower formation, and the presence of a substantially thick separating clay member. Where this clay member exists (see logs of wells JI-14, H-16, and K-16), it often separates the upper brown phase of the ^Merged Silverado Zone ' See Bibliography. from the lower gray phase, both of which are de- scribed below. This appears to indicate, as logged by the State Division of Water Resources, that the upper brown phase constitutes the seaward correlative of the "200-foot sand" zone, also previously mapped by that agency.* Continuity with the "200-foot sand" zone is significant in that barrier operations not only check sea water intrusion into the zone, but also permit its recharging with fresh water. Merged Silverado Zone (Upper and Lower Pleisfocene) The term "Merged Silverado Zone," proposed in a prior report on the West Coast Basin,'' was expanded from the name "Silverado Zone" used by the U.S.G.S. in 1948.* This zone, in the subject area, spans both the Upper and Lower Pleistocene and can be divided lithologieally into several components: an upper brown phase, a lower gray phase, and the Silverado — a generally very permeable portion of the San Pedro formation. Within the subject area the zone varies in thickness from 29 feet at Manhattan Beach City well No. 4 (F.C. 701-C) to nearly 155 feet at Project Well No. G-8 ; however, it thickens appreciably basinward. UPPER BROWN PHASE Although possibly distinct from the underlying Sil- verado, the upper brown phase is generally in hy- draulic continuit}'' with, and hence is included as a portion of, the ilerged Zone. Although this phase is quite extensive, local gaps are noted. The upper por- tion of the phase normally consists of yellowish brown sands and silts, and the lower portion of fine sands and gravel stringers with occasional clay bands. Sand fractions are angular to subaugular while the gravel in the lower portion contains fragments that are somewhat more rounded. It is arkosic, having a feldspar to quartz ratio of about 3 to 2. The heavy mineral content varies from 1% to about 15% and consists largely of magnetite and ilmenite, pyroxenes, amphiboles, micas, epidote, topaz, garnet, and minor amounts of other minerals. Study of rock fragments (See table 1) indicates only minor differences between this phase and an underlying lower gray phase. The upper brown phase is much more prominent in the seaward portion of the area and does not appear to exist in certain of the more inland wells. (See logs of wells K-16 andM-18). LOWER GRAY PHASE A gradation is noted from the upper brown ])hase to the underlying lower gray phase. The upper por- tion of the lower gray phase normally is rather gi-een- ish and becomes progressively gray and then bluish- gray with depth. The gravel horizon constituting the basal portion of the upper brown phase seems to ' See BlbUography. 82 SEA WATER INTRUSION IN CALIFORNIA merge with the upper portion of the lower gray phase, which contains the gravel horizon of that phase. The lower gray phase becomes finer and more compact with depth. Lithologieally, there is very little appar- ent difference between the two zones despite the great difference in color. This color difference seems prima- rily to be the result of the weathering of iron-rich biotite. In the upper phase, this produces an iron staining which imparts a distinct yellowish brown color. In the lower phase weathering is not noticeable, and the black coloration of the biotite lends a gray appearance to the sediment. A more detailed petro- graphic analysis may establish diagnostic criteria dif- ferentiating the upper and lower phases. Below the gray gravel horizon, as previously stated, the sediments become progressively finer and darker in color. They consist of fine to very fine silty sands and clay bands. Occasionally lenses of medium to fine sand are noted. Examination reveals the feldspar to quartz relation to be about 2 to 1 and the heavy mineral content, which is composed largely of biotite, to vary from about 2% to about 8%. The larger fragments are largely granitoid type rocks, some metamorphics, such as quartzite and gneiss, some basaltic or dark fine-grained volcanics, pegmatites or vein type quartz, and a minor amount of sedimentary rocks. SILVERADO PHASE That portion of the San Pedro formation in the West Coast Basin which is most permeable and com- posed largely of sand and gravel is know as the Sil- verado water-bearing zone. This zone is the most im- portant and most prolific aquifer in the basin. In the subject area, it consists largely of sand with scattered gravel and is combined or merged with the brown and gray phases previously described. In some instances, it is gray in its entirety. Another aquifer of Lower Pleistocene age, the ' ' 400- foot gravel" zone, which is important in the basin interior, does not exist recognizably in the investiga- tion area. tower Member of fhe San Pedro Formation (Lower Pleisfocene) In this area the lower member of the San Pedro Formation is composed of bluish gray compact very fine sands, silts and clays. The relatively tight sedi- ments are important aquieludes which retain waters in the overlying permeable members of the Merged Silverado Zone and constitute the lower boundary of the aquifer. Recognition of this member is based on the presence of megafossils of a cold water type and, where they occur, foraminifera. Obviously, the finer grained sediments underlying the merged zone are not impermeable but relatively so ; hence, salinity intru- sion within these sediments is of relatively minor con- cern. Transmission of pressure effects from the Merged Silverado Zone to the underlying sediments eventually occurs within those portions having at least some degree of hydraulic continuity with the Merged Zone. GENERAL PARAMETERS OF THE AQUIFER AS RELATED TO RECHARGE Geologic sections showing the extent, thickness, and character of the zones described may be noted on plates 5, 6, and 7. The Merged Silverado Zone, for the purpose of this report, has been chosen as that mate- rial below the ' ' clay cap ' ' and above the fine, compact, blue-gray portion of the lower San Pedro formation except where the correlative of the "200-foot sand" zone is distinguishable. In some cases where minor clays or claj'ey or silty strata are interbedded with the coarser material, particularly near the bottom, they have been included in this zone. Although the thickness of the Merged Silverado varies greatly, both seaward and landward, along the recharge line, the zone averages about 110 feet in thick- ness and the bottom elevation averages about 130 feet below sea level. Along geologic section lines C-C, G-G, and K-K extending in a direction transverse to the recharge line, the depth is somewhat variable, being in general deeper inland and toward the south. The thickness along section C-C is about 90 feet with the average elevation of the bottom being about 130 feet below sea level. The thickness along G-G averages about 100 feet and the bottom elevation is about 140 feet below sea level. The average thickness along K-K is about 120 feet and the elevation of the bottom is about 160 feet below sea level. Extending eastward from the ocean, the IMerged Silverado Zone thickens irregularly. Contours drawn on tlie top (Plate 8) and bottom (Plate 9) of tlie aqui- fer indicate several large irregularities in the vicinity of the test site and reveal the bottom to be somewhat more variable than the top. These surfaces are signifi- cant in that they indicate channeling transverse to the present shore line. There appears to be, in addi- tion, a rather strong suggestion of parallelism with the present shore line which may be a reflection of deposition controlled to a degree by sharp flexuring or faulting. Such a zone of sharp flexuring or fatilting further evidenced by a discontinuity of stratigraphic surfaces, may act as a partial barrier to sea water intrusion, as indicated by observation well water sur- face levels noted prior to recharging. Tliis pattern of thinning of aquifer sediments with approach to a fault-controlled coastline is evident along otlier coastal reaches of the State and is of significance in that, where applicable, geologic delineation of the zone of SEA WATER INTRUSION IN CALIFORNIA 83 thinning can define the most hydraulically effective Test well drilling delineated a channel transverse ami euononucal route for injection wells, jirovided to the present shore line between test wells Cr and I, rifrlit of way acnuisitiou is economically feasible. Iso- extending from the proposed recharge inland line to pachs on the Merged Silverado, (see plate 10) based the City of Manhattan Beach well fields. This feature, iu part on old water well logs, indicate differences in coupled with the absence of the "clay cap" along localized areas of special interest : shore, doubtless permitted more direct contact of the , , , ^ ^i- • ^ 1.1 1. 1 1 rr-,. e Merged zone with sea water at or near the .strand. ill An abrupt thinning, at the abandoned City of o i- • . • i? .i -it.,, ,, , AT 1 ^^ 15 1 11 /? 1 I . o.i C-, *■ 1 Saline intrusion of the original Manhattan Beach Manhattan Beach well field at Mh Street and n /; 1 1 i ii , * i i, ^i i i- „ , , T, , , -^i •■ ..1 • , well field probablv was accelerated bv the channeling Sepulveda Boulevard, with a minimum thick- , . i i i v i- / . , . ..i . , «„ 1. ' , -.r 1 .. -r, 1. 11 aii?^ LEGEND Cos'ng Pef(0fOl>on^3G'0vel EFme To Very Fine Silty Sand @Gniv(iaiS(n]BSil1or Cloy DSofid □ailorCkiy Phases QRilaiixly Impcmoui S'loto SeoM of Merged StNeiodo Zone HORIZONTAL SCALE 200 400 00( IN FEET ELEVATIONS IN FEET USGS OflTUM LOCATION OF SECTIONS SEE PLATE I GEOLOGIC SECTION G-G LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST GEOLOGIC SECTIONS 1-1 AND G-G PLATE 8 MANHATTAN BEACH LEGEND UNES OF EQUAL ELEVATION ON THE TOP OF MERGED SILVERADO ZONE. INFERRED LINES OF EQUAL ELEVATION ON THE TOP OF MERGED SILVERADO ZONE HERMOSA BEACH ^°^^-=^ ,^^°° ^000 zooo 1000 o LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST CONTOURS ON TOP OF MERGED SILVERADO ZONE PLATE 9 PLATE 10 SEA WATER INTRUSION IN CALIFORNIA 85 TABLE I PEBBLE COUNT OF AQUIFER GRAVELS-WEST BASIN BARRIER TEST Well Number. Upper Brown Phase Elevation. Rock typo* Granitoids Volcanics Metamorphics. Sediments Total %. K-8 —73 63.7 6.0 39.8 0.0 99.5 A-14 —62 61.6 6.8 29.1 2.6 100.1 K-12 —79 50.0 8.4 41.6 0.0 100.0 M-14 —53 55. 3 12.0 29.9 3.0 100.2 K-16 —75 76.8 2.5 30.6 0.0 99.9 Acoum. Avg. 56.6 8.3 34.0 1.1 Lower Gray Phase Well Number C-12 —73 67.2 1.4 31.5 0.0 A-14 —84 56.2 5.5 38.4 0.0 K-12 —94 74.0 4.6 20.5 0.9 M-14 —83 74.2 5.7 19.3 0.6 H-16 —159 26.6 14.0 41.1 18.3 M-18 — «8 38.1 10.5 51.4 0.0 Accum. Ate. Rock t>-pe» 64.3 6.4 27.6 2.4 Total % - 100.1 100.1 100.0 99.8 100.0 100.0 100.1 I^wer Gray Phase — Continued C-8 —76 65.0 1.1 31.9 2.1 C-12 —89 77.2 1.6 21.3 0.0 H-16 —206 83.5 3.2 8.9 4.5 M-18 —141 80.8 6.5 11.5 1.3 Rock tj-pe* Included Total % 100.1 100. 1 100.1 100.1 ' Grinltoids Include monzonltc. granite, granodiorlte, diorite, gabbro, and rein and pegmatiUc quartz. Volcanics Include basalt, rhyolite and felslte. Metamorphics include gneiss, schist, and quartzite. Sediments include shale, sandstone, and chert. 6—52568 86 SEA WATER INTRUSION IN CALIFORNIA MEGAFOSSILS NOTED DURING TEST DRILLING (Age occurrence in California only) (See Photo 1 ) Species Range 1. Nueulana taphria (Dall) Miocene — Recent 2. Aeila (Truncacila) castrensis (Hinds) Miocene — Recent 3. Peeten (Patinopecten) caurinus Gould Pliocene — Recent 4. Peeten (Peeten) caurinus Gould Upper Miocene — Recent 5. Trausenuella tantilla (Gould) Pleistocene — Recent 6. Lueina (Myrtea) acutilineata Conrad Miocene — Recent 7. Solen sicarius Gould Miocene — Recent 8. Mitrella carinata (Hinds) var. gausapata (Gould) Pliocene — Pleistocene 9. Mitrella tuberosa (Carpenter) Upper Miocene — Recent 10. Retusa (Acteocina) culcitella (Gould) Pliocene — Recent 11. Olivella pedroana (Conway) Miocene— Recent 12. Olivella biplicata (Sowerby) Pliocene — Recent 13. Nassarius (Schizopj'ga) perpinguis (Hinds) Miocene — Recent 14. Nassarius (Schizopyga) mendicus (Gould) Miocene — Recent 15. Turritella cooperi Carpenter Upper Miocene — Recent 16. Bittium (Semibittium) rugatum Carpenter Pliocene? — Recent 17. Epitonium cataliuae Dall Pliocene ? — Recent X2 PHOTO No. 1 X2 -«!»-■ X2 i X2 i X2 f X2 X2 ^ X2 10 • X2 14 II X2 I X2 16 13 k ■■ ■ \ 1 • X2 17 koo AnotLca couNTT rcoos COHTBOi. Oi«Ta-av 176 K Ja,Uk WEST BASIN BARRIER TEST AAegofoss!ts noted durlnrj lest drltfing (Age occurrence in California only) LOS ANGELES COUNTY FLOOD CONTROL DISTRICT TESTING DIVISION APPENDIX B LABORATORY STUDIES AND RESEARCH RELATIVE TO INVESTIGATIONAL WORK FOR THE PREVENTION AND CONTROL OF SEA WATER INTRUSION December 20, 1954 By CHARLES GREEN Submitted by: F. O. FRICKER, Division Head Recommended by: Approved by: PAUL BAUMANN H. E. HEDGES Assistant Chief Engineer Chief Engineer TABLE OF CONTENTS Page A. Summary 91 B. Introduction 91 C. Scope of Report 91 D. Description of Tests and Methods 92 E. Presentation of Data and Results 93 F. Discussion of Results 95 1. Preliminary Discussion of Geochemical Changes 95 2. Geochemical Changes in Groundwater and Soil Resulting from Sea- Water Intrusion 96 3. Geochemical Changes in Groundwater and Soils Resulting from Re- charge 98 4. Control of Chlorine Dosage 100 5. Relationship of Permeability to Cation Exchange 100 6. Presence of Deleterious Substances and Their Effect on Recharge 101 G. Conclusions 102 H. References 102 I. Tables 103 J. Charts 103 (90) LABORATORY STUDIES AND RESEARCH RELATIVE TO INVESTIGATIONAL WORK FOR THE PREVENTION AND CONTROL OF SEA WATER INTRUSION A. SUMMARY This Appendix is a report on tlie laboratory studies and researeli eoiidueted by the Testing Division of the Los Angeles County Flood Control District as a part of its responsibility in formulating plans and design criteria for the correction or prevention of damage to underground water by sea-water intrusion in the West Coast Basin of Los Angeles County. The tests procedures are described, results indicated and conclusions presented. The summary of conclu- sions being: 1. The tests establish the fact that the source of saline pollution of the groundwater at the West Basin Barrier Test site is sea-water. 2. The saline water has converted the soil in con- tact with it to a high sodium soil through cation exchange. The degree of conversion is in propor- tion to the concentration of the saline water and therefore decreases with increased distance from the shore line. 3. As a consequence of recharge, the sodium sat- urated soil has given up its exchangeable sodium to saline water as its dilution with recharge wa- ter increases. As the recharge water displaces the saline water completely this exchange continues as long as enough exchangeable sodium is avail- able in the soil. 4. The changes in permeability resulting from the above cation exchange processes in the soil are obscured by other factors difficult to evaluate. 5. The optimum chlorine dosage was best deter- mined by the maintenance of maximum accept- ance rather than bj' chlorine demand evaluations or bacteriological analyses. 6. The effect of suspended solids in the recharge water on acceptance rate could not be evaluated. B. INTRODUCTION The purpose of this report is to present information obtained from the laboratory research studies of the general program conducted by the Los Angeles County Flood Control District. This program was undertaken as a result of an agreement, entered into by the State Board of Water Resources and the Los Angeles County Flood Control District, to perform investigational work for the prevention and control of sea-water intrusion. C. SCOPE OF REPORT The laboratory research studies involved the con- duction of experimental tests and investigation of phenomena resulting from prototype application of injection wells toward creating a pressure ridge in confined aquifers to prevent and control this sea-water intrusion. The data obtained has been tabulated, as- sembled and evaluated where feasible for the items that were investigated. A brief outline of these in- vestigations follows: 1. Evaluation of Water Quality a) Pre-recharge analyses b) Quality changes resulting from recharge 2. Origin of Various Tj'pes of Groundwater Present in the Area a) Criteria used for determining pollution sources b) Special tests 3. Effects of Deleterious Constituents in Ground and Recharge Waters a) Iron b) Calcium carbonate (sludge) c) Bacteria d) Suspended solids Under this heading is included deleterious microbiological and chemical constitutents na- tively present in the merged aquifer and also in the recharge water. 4. Ion Exchange Ion (base) exchange reactions will be covered in the "Evaluation of Water Quality" section of this report and in other sections where the subject is pertinent. 5. Laboratory Soil Permeability Tests This consists of a compilation of laboratory permeability and supplementarj' test data. Eval- uation is presented where permeability is related to cation exchange. 6. Control of Chlorine Dosage An evaluation of methods for control of chlo- rine dosages is presented. This evaluation is based on an investigation of the effect of suspended (01) 92 SEA WATER INTRUSION IN CALIFORNIA solids and micro-organisms in water, on the per- colation rate through Ottawa Sand. D. DESCRIPTION OF TESTS AND METHODS 1. Water Quality Analyses a) Routine Analyses Samples received from the West Coast Basin Barrier Test were subjected, in the greater majority of the cases, to complete analyses which included the determination of the following items : (1) Calcium (2) Magnesium (3) Sodium with Potassium (by calculation) (4) Carbonate (5) Bicarbonate (6) Chloride (7) Sulfate (8) Ammonium (9) Nitrates (10) Total dissolved solids by evaporation at 103°C (11) Total dissolved solids by summation of items (1) to (9) inclusive (12) Hardness (13) pH (14) Electrical conductivity — micromhoes per em at 25° C. (15) Description of sample: turbidity, sedi- ment, color and odor. Occasionally where a complete analysis was thought unnecessary a partial analy- sis was made in which only items 4, 5, 6, 7, 10, 12, 13, 14 and 15 were determined. b) Special Analyses Certain constituents not included in the above list of routine determinations were ana- lized in specific instances. (1) Bromide, borate and iodide ions were de- termined in samples from Wells C-12, Manhattan Beach #4 and K-12 vrith the view to determining polluting source, i.e., connate oil brine versus sea-water. (2) Iron was determined in some water sam- ples in order to evaluate effects of corro- sion. (3) Dissolved oxygen determinations were made on samples of recharge water and pumped native water in order to evaluate the degree of air entrainment. These tests were conducted at Manhattan Beach. (4) Other tests on the recharge water, per- formed at Manhattan Beach, were pH, chlorine residual and chlorine demand. Wherever possible the "project person- nel" were instructed in the methods of performing the above tests, and the data obtained by them are not included in this report though reference may be made to these data. (5) Analyses were made of residues and de- posits taken from valves, meters, chlorin- ator equipment and other locations. 2. Method of Reporting Analytical Results The "water quality reports" on the West Coast Basin Bai-rier Test water samples, sub- mitted periodically, presented the above itemized constituents, where applicable, in terms of parts per million by weight (ppm) and as percent equivalents per million (%epm). To complete the information in terms of equivalents per million (epm), the sum of the major ions as equivalents per million has been recentl}^ in- cluded in the water quality reports. The purpose of reporting percent epm is that it reduces the analyses to a form which may be plotted on a tri-linear chart. The value of plot- ting the data on a tri-linear chart graphically, to determine the source of pollution or dilution, is described in more detail later. Percent epm further provides a means for quantitatively measuring cation exchange in water. 3. Constant Head Permeameters Constant head permeameters were iised to de- termine permeabilities of soil samples obtained during drilling of the West Coast Basin Barrier Test wells. Initially the permeameters were machined from steel and were either black oxide finished or chrome plated to inhibit corrosion. This pro- tection was soon found to be inadequate. The products of corrosion formed in the presence of the highly saline water in the samples tended to clog the sample and give decreasing values of permeability. In addition, it was found that the porous plate supporting the sample eventually became clogged either with the products of cor- rosion or suspended material in the water. The permeability values reported are the first values obtained in the test after correction was made for porosity of the porous plates. In order to correct the deficiencies apparent in the permeability apparatus as originally de- signed, the permeameters were redesigned along the lines indicated in Figure 1. The features which have been incorporated in the new design are: a) AU plastic construction SEA WATER INTRUSION IN CALIFORNIA 93 b) Piezometer connections at one-inch intervals along the length of the permeameter cjlinder c) Means for evacuating the sample to prevent decreased permeabilities due to entrained air Sea-water, filtered through diatomaceous earth under vacuum was used in the tests. The water was filtered to minimize the effect of suspended solids, micro-organisms and air in the water. Sea-water was used to duplicate the conditions existing in the aquifer prior to recharge. In conjunction with permeabilit.y tests, ion ex- change plienomena were quantitatively deter- mined by chemical analj'ses of the calcium, magnesium and chloride ions in the influent and effluent. Two fluids were used in these determi- nations: filtered sea-water followed by filtered Metropolitan Water District water, both ob- tained at Manhattan Beach. Changes in perme- ability were noted as sea-water was replaced by Jl.W.b. water. 4. Falling Head Permeameters Two percolation tubes, each consisting of a 1|" I.D. Incite cylinder 2 feet long, closed at the bottom with a fine screen and approximately half filled with a weighed amount of screened, sterile Ottawa Sand, were set up in the chlorine storage shed at Manhattan Beach. Water taken from the line leading to the chlorinator (pre-chloriuator water) passed con- tinuously through one cylinder; and, water taken from the line leading away from the chlorinator (post-chlorinator water) passed through the second cylinder. The cylinders were supported on a board with a third cylinder placed between them similarly loaded but closed at the bottom and filled with water. This cylin- der was included as part of the apparatus ar- rangement to furnish a visual comparison with the other cylinders. Permeabilities were determined periodically by considering the percolation tubes as falling head permeameters. Although this assumption is not entirely valid, the results obtained are suffi- ciently close to the true values to be comparable with each other. E. PRESENTATION OF DATA AND RESULTS 1. Water Qualitj' Evaluation a) Pre-recharge conditions The selection of analyses to represent pre- recharge conditions for each observation and recharge well was determined by the follow- ing considerations: (1) Proximity of sample date, to date of re- charge in the nearest recharge well so tlial the sample is not influenced b.y re- el large. (2) "Typicalness" of sample as determined by comparison with analyses of other samples from the same well taken at ap- proximate]}' the same time. (3) Proximity of the chloride concentration of the sample with the chloride concen- tration selected as representative of pre- recharge concentrations. (4) Availability of analyses that would fill the above conditions. In case one was not available a United States Geological Sur- vey analysis was taken to represent the pre-recharge state. Pre-recharge analyses are tabulated in Table I. The calculated and analysed percent epm for each well are given in order to determine the loss or gain of cations due to cation exchange. This in- information is li.sted in the three columns under cation exchange. Negative values indicate migration of cations from water to soil* and positive values the reverse. Percent dilution in the fifth column is de- termined from the chloride concentra- tions of native and sea-water as shown in Table I. Percent dilution in turn was used to calculate cation percent epm. The native water analyses given in Table I is the mean of the analyses of samples taken from wells G-8, H-16, K-12, Test Hole No. 2 and General Chemical Well No. 4. In order to determine the degree of saturation of the soil with sea-water at each well, the Sodium Absorption Ratio (SAR) is first calculated from the rela- tion: SAR = Na \/iCa + Mg) /2 where Na, Ca and Mg are the analysed epm of sodium, calcium and magnesium in the water in contact with the soil. The above relationship combines the effect of total cation concentration and the ratio of sodium to calcium pli:s magnesium, each of which influence the equilibrium reached between the soil and the soluble cations in the water. Since a linear rela- tionship exists between SAR and the ratio of exchangeable sodium to ex- changeable calcium plus magnesium pres- • The term "soil" as used In this report refers to the unconsoli- dated deposits present In the merged Silverado Zone which have cation exchange properties. ' Sec Bibliography. 94 SEA WATER INTRUSION IN CALIFORNIA eiit iu the soil/ a similar relationship ex- ists between SAR and the degree to which the soil is saturated with exchange- able sodium. Therefore one hundred percent satura- tion of the soil is arbitrarily taken as the condition where the exchange reaction has ceased and the intruded sea-water is unaltered; zero percent saturation is where only native water exists. Be- tween these two limits the percent satu- ration existing at each well is interpo- lated using the SAR values for each weU. The analysed percent epm data in Ta- ble I axe plotted on a tri-linear chart (Fig- ure 2). Zones delineated in the tri-linear chart, representing different ranges of water quality distinct from concentra- tion are then geographically represented on the map of the West Coast Basin Bar- rier Test (Figure 3). The chloride con- centration, tabulated both in ppm and as a decimal ratio to sea-water is indicated on this map as the decimal ratio. The groundwater-sea-water chloride ratio re- ferred to, has been selected as a para- meter rather than the chloride concentra- tion in ppm, since it is then possible to correlate and amplify concentration data with a similar ratio of electrical conduc- tivity. b) Effect of Recharge WeU GH and the G line of wells, G-2, G-4 and G-8 have been selected to show change of water quality and concentration with re- charge. The data tabulated in Table 2-A for WeU GH are plotted in Figure 4. The re- charge rates in adjacent wells G and H are plotted to show their influence on concentra- tion, degree of dilution and water quality. The difference between calculated and ana- lysed percent epm of the cations are plotted to show cation exchange effects. The tri-linear plotting of analysed percent epm is shown in Figure 5. This plot shows the water quaUty "path" taken by the water in weU GH as re- charge progressed. The data tabulated in tables 2B, 2C and 2D for weUs G-2, G-4 and G-8 are plotted in Figures 6 and 7. Figure 6 shows the change in the concentration (as epm) of the anions and cations in each well with increased dilution. Each triangle can be considered as an integration of several bar- graphs with the anion and cations bars sepa- rated. Figure 7 shows the variation in cation exchange, as percent epm, with dilution for each well. Dilution is represented in absolute terms of equivalent per million so that the products of ordinate and abscissa equals ex- change as equivalents per million. Tri-linear chart (Figure 8) is drawn to show the sources of dilution in wells K-8 and K-12. c) Results of Special Analyses (1) Source of Pollution Table 3 presents the results of special analyses conducted to determine polluting source in Wells C-12, MB-4 and K-12. For comparison, analyses of sea-water and connate waters are included.^ (2) Dissolved Oxygen in Groundwater Table 4 lists the results of dissolved oxygen determinations on pump dis- charges. (3) On four occasions recharge water was analysed in the field for pH, dissolved oxA'gen and residual chlorine. The results are tabulated in Table 5. (4) Iron in Ground and Recharge Waters A series of tests involving the detection of iron in ground and recharge waters was conducted during the course of the West Coast Basin Barrier Test. One test was to determine the source of iron present in Well G following the cave-in. Table 6 presents the data ob- tained during this test. A second test was conducted when the following reaction was noted during the course of a field chlorine demand test. A 160 ppm chlorine solution precipitated an iron hydroxide floe when added to clear thiefed samples. The samples were filtered and sent to the laboratory witli samples of unfiltered (raw) water from the same sources. The results of the iron determinations on the samples are shown in Table 7. Samples of M.W^.D. water were taken from the "vault" at Manhattan Bcacli Boulevard and Redondo Boulevard and at the Field House in order to determine the iron picked up in the feeder line. The results of the test are presented in Table 8. • See Bibliography. ' See Bibliography. SEA WATER INTRUSION IN CALIFORNIA 95 (5) Analyses of Deposits and Sludges The laboratory has received samples of deposits and slndtres found on equipment and in the feed lines. The results of the analyses of these are shown in Tables 9, 10 and 11. Table 12 contains tlie analyses of the water found in the ehlorinator bell- jar from which the sludp:e reported in Table 11 was taken. For comparison an analyses of the recharge water is in- eluded. 2. Constant Head Permeameter a) The permeability results of soil samples de- determined in the steel permeameters are reported in Table 13. Those determined in the lucite permeameters are reported in Table 14. The tests performed in the lucite permeame- ters were conducted using sea-water. M.AV.D. water was used in some of the tests after the application of sea-water. The permeability values listed in these tables are based on a hydraulic gradient of unity. b) Influent and effluent fluids were chemically analysed in some of the tests reported in Table 14 to determine relationship of permea- bility to ion exchange. The complete analytical and permeability data for the soil sample from Well G-8 are tabulated in Table 15 and indicated graph- ically in Figure 9. 3. Supplementary Soil Test Data Table 16 summarizes all the data (other than permeability) obtained from tests on disturbed and undisturbed soil samples. These data include, percent moisture, dry density, specific gravity, porositj', void ratio and sieve analyses. 4. Falling Head Permeameter The results of the permeability runs made on the "pre-chlorinator" and "post-chlorinator" percolation tubes installed at ^Manhattan Beach are reported in Table 17 and shown in Figure 10. F. DISCUSSION OF RESULTS 1. Preliminary discussion of geochemical changes. Before entering the discussion of the geochemi- cal changes that apply to the intrusion and re- charge phases of the West Basin Barrier Test, a brief discussion of the geochemical changes that occur in groundwater is in order. Tlie following has been taken from various sources.-''" The jzeocliemical changes that occur in groundwater result from : (1) Evaporation near the surface (2) Solution of soil minerals (3) Precipitation of salts (4) Ion exchange between cations in water and in the soil (5) Sulfate reduction from microbiological activity (6) Admixture with waters of different types and concentrations a) Evaporation, Solution and Precipitation These processes affect the amounts of an- ions and cations present. These processes are not important in the geochemical changes occurring during sea-water infiltration or recharge. b) Ion Exchange This process affects only the cations present in water. The sum of the cations as equiva- lents per million will not be influenced, though the amounts of each cation will be altered because of ion exchange. The relative ion ex- change activities of the three dominant cations in water are : calcium > magnesium > so- dium, that is, when the concentration of each cation is relatively low, the tendency is for the calcium to leave the water and enter the soil and sodium to enter the water from the soil. The reverse will occur w-hen sodium is present in water in great excess as it is in sea- water. Both these tendencies are readily ap- parent in the analyses of the West Coast Basin Water Samples, the first during re- charge and the second prior to recharge. The influence of cation exchange on permeability will be included in the discussion of perme- bility. c) Sulfate Reduction This process which affects only the anions, sulfate and bicarbonate (the latter increasing at the expense of the former), requires the presence of both sulfur reducing bacteria and organic matter. This process is not active in the geochemical changes observed in the West Coast Basin Barrier Test. d) Mixture of Waters of Different Types and Concentrations When groundwaters mix, little if any re- action will occur between them and the only change will be that of concentration. When two waters of different types are plotted on '• " See Bibliography. 96 SEA WATEK INTRUSION IN CALIFORNIA tri-linear charts (see Figures 5 and 8) a line drawn between them will be the locus of all waters that could result from mixing the two waters. In Figure 5 the recharge (M.W.D.) water is represented by point "R" and the pre-recharge analyses of Well G-H by point "1". The dotted line connecting these two points contains all the mixtures that could result from mixing "R" and "1". However this relationship is valid only when geochem- ical changes resulting from cation exchange (aifecting only cations) and sulfate reduction, (affecting only anions), do not occur. In fact any deviation from this straight line relation- ship is an indication as well as a measure of the geocliemical changes that result from ion exchange, sulfate reduction, precipitation and solution. This straight line relationship be- tween "original" and "diluting" waters is also used to determine which one of two or more "original" (or "diluting") waters when mixed with a known "diluting" (or "original") water yields a known mixture. Where sulfate reduction is known to have little or no effect on the water the anion tri- angle can be used for this purpose. By this means it is possible to determine whether a known connate or sea-water is responsible for an increase in chlorides, or whether a reduc- tion in chlorides has resulted from dilution by native or recharge water. Although a mixture may not be altered by sulfate reduction, if one of the "original" waters has been so altered the resulting mix- ture may deviate from the straight line established by the "original" waters when plotted on the tri-linear graph. The reason for this is that connate waters which contain very little to no sulfate ions due to sulfate reduction may pick up by solution, appreci- able quantities (24 to 142 ppra) ^ of barium. Because of the high degree of insolubility of barium sulfate it is not usually present in waters containing any sulfate ions. When connate water containing sufficent barium ions is mixed with normal groundwater con- taining sulfate ions, barium sulfate will pre- cipitate thus depleting the concentration of sulfate ions. This depletion will be reflected as a deviation from the mixture line on the anion side of the tri-linear chart. ^ Of the 6 major ions in groundwater, cal- cium, magnesium, sodium, bicarbonate, sul- fate and chloride, only the chloride ion is unaffected by either ion exchange or sulfate reduction and thus becomes a reliable param- eter for measuring sea-water intrusion or effect of recharge. The tri-linear chart referred to, is a means of graphically characterizing water quality. This form of tri-linear plotting is described in a paper by Raymond A. Hill previously referred to." Cations in %epm can be repre- sented as a point on the cation triangle and similarly, the anions on the anion triangle. The combination of anion and cation is rep- resented on the "diamond" as a point of intersection of lines drawn from the cation and anion points parallel to the sodium and chloride base lines respectively. This is shown on Figure 2 for Well TH-3. 2. Geochemical Changes in Groundwater and Soils Resulting from Sea- Water Intrusion The quality of the groundwater at the Man- hattan Beach site of the West Coast Basin Bar- rier Test has been altered as the result of the admixture of native groundwater and sea-water, and also cation exchange. The concentration as determined from chloride analyses ranges from essentially sea-water near the ocean to slightly polluted water near Sepulveda Boulevard. Par- allel and indirectely related to the change in concentration is the change in water quality re- sulting from cation exchange. Cation exchange was evident ; though slight near the ocean it in- creased rapidly landward with the decrease in concentration of chlorides. This change is re- flected in the decrease in sodium and increase in calcium in the water and the decrease in percent saturation of the soil. Seaward the soil contain- ing montmorillonitic minerals responsible for cation exchange, have by long exposure to sea- water become saturated with exchangeable sodium. Recently intruded sea-water exposed to the sodium saturated soil will show little or no exchange of sodium in the only direction pos- sible for sea-water, i.e., water to soil. However inland the soil becomes progressively less satu- rated with exchangeable sodium thus permitting cation exchange to increase landward. This cation exchange phenomenon is demonstrated in Table 1 and Figure 2. The zones indicated in Table 1, Figure 2 and the map in Figure 3 were arbitrarily defined on the basis of %epm of calcium as follows: Zone I, to 2.50 %epm calcium, includes 13 wells and sea-water, % saturation 89-100 Zone II, 2.5 to 4.0 %epm calcium includes 9 wells, % saturation 85-94 Zone III, 4-10 %epm calcium includes 7 wells, % saturation 52-79 ' See Bibliography. " See Bibliography. SEA WATER INTRUSION IN CALIFORNIA 97 Zone IV, 10-20 %epm calcium includes 8 wells, % saturation 7-27 Zone V, 20% plus epm calcium includes 7 wells, % saturation 0-9 When the analyses (in terms of %epm) of the water in the wells in each zone are plotted on the tri-linear chart of Figure 2, the boundaries of each zone are well defined with certain excep- tions as follows: a) Zone 1-A — Well K-9 ; geographically this well is in Zone 1 ; on the basis of the chloride concentration it should be in Zone V. It is the only well where cation exchange is in the re- verse direction, that is the direction of sodium is soil to water. This well contains water with a sodium concentration as ppm about 8 times that of its combined calcium and magnesium con- centration. The reason for this is that the water has been exposed to a soil saturated with sodium. A high sodium soil, as has already been stated, is indigenous geographi- cally, to Zone I. The source of the water origi- nally was meteorological, garden or lawn or possibly septic tank or cess-pool. The low nitrate and ammonium content renders the last two possibilities doubtful. b) Zone IV-A, Wells TH-2, TH-3, H-16 ; these 3 wells on the bases of water quality, geo- graphic contiguity and chloride concentra- tion, belong in one group. They are quite dis- tinct from the cluster of six Zone V wells to the northwest. They contain native water in two of the wells and slightly polluted native water in the third. c) Zone V-A, Well M-18; this well is very closely associated to the ones in Zone V. It is however, less polluted with sea-water than the Zone V wells. The bicarbonate content is correspondingly higher and is responsible for setting this well slightly apart from the Zone V weUs. Examination of Table I and Figure 2 yields the following additional information concerning the groundwater in this area: a) Although sodium in the water shows a defi- nite tendency to replace the calcium in the soil, magnesium (with sodium) may replace the calcium at one location and be replaced by sodium at another location. The reason for this inconsistent behavior of magnesium is because of its relatively high concentration in sea- water compared to calcium (calcium (ppm) : magnesium (ppm) := 1:3). It will therefore have the tendency under certain conditions to be absorbed by the soil minerals along with sodium at the expense of calcium. Experiments in which soil was mixed with sea-M'ater showed the absorption of both sodium and magnesium.'' The results are pre- sented in the following table. Percent Exchangeable Cations in Soil Afagne- Potat- Calcium sium sitim Sodium 71.7 19.8 4.5 4.0 20.3 37.1 6.7 35.9 —51.4 -1-17.3 4-2.2 -(-31.9 Before Treatment After Treatment. Percent Change It is apparent that there is as much ex- changeable magnesium present as exchange- able sodium in the treated soil, and that the increase in percent, of exchangeable magne- sium (17.3%) is appreciably less than the gain or loss of exchangeable sodium and calcium (31.9%, 51.4%). This explains why magnesium does not show as much exchange activity as the other two cations. The satura- tion of soil with magnesium occurs before saturation of the soil with sodium, and as the absorption of sodium progresses the ab- sorption of magnesium decreases to zero, and magnesium may then be displaced by sodium. Table I illustrates the changes in the direc- tion that magnesium takes with changes in degree of saturation. In more highly satu- rated Zone I the tendency is for magnesium to be displaced by the sodium. In less saturated Zones II and III the proportion of wells in which magnesium displaces calcium increases. Zone V wells are the exception. The magne- sium is displaced by sodium in all the weUs. During the 1950 pilot recharge test one of the weUs in Zone V, M.B. No. 7 was used as the recharge well, and four of the other wells in this zone were used as observation wells. This might have altered the exchange- able cation status of the soils in Zone V. b) The sulfates show a narrow range from 1.85 %epm to 5.60 %epm. In the higher concen- trations this range is even narrower. This slight variation in %epm of sulfate is an indication that the polluting source is sea- water rather than connate water in all the wells listed in Table I. c) The concentration of wells into small con- tiguous zones on the anion side of Figure 2 as compared with the much greater spread on the cation side is an indication as to the relative changes in water quality resulting from dilution of sea-water with and without the influence of ion exchange. ' See Bibliography. 98 SEA WATER INTRUSION IN CALIFORNIA The source of the pollution iu all the wells is believed to be sea-water as indicated in the above paragraph. The analyses of the trace elements as borate, iodide and bromide (see Table 3), in Wells C-12, MB-4 and K-12 confirms this to some degree. The amount of borate ion iu sea-water is within the connate range but close to the minimum. The borate found in Wells C-12, MB-4 and K-12 is even less than what should be expected in sea-water when diluted to the chloride content in each sample. This is not definite proof of sea-water pollution as against connate water pollution. The iodide content iu sea-water is consider- ably below the range found in connate waters. The iodide found in Wells MB-4 and K-12 within the limits of analyses, corresponds to the theoretical iodide concentration for sea- water pollution. The iodide iu Well C-12 is above this theoretical value but still well below the connate range. This is a more conclusive proof of sea-water pollution. The bromide concentration in sea-water like the borate is within the connate range but toward the lower limit. Considering the range of bromide possible if connate water was the source, good correlation therefore exists be- tween theoretical sea -water concentration and actual concentration particularly in Well C-12. The %epm for sulfate in these wells is well above the connate range and approaches the %epm for sea- water. Very good correlation exists in the chloride %epm between the weU analyses and sea-water. 3. Geochemical Changes in Groundwater and Soils Resulting from Recharge a) WellG-H The quality of the water in Well G-H was altered by the recharge in the adjacent Wells G and H. The rapid decrease in concentration shown in Figure 4 between June and Septem- ber of 1953 reflects the increase in recharge rates in this period. The fluctuation iu con- centration between September of 1953 and February, 1954, was caused by the temporary cessation of recharge in Well 11 and a subse- quent decrease in Well G. It is apparent that an appreciable lag exists between changes in recharge rates and couceutration. The increa.se in concentration in November and December of 1953 resulted from the return of the more saline water which earlier had been pushed toward Well H. 'I'lie resiiiiiption of a steady rate of recharge eventually succeeded in slowly pushing this saline water either sea- ward or inland. With recharge the cation exchange reverses direction. At first it is mainly between sodium and calcium with the calcium entering the soil at the expense of the sodium. At a con- centration of about 9000 ppm of chlorides, magnesium enters into the exchange dis- placing the sodium. With further dilution the exchange increases, with magnesium showing greater activity than calcium. The direction taken by magnesium (water to soil) is the same for recharge and pre-recharge condi- tions (except, as noted, for saturated condi- tions, i.e., Zone I) with the greater activity displayed during recharge. This unidirec- tional activity of magnesium is explained by the fact that sodium is no longer competing with magnesium in the exchange reaction but, in fact, is now on the opposite side of the exchange equation, viz : Pre-rechaxge — Na* + Mg** + CaX ? Post-recharge — Ca** + Mg^* + NaX ; : NaX + MgX + Ca** ± CaX -f MgX + Na* where cations in solution are represented as Na* etc. and cations combined with soil is represented as NaX etc. Magnesium, being the more abundant cation of the two, enters into the exchange reaction to a greater degree than calcium. The tri-linear chart Figure 5, shows the water quality "path" taken by the ions as recharge progresses. The deviation from the straight line defined by the original water in G-H and the recharge water (points "1" and "R" in the cation triangle) clearly shows the influence of cation exchange. No such devia- tion appears on the anion side, the points following very closely the mixing line as anticipated. b) G-Wells The analyses of the change in the quality of the water with dilution in the G Wells as shown in the integrated barograph, (Figure 6), demonstrates that the degree of cation exchange increases as the recharge water pro- gresses inland from the recharge wells. This is indicated by the increase in the separation of the calculated from analysed dilution lines from Well G-2 to Well G-8. This separation occurs only on the cation side of the baro- graph. On the anion side the two lines coin- cide indicating dilution is the only factor in- SEA WATER INTRUSION IN CALIFOKXJA 99 volved on this side. This increaso in cation exchange Landward is ehiborated in Fignre 7. Well G-2, the one nearest the recharge -well, shows an increase in cation exchange only at extreme dilntion (concentration 20 epm). The next well, G-4, shows a peak exchange at 95% dilution (concentration 74 epni) and Well G-8 has its peak at 35% dilution (con- centration 200 epm). In terms of equiva- lents per million of exchange these peaks are represented by the following quantities : Calcium Magnesium Sodium G-2 —1.2 —0.6 -f 1.8 G-4 —1.1 —3.2 +4.S G-8 —2.0 —7.0 +9.0 Subsequent to the peaks observed in wells G-i and G-8 cation exchange falls off rapidly with dilution so that no (or very little) cat- ion exchange will be evident by the time the unpolluted recharge water reaches these weUs. The following discussion is presented to explain the difference in cation exchange ac- tivity from Well G-2 to Well G-8 as evi- denced in Figures 6 and 7 and discussed in the preceding paragraphs. As recharge water displaces the sea-water inland, the sea-water will become increasingly diluted with the re- charge water at the interface. Very slight exchange activity is observed until a con- centration of 200 epm to 300 epm is reached at which point the concentration of sodium in the water has been reduced enough to permit the reaction : Ca** + Mg- + NaX ?=> CaX -f MgX + Na* to advance to the right. The amount of ex- change that will then take place naturally depends on both the quantity of sodium saturated soil the water is exposed to, the degree of saturation and the time in which the water is in contact with the soil. The water reaching Well G-8 has been exposed to more high-sodium soil than water reaching Well G-4 although the soil between Well G-4 and Well G-8 shows a reduction in exchange- able sodium saturation from 82% to 27%. The same thing can be said of the water reaching Well G-4 compared to Well G-2, although here the sodium saturation de- creases slightly, from 84% to 82%. Follow- ing the increase in exchange activity the soil becomes gradually depleted in exchangeable sodium and the exchange activity lessens. At Well G-8 the fall-off in exchange activity starts at a higher concentration than G-4. This is because of the reduction in exchange- able sodium saturation of the soil between these wells. The groundwater which reached Well G-8 as result of recharge, first showed a decrease in concentration probably due to the dis- placement of a fresher body of water existing seaward of Well G-8. As the saline wave proceeded inland the concentration in- creased and as the wave passed Well G-8 the concentration decreased again. Therefore waters of the same concentration reached Well G-8 ahead of the saline wave and then later following the wave. The water follow- ing the wave shows a greater degree of ex- change than the initial water. The curves for Well G-8 in Figure 7 show that this increase amounts to 3% epm at a concentration of 220 epm. Evidenth- tlie water following the wave was in contact with a greater mass of sodium saturated soil than the water which pre- ceded the wave which of course was the case. The water at G-2 unlike the water at G-4 or G-8 shows no fall-off in exchange activity although the water at this well is unpolluted recharge water. The reason for this is two- fold, (1) the higher level of exchangeable sodium saturation in the soil, (94% to 84%) between W^ells G-1 and G-2, and (2) a more rapid displacement of the water at those concentrations that would permit re- lease of much of the sodium from the soil at this location (see Table 2B) ; thus the con- tact period required for cation exchange is relatively brief. The soil therefore, between Wells G-1 and G-2 has not been depleted in exchangeable sodium to a degree that would inhibit exchange, particularly with diluted water that exists between these wells. Just how long this exchange capacity will be maintained can only be determined from sub- sequent analyses. c) Wells K-8 and K-12 The pre-reeharge and latest analyses for Wells K-8 and K-12 are plotted on the tri- linear chart (Figure 8). The analyses of the two possible sources of dilution of the re- charge water, (viz: native and recharge wa- ter) are also plotted. As stated previously (see page 95) one of the lines drawn down between the pre-reeharge analyses (on the anion triangle) and the diluting water anal- yses will contain the point representing the latest analyses. For Well K-8 that line is the one connecting recharge water analysis with pre-reeharge analysis. For Well K-12 the line connects pre-reeharge water analysis with 100 SEA WATER INTRUSION IN CALIFORNIA native water analysis. Therefore the water in K-8 is being diluted by recharge water and the water in K-12 by native water. The pre- recharge analyses for Well K-12 falls on the K-8 native water mixing line as would be ex- pected. 4. Control of Chlorine Dosage Since chlorine is added to the recharge water to prevent bacterial slimes from accumulating at the well perforations, a criterion for controlling the dosage to achieve this should be developed. The criteria used at Manhattan Beach were : (a) Adding sufficient chlorine to the recharge water to obtain a miuumum residual in the adjacent "20 foot" well. (b) Adding sufficient chlorine to obtain a minimum bacterial count in the "20 foot" well. (e) Adding sufficient chlorine to obtain maxi- mum acceptance of the recharge water. The chlorine dosage was gradually reduced from the initial 20 ppm dosage to 1.5 ppm, the lowest concentration that would satisfy the first two criteria. However the recharge well head (at constant recharge rate) started to increase at this low dosage. The dosage was then increased to 5 ppm which resulted in an appreciable de- crease in the recharge well head. A low bacterial count and a detectible residual chlorine in the "20 foot" well are therefore not reliable criteria for controlling chlorine dosage. The following discussion is presented for a better understanding of the factors responsible for the inadequacy of the above two criteria and to shed some light on this problem.^ When chlorine gas is added to water it com- bines with the water to form hypochlorous and hydrochloric acid. The hypochlorous acid further dissociates into hydrogen and hypochlorite ions. The proportions of these components present de- pends on the pH as follows : Below pH 5 — molecular chlorine pH 5-6 — ^hypochlorous acid pH 6-7.5 — hypochlorite ions increa.se Above pH 7.5 — hypochlorite ions predominate The recharge water with a chlorine dosage up to 20 ppm will therefore have the chlorine as hypochlorite ions, (See Table 5). Chlorine in this form is defined as free available chlorine. Chlorine in this form is highly reactive and will either combine with inorganic matter or oxidize inorganic matter, (See Table 7 for its effect on iron). Chlorine will also combine with and oxidize or- ganic matter, and in so doing it can, by coagula- tion or precipitation, alter the physical state of organic compounds. Chlorine in the combined form is defined as combined available chlorine. When a small dose of chlorine is added to wa- ter the reaction is mainly that of combination. As the dosage increases free available chlorine ap- pears. When a sufficiently large dose is added, the free available chlorine predominates and oxi- dation occurs. The oxidation reaction is desired in the re- charge water to kill bacteria and to effect a bene- ficial change in the physical state of the organic slime, at the perforations. The amount of chlo- rine required to assure an oxidation reaction de- pends on, (1), the chlorine demand, i.e. the amount of inorganic and organic matter in the recharge water and in the well itself that wUl combine with or be oxidized by the chlorine, (2), time of contact and, (3), the water temperature. Since these factors, particularly the first one, are difficult to evaluate no direct method for the determination of the optimum chlorine dosage is therefore possible. An indirect method, the 3rd criterion of adding sufficient chlorine to permit maximum acceptance would appear to he the logical one to use to adequately control chlorine dosages. The one difficulty inherent in this method is that even an optimum dosage as determined by the third criterion may permit an iusiduous bacterial or organic growth to collect at the per- forations without it being reflected in a reduc- tion in acceptance. By the time the reduction is noticed a sufficient quantity of slime may have accumulated that would be difficult to remove even by increasing the dosage in small incre- ments. However by giving the recharge well a "shock" treatment of 20 ppm or more, it may be possible to remove most of the accumulated slime. 5. Relationship of Permeability to Cation Exchange and Concentration Sodium soils are more impermeable than cal- cium soils.'' It would therefore be expected that a soil in contact with saline water woiald on con- version to a sodium soil become more imper- meable, and the reverse would occur when the saline water was displaced by a recharge water containing a normal amount of calcium. This change in permeability, resulting from cation exchange, is appreciably influenced by the con- centration of salts in the water to which the soil is exposed. The high concentration of salts in ' See Bibliography. » See Bibliography. SEA WATER IXTRUSION IN CALIFORNIA 101 sea-water may offset the reduction in \>or- meability from cation exchange, due to the flocculating effect of the charged ions. When the sea-water is displaced by recharge water any increase in pornicability result iiig from ion ex- change will be olTset again by the removal of the flocculating effect of the charged ions. This has been demonstrated in the article on the '"Sealing of the Lagoon Lining at Treasure Island with Salt" by Charles H. Lee.^" When fresh water was pumped into the clay lined lagoon, the seepage rate was 0.90 inches per day, which was excessive. Sea-water then replaced the fresh water to convert the clay to the sodium type. With sea-water the seepage rate fell to 0.60 inches per day. When fresh water replaced the sea-water in the lagoon, the seepage rate fell to 0.10 inches per day in 3 months. Only a 33^% decrease resulted in the conversion of the soil to the sodium type in the presence of sea-water, but when the sea-water was displaced with fresh water an additional 83^% decrease ensued, making a total decrease of 88.9%. The same type of decrease, resulting from dis- placing sea-water with fresh water, was ob- served in a laboratory permeability test of sample #ol from Well "G-8, (see Table 15 and Figure 6). The mean permeability with sea- water was 19.9 feet per day. Following the change to M.W.D. recharge water, the mean permeability fell to 16.8 feet per day. After the soil was exposed to the fresh water over the week end the mean permeability rose to 19.3 feet per day. This final increase in permeability was due to conversion of the soil to the more permeable calcium type. Though most of the conversion to the calicum soil occurred the first day, (see Table 15, last 3 columns) some addi- tional time was required for the physical change to occur. The maximum displacement of the sodium ion from the soil (7.14 %epm) occurred at 10:03 AM, one and one-half hours after the influent was changed to fresh water. At this time the permeability was the maximum for the first day, i.e., 18.6 feet per day, but the effluent still showed an appreciable proportion of sea- water. For the remainder of the day the per- meability fluctuated above and below a mean of 16.7 foot per day and the amount of sea-water in the effluent decreased rapidly until prac- tically none was present. The increase of chlorides (see Table 15 and Figure 6) in the effluent, following the week end standover, probably resulting from the interim diffusion and mixing of the more saline water trapped in "dead spots" during the first day's run. 6. Presence of Deleterious Substances and Their Effect on Recharge a) Dissolved Oxygen in the Recharge Water The average temperature of the recharge water is about 18° C which would permit a dissolved oxygen content of 9.6 ppm. Table 5 shows that in some cases, particularly on September 15, 1953 and January 11, 1954, the concentration exceeded the maximum solubility of oxygen. The excess was appar- ent in the entrapped air bubbles observed during sampliug. Trapped air is considered deleterious because of its detrimental effect on permeability of soils. b) Iron in the Recharge Water Table 8 shows an appreciable increase of iron in the recharge water resulting from its passage through the 20" asphalt coated iron pipe. Although the amount of this iron i)ick- up is too small to adversely affect the recharge water it does indicate corrosion in the line. Tables 6 and 7 on the other hand show the presence of sufficiently large amounts of iron in the recharge and observation wells to indi- cate a high degree of corrosion of the casing. The formation at the perforations of a gela- tinous iron hydroxide which is encouraged by the presence of oxidizing substances, chlorine and oxj'gen, would tend to reduce the accept- ance of the recharge water. In addition to the damage to project facili- ties and reduction of permeability resulting from corrosion, the accumulation of corrosion products in back pressure valves and metro valves (see Tables 9 and 10) resulting from the corrosion of aluminum as well as iron may seriously affect the function of the equipment. c) Bacterial Slimes and Organic Matter The effect of these substances on permeabil- ity has already been discussed. Another effect of the bacterial slime, already referred to but not explained, is its filtering action on bac- teria. The low bacterial count that has been observed in the water samples taken from the adjacent "20 foot" well, may have resulted from this filtering action rather than the bac- tericidal action of the chlorine. This explains why the second criterion for control of chlo- rine, i.e., a low bacterial count, is not a reli- able one. This filtering action of the slimes has been observed during pollution -travel studies conducted by the University of Cali- fornia at its Richmond station. -52568 102 SEA WATER INTRUSION IN CALIFORNIA The results of the tests involving the two percolation tubes that were installed at Man- hattan Beach show an appreciable difference in the clogging action of chlorinated and non- chlorinated water (see Table 17 and Figure 9). Water without chlorine left a brown deposit on the top layer of Ottawa Sand at the outset, which was reflected in a persistent decline in permeability. The deposit in the other tube appeared later and was never as great. The permeability in this tube after the initial drop fluctuated about a mean value appreciably above that found in the other tube. No attempt was made to prevent the suspended solids in the water from contrib- uting to the decrease in permeability so that the changes in permeability reflect the pres- ence of suspended solids in the water as well as the slime forming bacteria. d) Presence of Sludge in the 6" Line and in the Chlorinator Bell-Jar Appreciable amounts of a white powdery sludge have appeared in the Chlorinator Bell- Jar and in the 6" line. Samples of the mate- rial taken from these sources indicate that the sludge is primarily calcium carbonate (see Table 11). This sludge has been brought in by the recharge water and may have resulted from the erosion of the concrete M.W.D. feed lines. It could be considered a deleterious sub- stance for two reasons: (1), It may affect the operation of project equipment, such as the chlorinator, if permitted to accumulate to a sufficient degree; (2), when combined with bacterial and iron hydroxide slimes it may augment the clogging action of these sub- stances. G. CONCLUSIONS 1. Source of Groundwater Pollution No evidence has been found to indicate that the source of pollution at the Manhattan Beach site of the West Basin Barrier Test is other than the ocean. 2. Effect of Recharge on Soil and Groundwater a) Soil As the fresh water displaces the sea-water landward a gradual conversion of soil will occur. The sodium will be exchanged for the calcium in the soil and the permeability will tend to decrea.se at first but then increase as the sea-water is completely displaced. These changes in permeability may be obscured by other factors difficult to evaluate. b) Groundwater (1) Recharge water is altered by cation ex- change when it contacts soil which previ- ously had been exposed to highly saline water. This cation exchange activity which softened the water is appreciable within 300 feet of the recharge well but diminishes further inland. (2) The dilution of water observed in Well K-12 was found to be caused by admix- ture with native water rather than re- charge water. 3. Control of Chlorine Dosage The chlorine demand studies and quantitative bacterial analyses have been proven an unreli- able means of controlling the chlorine content of the recharge water. Maintenance of a maximum acceptance is a more reliable method as long as an excess of chlorine is present to prevent insid- uous growth of bacterial slime. Any change in the recharge well head would not be sensitive enough to reflect such a growth, until too late. An Ottawa Sand percolation tube similar to the ones installed in the chlorine storage shed at Manhattan Beach is a more sensitive indicator. Periodic determinations of the permeability are sufficient to indicate any tendency toward the formation of slimes. For confirmation, a micro- scopic examination and bacterial analyses of a few surface grains could be made. Precautions should be taken to avoid passage of sludge and rust particles into the percolation tubes which would also reduce the permeability. This could be accomplished by introducing traps in the line feeding the percolation tubes. 4. Suspended Solids in the Recharge Water Suspended solids which may include sand, cal- cium carbonate, rust particles and zeolite par- ticles will not be apparent in a closed system except where some may be caught in the chlori- nator bell- jar H. REFERENCES 1. S.'ilinp and Alkali Soil.-* — liy L. A. Hiohards et al.. Ajiricul- tiirc Haiulljodk No. 60, U. S. Department of Agriculture, llir.4 2. United States Geological Survey, Water Supply Paper 1136 — Native and Contaminated Ground Waters in Long Beach, Santa Ana Area, California — by Piper, Garrett et al., pg. m. Table 8, l!tr)3 .3. Hydrology — edited by O. L. Meinzer. Chapter on Chemistry of Groundwater 4. Criterion for Recognition of Soa-Water and Ground-Water —by Roger Revelle. Trans. A.G.TL, Part III, pg. 593. 1941 5. Cation Exchange in Ground-Water Contaminated with Sea- Water near Miami, Florida — liy S. K. Love, Trans. A.(;.U., Part VI, pg. 951, 1945 G. Salts in Irrigation Water — bv Raymond A. Hill, Trans. A.S.C.B., Vol. 107, pg. 1478, 1942 SEA WATER INTRUSION IN CALIFORNIA 103 7. Rase Exchange in Relation to Composition of Clays with Special Ueference to Sea-Water — Kelly and Liebig, Am. As^s.H■. of ri'troleiim Ccol. lUill. IS. w. :i.'.S-;{(;T. VXi', 8. Water Quality and Treatment — American Water Works Association, pg. 205, 1951 9. Cation Exchange in Soil.s — by Walter P. Kelley, 1948 10. Trans. A.S.C.E., Vol. lOG, pg. 577, 1941 T-l'le I. TABLES Number 1. Pre-Reoharge Analyses 2A. Effect of Recharge Well GH 2B. Effect of Recharge Well G-2 2C. Effect of Recharge Well G-4 2D. Effect of Recharge Well G-8 3. Source of Pollution, Wells C-12, MB-4, K-12 4. Dissolved Oxygen in Ground Water 5. Field Analyses of Recharge Water 6. Iron in Well G after Cave-In 7. Iron Precipitated in the Presence of Chlorine S. Iron Pick-Up in 20" Feed Line 9. Deposit Inside Back Pressure Valve 10. Deposit Iiisiile Metro Valve 11. Sludge in CbUiriuatnr Bell .Jar and (J" Line 12. Analysis in Chlorinator .lar Containing Sludge ];!. PeriMi'ability of Soil Samples — Metal Pernieanieter 14. Permeability of Soil Samples — Lucite Permeameter 1."i. I'crincabilily of Sample #."i."i. Well (J-S IG. Summary of Soil Test Data 17. Falling Head Permeameter Test at Manhattan Beach J. CHARTS Figure Number 1. Permeameter Sketch 2. Tri-linear Chart of Pre-Recharge Conditions 3. Map — Pre-Recharge Conditions 4. Graph^ — Effect of Recharge on Well GH 5. Tri-linear Chart — Effect of Recharge on Well GH 0. Bar-graph — Effect of Recharge on G Wells 7. Tri-linear Chart^Effect of Recharge on G-Wells 8. Tri-linear Chart — Source of Dilution Wells K-8 and K-12 9. Graph— Permeability Test, Soil Sample S55, Well G-8 10. Graph — Falling Head Permeameter Test 104 SEA WATER INTRUSION IN CALIFORNIA 6? ©■^■^OOOCOOOt^OiO ■* © lo to t* OJ • « ■* 00 ■* CO Oi o to i« CO O W5 ^ t^ to a to CO o t- 00 M W ^ 11* — ' N M CO r- CO lo CO N ©i 9 9 9 9 9 9 9 ° 9 o o P <^ o -^ o ^ Q^ oi O) a MWiMfOCOCOCO'^'^ lOifjiOiCiCi^iOiOi** ■< « M M CO CO ■* CJ CO CO CO lO - < uj UJ ^ o »- < X u ■^ 'J' ■^ -J" Tjt ■^J" ■. to >o CO r- r- 00 O t- -^f 00 Tf -fJ" CO ■* o ^ ^ ^ ^ ^ O O lO t- O 00 r- . X CJ O O; t- to o oi X X r^ t-- to »o in Tf (N -^ Tf --. to Tj" O Tj" CO ■— .-H -^ "* Tf« O iM lO to 8^1 "IS ooocoooooooo OlOC'^-S'OOOO-rftOO t--tqcO_'^'«3^"*COOOtDCO—' 00 X' X' 00 00 00 X* x" x" t-' t-* 1^* ooooooooo OOOOtDWOOO cjtotqooi^oot^»0'«j' r^cDcotO'Ou^riC'C'i^ o o o c O O ?J c d r- lo c^ coo C CJ X O CO -J- i Cft LO o o CM — ' CO lO o to o CM CO "* CM ■-> •-< h, lO K5 ■<)' CM w o r- TT CM fM — < k me IV C-I2. G-8- K-12. A-U- M-14. a o N COtc! W ^a o o o to c^ o O CM to — I-- CM O C: t^ to OS 01 ro K N K J cj I -^ •^ O . SEA WATER INTRUSION IN CALIFORXIA 105 O o < I u O I— o 1 o M Ed a o s SSoggSSgSS3t:2 « — N — -H — «NNe0M»O f?7tttttTT7t d —0.99 —1.38 —0.91 —0.09 +0.12 —0.24 —0.69 —0.95 -0.59 —1.10 —0.88 —1.75 1 1 s 3 O 6 CO CS CO COCCCOCCCOCOCOCOCO M s •-■oaoor»ooa)909>ao iflt-iCiOOOiMOO'-'^WON CJCO»COOCOO:>.';'^».'5«0001 " s fe CO o CO ooaoo — ^■^ — (Ct-cooo — t^t^O^ — — N0O'ocj>.'5asr-»oco coOOTi«t^rC(^'^coco«o NCIfNCOCOeOC^C^NNCl^ Total Dis. Solids opm s 01 mocowcjt-OrtOioo^ c o 6« ^ t-WCCN— 'NCOCOCOMNi-H ssg o o «o- «D ■*" CO V © r-' <©" "5 V M Days of Roolig. o QOt^ooffl'C-'cecoaoosooco C*r-CaCM'^K5OCiC0t^O00 — — — MC4CJCOCOCCeO-*'-0 0! «ooaot-o*oeoooooo«5 tf — ^ « — 1-. 6 O < X u UJ Of U Ui s « , r- o r* u5 «5 o ® iS O — — OJ ■* © 00 +++++++ feS '* 04 (D CO !>• r* t^ O -ff Oi « h- -!■ I^ s s TTTTTTT u OS -• © Oi p o w -H ' •^ is. t, (O © "^ 01 1-t -H CO CO r- "tp o o 5 N >.M J3 „ « « « ^ 01 ^ «°l 1 S§SS5SSSS Ol O © tO © t^ ^ s .51Si^Sl^Si-, «t*(^0000C!0»O & OS Q 106 SEA WATER INTRUSION IN CALIFORNIA 6 I o Of < u O 1- o 6S 5 is3 S o« 03 o V eg +++++++++ o o o o + ■ ■ ■ iTTTTT' OppOOOr-1— .po I IT I++I III COOOOOXOOOOt-b-tOiO MCOrCCOMCOCC'^'i'CO C000O0COt^(O'^ O CO OO0O0OOM"rf*O cDOCJtDOCGtOCOCDO '^^ ''1 ®. '^- 00^ (N t-_ CB »ra (N ITS Tf -«"■ oj' co' ■*' i-T Cs) ^ ^ ^ c^ c^ 00 6 o < X u u s a. e? s? ^-t m jQo o« &-S Oj o " t- W 'J" ^ ^ cc c^ +++++++ tl T §5 00 N C "". TTTTTTT GO 00 00 00 00 X 00 O O O O O Oi Ci ^ If ^ rf ^ iC lO t^ t^ ^. t^ X « b- oi 00 t- r- CD lO 00 .-. t- — tC Cft Ci « ^ CD -*• —" t^ — ' 00 O) (N CM C4 ■-« ^ t- b- CO ec >o i-< cc o o o o o o o "Tf (N O CD CO O O CO ci^ oo_ co_ r-_ 00 « co" ■^' co" co" w" —«" f' CO t-- O C-l (M CO h. CO ■* Oi Ol I^ CD (N N IN W CO CO -^ lO 3 to X SEA WATER INTKUSION IN CALIFORNIA 107 TABLE 3 SOURCE OF POLLUTION Source Date of Sample Depth Chlor. ppm Borate ppm Iodide ppm Bromide ppm Calrium % epm Magnesium % epm Sodium *7 epm > Bicarbonate % epm Chloride % epm Sulfate '~c epm •Theor. Bor. C-12.. MB-4 K-12. I •Theor. I. C-12.. MB-4 K-12. •Theor. Br. C-12.. MB-4 K-12. Sea NVatcr 18.980 25 .05 65 1.7 8.8 39.5 .2 45.2 4.7 5.5 3.2 4.2 .010 .006 .009 14 S 11 Connate Water Max. 13.319 386 80 200 6.6 4.8 47.9 12.3 49.8 .7 120 71 92 25 15 19 62 37 49 Min. Wells C-12 6/18/52 219' 3,384 4.140 1.6 30 2.0 25 13.0 1.0 15.6 0.1 9.1 41.8 25.3 .2 1.6 37.5 44.0 4.4 9 6 7 8 5 6 MB-4 5/29/52 311' 440 .8 <1 14 23.5 12.0 14.5 1.4 45.1 3.5 6/17/S2 248' 3,220 1.5 <1 8 20.6 9.2 20.2 1.3 44.6 4.1 * Theoretical ccnceotration based on a dilution ratio calculated rrom chloride concentrations. TABLE 4 DISSOLVED OXYGEN-GROUND WATER Date Dissolved 0.xygen Remarks Well Max. Min. Mean E..-- 12/ 5/52 2/12/53 2/ 4/53 1/11/54 2.6 .8 1.1 1.8 1.3 .6 .6 1.5 2 .7 .8 1.7 Turbulence caused a steady increase in dissolved oxygen. H I G-2 TABLE 5 FIELD ANALYSES OF RECHARGE WATER Date Source pH Dissolved Oxygen ppm Residual Chlorine ppm Remarks 6 23 '53 Pre-Chlor 8.35 7.45 7.60 7.64 7.44 7.50 7.65 8.8 7.0 4.6 7.4 8.1 8.4 4.6 li'.s 10.4 12.2 9.2 7.6 All analyzed in field house. SampUng technique at WeUs 6 23 '53 Post-Chlor. caused air to be entrapped during sampling. 6 23/53 Well E ... «/23/53... Well G 6,23/53 WeU H «/23/53.— Well I-- 6/23/53 WeUK - 7/ 1/53 Pre-Chlor. 8.53 7.51 7.50 7.55 7.40 7.51 7.58 9.8 7.0 10.2 8.7 7.7 12.1 8.4 10.3 7.3 7.6 7.6 8.0 9.7 M\ analyzed at point of sampUng. No air entrapped during sampling. 7/ 1/53. 7/ 1/53 WeU E 7/ 1/53 Well G 7/ 1/53... . WeU H 7/ 1/53. WeU I .... 7/ 1/53 WeU K •/I5/53 Pre-Chlor 8.40 7.94 7.85 7.85 7.85 7.85 11.6 9.4 11.2 10.3 11.0 10.0 4^8 5.1 SO 5.2 5.2 9/15/53 Post-Chlor •/15/S3... WeU E •/15/53 WeUH •/15/53... . WeUG 8/15/53 WeUK 1/11/54. Pre-Chlor. 8.50 7.80 7.85 11.1 10.9 10.1 Same as for July 1, 1S53. 1/11/54 WeU E 1/11/54 WeU K 108 SEA WATER INTRUSION IN CALIFORNIA TABLE 6 IRON IN WELL G AFTER CAVE-IN Source Date Time Depth (Ft.) Conductivity ECXIO* Iron, ppm G 3-18-53 3-18-53 3-18-53 3-18-53 Pre-Chlorination 3-27-53 3-27-53 3-19-53 3-19-53 3-20-53 1250 1305 1315 1325 1215 1515 1450 1130 1520 188 189 155 145 193 144 158 193 185 1046 1048 1050 1049 1037 3 8 G 3 8 G.- - 3 5 G 4 MWD 1 FG 0.5 GH 5 G-1 0.2 G-2 2 G-3 . 2 TABLE 7 IRON PRECIPITATED IN PRESENCE OF CHLORINE Source Date Depth (Ft.) Iron (Raw) Iron Filtered FG 11-25-53 12- 7-,53 12- 7-53 0.6,0.2, >0.1 K-1 152 160 3.0 3.5 >0.1 E-1 0.6 TABLE 10 DEPOSIT FOUND INSIDE METRO VALVE Source: Well E Date:- August 1953 Description : White Powdery Crust Aluminum: About 20% TABLE 8 IRON PICKUP IN FEEDER LINE TABLE 11 SLUDGE FOUND IN CHLORiNATOR BELL-JAR AND 6" FEED LINE Source Date Iron ppm Remarks Vault 4-28-54 4-28-54 0.05 0.20 Chlorinated 0.15 Chlorinator Bell-Jar 6 Inch Feed Line 97.8% 96.0% 2.6% 0.6% 0.0% 3.2% *89.3% t88.1% Sihca- 1.9% .1% .2% 4.7% TABLE 9 DEPOSIT INSIDE BACK PRESSURE VALVE Source: WeU G Date: April 1953 Description: Red Brown Powder Iron and Al. Oiride: 75.0% Silica 1.5% • Calculated from analyses of calcium, t Calculated from loss on Ignition. TABLE 12 ANALYSIS OF WATER IN CHLORINATOR BELL-JAR CONTAINING SLUDGE Chlorine pH Calcium Magnesium Sulfate 860 ppm 2.6 313 ppm 5.2 ppm 306 ppm Typical MWD Water 8.5 33.7 ppm 10.5 ppm 348 ppm SEA WATER INTRUSION IN CALIFORNIA 109 TABIE 13 LABORATORY PERMEABILITY RESULTS-METAL PERMEAMETERS Depth Field Description Permeability Kt/Dny 96.4- 96.9 208.3-208,4 116.0-118.0 140.0-140.5 236.1-236.5 138.9-139.2 158.3-158.6 134.7-135.6 305.7-306.0 Li^ht brown sandy silt A brown silty fine sand Bl Gry Vy Fine Sand, Woodclups, Vy scat. ^" grav Yel brn fine to \-y coarse sand & grav. to 1", (average H"), occaaional pebbles to 2" Dark sray gravel to H" (average }'i") with fine sand, thin silty clay layers abund., shells- Gray fine to coarse sand & grav. ^"-2" Light brown fine to med. sand with scattered gravel toH" Tan fine to coarse sand Gray brown fine to coarse sand & gravel to H" Yellow brown very coarse to medium sand 5-20-52/ 8-19-52/ 4-15-53/ 3-26-53/ 5-12-53/ 8-22-52/ 6-12-52/ 8-29-52/ 7- 2-52/ 5-23-52 8-22-52 4-28-53 4- 8-53 6-13-53 8-27-52 6-20-52 9- 2-52 7- 9-52 5.0 3.45 21.55 6.70 30.38 2.80 20.! 18.4 57.5 •Hifsc ti'st'i run on the portion of the sample passing No. 4 sieve (less than one-quarter Inch). TABLE 14 LABORATORY PERMEABILITY RESULTS Sample Number Depth (Ft.) Field Description Dates of Test Permeability Ft./Day at 60° F Permeability Ft./Day at 60° F Well Sea-Water Influent Fresh-Water Influent Remarks Max. Min. Mean Max. Min. Mean I 34 202.5 to 202.8 Light gray medium to very coarse sand with scattered gravel — H " 4-21-54 to 4-28-54 45.6 27 37.4 44 27 34.4 Run intermittently. Partially an- alyzed influent & effluent K-4 27 255.7 to 256.0 Blue gray medium to fine sand 4-21-54 to 4-28-54 18.5 11.0 15.0 14.0 12.0 12.9 Run intermittently. Partially an- alyzed influent & effluent 0-1- 13 133.7 to 134.0 Gray silty fine sand scat, coarse sand and pebbles — ^ s " 4-12-34 to 4-15-54 4.4 1.5 2.7 Run intermittently. Partially an- alyzed influent & effluent L-1-- - 18 124.0 to 124.5 GrAy Rilty finp sand 4-12-54 to 4-15-54 6.0 .24 1.4 Run intermittently. Partially an- alyzed influent & effluent F-G 12 140.8 to 141.1 Gray brn silty fine crse sand w/scat. ert^v.H"—H" 3-18-54 to 3-19-54 HI 51 91 107 26 59.4 Run continuously. Partially an- al3^ed influent & effluent. Rust in sample K- 20 194.9 to 195.2 Blue gray fine very crse sand w/scat. gravel — Is" 3-20-54 to 3-23-54 17.4 14.2 15.4 20.4 14.6 16.6 Run continuously. Partially an- alyzed influent & effluent. Rust in sample M-18 13 225.3 to 225.6 Gray medium to very coarse sand and gravel — 1" 5- 6-54 to 5-11-54 58.0 39.0 46.3 61 41 52.7 Run intermittently. Analyzed in- fluent & effluent. Rust in sample; anthrone sugar test G-8 51 279.9 to 280.2 Gray medium to coarse sand 5-25-54 to 6- 1-54 22.6 19.0 20.6 21.4 15.4 18.0 Run intermittently. Analyzed in- fluent & effluent G— 52568 110 SEA WATER INTRUSION IN CALIFORNIA TABLE 15 LABORATORY PERMEABILITY TEST ON SAMPLE No. 51 FROM WELL G-8, DEPTH 279.9-280.2 FEET, GRAY MEDIUM TO COARSE SAND Date Time Permea- bility Ft./Day at 60° F Analyses of Inflvient and Effluent ppm Theoretical ppm' Difference ppm 2 Difference epm Chlor. Cal. Mag. Cal. Mag. Cal. Mag. Cal. Mag. Sodium 5-25 11:20 11:35 12:13 2:00 4:00 8:00 840 10:30 12:00 2:00 4:00 8:00 8:45 10:45 1:30 8:15 8:45 18640 14900 18360 18680 18680 18720 18680 18700 18760 18740 18800 18760 18880 18600 18580 18795 18760 402 665 466 397 393 426 399 393 397 396 390 394 390 402 392 409 390 1258 1012 1255 1261 1271 1272 1275 1279 1279 1287 1281 1289 1294 1263 1270 1290 1270 Start 1st Bottle-Sea water Influ ent *5-25 5-25... 22.2 22.6 22.3 + 64 —5 —9 +29 + 2 -^ Bottle-Sea —1 —7 —3 —7 + 5 —5 + 12 —7 —3 +3 + 13 —7 — 4 —3 water Influ + 8 + 2 + 10 + 15 —16 —9 + 11 —9 + 3.2 — .25 — .45 + 1.4 + .10 — .20 ent — .05 — .35 —.15 — .35 + .25 — .25 + .60 — .35 — .25 + .25 + 1.7 — .58 — .33 — .25 + .65 + .16 + .82 + 1.2 —1.3 — .74 + .91 —.74 —2.95 5-25 5-25. _ —1.25 *5-26 — .82 5-26 20.9 19.7 + .23 5-26- + .45 5-26 Start 2nd 5-26 20.5 19.8 — .60 5-26 + .19 *5-27 — .67 5-27 19.8 19.3 19.0 17.7 15.4 — .85 5-27-- - -- + 1.05 5-27 + .99 •5-28 —1.51 5-28 + 1.09 5-28- .. 8:46 9:50 9:57 10:03 10:05 10:15 10:28 10:45 11:00 11:18 11:32 11:50 12:17 12:50 1:15 1:45 2:17 2:45 3:15 3:45 80 17980 15440 11020 7400 4300 2102 1130 680 440 385 312 192 182 134 124 116 108 106 100 402 196 150 114 90 86 84 84 80 158 90 112 88 32.6 377 318 229 147 79.5 37.7 22.3 16.2 12.8 12.0 14.2 11.5 15.0 15.3 16.0 18.1 19.0 19.7 20.0 28.2 20.1 22.5 23.5 23.6 25.7 26.7 27.6 27.1 26.6 27.8 28.6 28.5 12.7 1220 1000 688 439 250 118 79 63 45 29.4 22.9 17.7 22.9 16.1 18.1 15.7 15.5 15.8 17.5 30.0 16.6 16.3 15.6 17.1 17.1 16.3 14.9 15.2 22.9 14.6 9.4 15.2 381.4 331.8 246.5 175.8 114.9 72.1 53.1 44.3 39.6 38.5 37.1 34.6 34.6 33.7 33.4 33.3 33.2 33.1 33.0 38.9 34.9 34.0 33.3 32.8 32.7 32.7 32.7 Start 3rd 1226.5 1054.3 754.3 508.7 298.8 149.7 83.9 53.4 37.1 33.3 28.4 19.8 19.7 16.4 15.7 15.1 14.6 14.5 14.1 34.5 20.6 17.4 15.0 13.3 13.1 13.0 13.0 Bottle-M —13.8 —17.5 —28.8 —35.4 —34.4 —30.8 —28.1 —26.8 —26.5 —22.9 —23.1 —19.6 —18.4 —17.4 —15.2 —14.2 —13.4 —13.0 —10.7 —14.8 —11.5 —9.8 —9.2 —7.0 —6.0 —5.1 WD Water —6.5 —54.3 —66.3 —69.7 —48.8 —31.7 — 4.9 — .4 +7.9 —3.9 —5.5 —2.1 + 3.2 — .3 + 2.4 + .6 + .9 + 1.3 +3.4 -4.5 —4.0 —1.1 + .6 + 3.8 + 4.0 +3.3 + 1.9 Influent — .22 — .69 — .87 —1.44 —1.77 —1.72 —1.54 —1.40 —1.34 —1.32 —1.14 —1.15 —.98 — .92 — .87 — .76 — .71 — .67 — .65 — .53 — .74 — .57 — .48 — .46 — .35 — .30 — .25 — .53 —5.5 —5.7 —4.0 —2.6 — .40 — .03 + .65 — .32 — .45 —.17 + .26 — .02 + .20 + .05 + .07 + .11 + .28 — .37 — .33 — .09 + .05 + .31 + .33 + .27 + .16 5-28 16.5 + .75 5-28 + 5.19 5-28 16.3 + 6.37 5-28 +7.14 5-28 5-28 18.6 + 5.77 +4.32 5-28 16.6 + 1.94 5-28 . + 1.43 5-28 16.4 16.5 + .69 5-28 + 1.64 5-28 + 1.59 5-28- + 1.32 5-28 17.1 + .72 5-28 + .94 5-28 16.7 + .67 5-28 + .71 5-28 16.5 + .64 5-28 + .56 5-28 --. 17.0 + .37 ♦5-31 . + .90 5-31 9:00 10:00 11:00 12:00 1:00 2:00 3:00 3:38 21.4 17.5 19.3 19.7 20.5 20.4 20.1 + 1.07 5-31 + .66 5-31- + .43 5-31 + .15 5-31 + .02 5-31 + .03 5-31... + .09 S-31 ♦6- 1 34.2 32.8 33.2 32.8 18.1 13.3 16.3 13.2 —7.6 —5.0 ^.5 ^.3 + 4.8 + 1.3 +3.7 +2.0 —.38 — .25 —.22 — .21 + .40 + .11 + .30 + .16 — .02 6- 1 11:00 12:45 4:00 19.2 18.8 18.3 + .14 6- 1-. — .08 6- 1 + .05 •First effluent after starting permeameter runs, represents. (1) Influence of original soil water, or, (2) water left in sample overnight or over week-end. * Calculated from dilution ratio based on chloride cimcentration. * With sea water as influent — effluent minus influent ppm; with MWD water as Influent — Theoretical minus Analytical. SEA WATER INTKUSIOX IN CALIFORNIA 111 TABLE 16 SUMMARY OF SOIL TEST DATA WeU No. Depth % Moist. Dry Density p.c.f. Speo. Grav. Sieve Analyses Porosity Void Ratio % Silt or Clay % Sand % Gravel •C-9 116.0-118.0 140.0-140.5 208.0-208.3 172.0 96.4- 96.9 162.8-163.2 169.4-169.7 197.0 156.0 157.0 236.5-237.0 236.1-236.5 156.0 150.0 158.3-158.6 169.0-171.0 177.0-181.0 204.0-209.0 211.0-215.0 218.0-224.0 225.0-230.0 230.0-234.0 248.0-250.0 138.9-139.2 134.0 155.0 201.0 136.0 134.7-135.0 136.0-138.0 149.3-149.6 183.4-183.7 210.7-211.1 259.8-260.2 305.7-306.0 8.0 10.1 10.0 22.6 36.1 12.0 8.0 6.8 15.0 15.0 12.0 31.6 21.1 23.6 21.9 14.0 115.0 120.2 131.8 99.8 So. 4 128.0 116.6 110.1 118.0 114.9 116.0 90.9 105.2 101.2 103.8 112.8 2.67 2.67 2.70 2.66 2.68 2.73 2.71 2.71 2.75 2.68 1 6 1 2 1 2 1 3 3 1 3 3 1 82 55 55 100 47 64 65 78 50 57 81 65 64 38 53 22 23 59 68 48 85 30 48 42 48 89 71 91 17 39 45 53 35 35 20 49 43 19 33 36 61 47 78 77 41 32 49 12 70 51 58 49 8 28 9 .31 .28 .22 .40 .30 .33 .30 .32 .32 .32 .45 •C-9 — - C-4 - --- .38 .28 c •C-1 .69 C-I C-l D E E-1 E-1 •E-4 - .43 .55 F F-G - •G-1 .43 G-2 G-2 . ... G-l G-« G-» 0-1 G-» G-l . . •G - .47 G-H H I J >J. .48 K K-1 K-1 K-1 K-1 •K-1 2 .48 * Slere analysis nm after permeabillts' test. 112 SEA WATER INTRUSION IN CALIFORNIA TABLE 17 FALLING-HEAD PERMEAMETER TEST AT AAANHAHAN BEACH Date Time Permeability Ft./Day at 60° F Remarks Pre-Chlorinator Water Post-Chlorinator Water 3-30-54 1335 1400 0803 1600 0810 0805 1400 1452 1110 1050 1300 1325 1125 1340 1200 1535 1330 1455 1119 1552 0815 0935 0950 0815 0810 1252 1303 1153 1339 1258 1551 1100 1045 1414 1115 1540 1257 315 277 177 165 146 129 92 55 56 44 46 47 43 42 39 49 57 58 41 41 38 30 28 26 25 23 22 21 22 22 20 20 24 25 19 21 16 317 242 132 132 119 124 86 68 87 90 84 90 86 88 90 84 90 87 93 96 91 91 89 87 83 82 83 78 84 86 86 83 81 88 80 82 72 Buff colored precipitate in "Pre" tube above sand. Slight darkening of top H " of sand in "Pre" tube. Sand in "Pre" tube darkening. Sand compaction — 5/8" Pre H' Top K" "Post" sand starting to turn brown. Sand fading in "Pre" tube. Sand getting darker in both samples. End of Teat — Sludge from lines deposited on sand. 3-30-54 3-31-54 3-31-54 4- 1-54 4- 2-54 4- 3-54 4- 4-54 4- 5-54_._ 4- 6-54. 4- 7-54 . . 4- 8-54. ._ ' Post. 4- 9-54 4-10-54 4-11-54 4-12-54 4-13-54 4-14-54 4-15-54 4-15-54. 4-16-54 4-17-54. 4-18-54 4-19-54 4-20-54. 4-21-54 4-22-54. 4-23-54 . 4-24-54 4-25-54 . 4-26-54... 4-27-54 4-28-54. 4-29-54 4-30-54 5- 1-54 5- 3-54. FIGURE I LTERED WATER iOTTLE EVACUATE specimen: Close all valves except no.6, allow specimen to evacuate completely ( approximately I hr ). SATURATE SPECIMEN: Close valve no 6, open valve no.8 to charging water until water rises to valve no. 7 8 specimen is at atmospheric pressure. BEGIN test: Close valve no.8, open all others except no.6. LOS FLOOD ANGELES CONTROL COUNTY DISTRICT WEST BASIN BARRIER TEST CONSTANT HEAD PERMEAMETER PIEZOMETER BOARD OUTFLOW HEAD CONSTANT HEAD OVERFLOW TO VACUUM RUBBER STOPPER OTTAWA SAND LUCITE CYLINDER OTTAWA SAND RUBBER STOPPER EFFLUENT WATER GRADUATE [figure T FILTERED WATER BOTTLE TO EVACUATE SPECIMEN: Close all valves except no.6, allow specimen to evacuate completely (approximately I nr ). TO SATURATE SPECIMEN: Close valve no 6, open volve no.8 to charging water until water rises to volve no. 7 a specimen is ot atmosptieric pressure. TO BEGIN test: Close valve no.8, open all others except no.6. LOS FLOOD ANGELES CONTROL COUNTY DISTRICT WEST BASIN BARRIER TEST CONSTANT HEAD PERMEAMETER [FIGURE 2 LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST PRE- RECHARGE CONDITIONS LEGEND t- PROJECT RECHARGE WELLS PROJECT OBSERVATION WELLS PRIVATELY OWNED WELLS CITY OF MANHATTAN BEACH WELLS STANDARD OIL COMPANY WELLS CALIFORNIA WATER SERVICE CO. LAC FC.O TEST HOLES GENERAL CHEMICAL CO. WELLS - PIPE LINE ZONE BOUNDARIES - ISOCHLOR-RATIO LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST PRE- RECHARGE CONDITIONS I Fl G UR E 4 05 JA N 1953 MAR APR MAY JUN JUL AUG SEPT OCT DEC JAN 1854 OCT LEGEND H ^ h» RECHARGE ■)GH OBSERVATION ^G RECHARGE No a K Co Mo LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST WATER QUALITY WELLGH EFFECT OF RECHARGE FIGURE 5 ® s LEG END SEAWATER RECHARGE WATER SAMPLES FROM WELL GH See Table 2A LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST EFFECT OF RECHARGE ON WELL GH —I — I — I — I — I — I — I — ( — I — I — I — I — I — I — 1 — I — I — I — h 10 20 30 40 90 SO 70 SO 90 KX> 90 80 70 SO 50 40 30 20 10 CATION ANION 10 20 30 40 90 SO 70 SO 90 100 90 80 70 60 50 40 30 CATION ANION 10 20 30 40 50 60 70 80 90 100 90 80 70 SO 50 40 30 20 10 CATION ANION P E R C N T PRE- RECHARGE WATER LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST EFFECT OF RECHARGE G WELLS UJ o 2 < X X' UJ S5 z o 1000 FIGURE 7 LEGEND N \ \ \ 2000" i k 235' -285- -660- © 7 G-2 G-4 G-8 800 400 300 200 100 CONCENTRATION E.P.M. LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST EFFECT OF RECHARGE G LINE WELLS FIGURE 8 %epnri CATIONS % e p.m, ANIONS LEGEND P PRE-RECHARGE WATER N NATIVE WATER R RECHARGE WATER L LATEST ANALYSES, 9/154 LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST SOURCE OF DILUTION WELLS K-8 a K-12 [FIGURE 9 ^°° < < tlJ o S ^ tr I- a. z < X o » X a: [-5 -10 20,000 15,0 00 CO I UJ Q — s a: a: o ""^ 5,000 I o ( l-.,^ ^ t t 1 1 1 j. 1 ■^ t "^ t • —~*ii — — ®— -— . -- V^ 1 y V* ®— ~*"~~^' 1 1 I, aK yi (ci » — ^^ s y^~^ ^ stri rt^^)A^<^>il3;M> r— — 'VTii_i_^L-?^_.^.^> — r-*^ H' t-J*- h^ i-= f—^ ? f-Hrl mT ^ f 1 1 -^ 1 ' — ~"* — »--< 1 . • a « t • r M V» W a t 1 , 1 1 1 i_ I ti 12 25 8 12 4 MAT 26 12 27 MAY 12 8 12 JUNE LEGE N D 1 1 NFLU ENT OFF t INFLUENT ON CALCIUM o MAGNESIUM A SODIUM LOS ANGELES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST PERMEABILITY a EFFLUENT ANALYSES OF SAMPLE 51 WELL G-8 DEPTH 279.9-280.2 FIGURE 10 300 200 a a> < 100 LEG END POST CHLORINATOR WATER __V__y_V PRE - CHLORINATOR WATER A A / \ TIME IN DAYS LOS ANG E LES COUNTY FLOOD CONTROL DISTRICT WEST BASIN BARRIER TEST PERCOLATION TESTS PERMEABILITY RESULTS LOS ANGELES COUNTY FLOOD CONTROL DISTRICT TESTING DIVISION APPENDIX C SPECIAL TESTS ON CLAY CAP MATERIALS RELATIVE TO INVESTIGATIONAL WORK FOR PREVENTION AND CONTROL OF SEA WATER INTRUSION December 10, 1954 By IRVING SHERMAN CHARLES GREEN Chemical Section Head Submitted by: F. O. FRICKER, Division Head Recommended by: Approved by: PAUL BAUMANN H. E. HEDGER Assistant Chief Engineer Chief Engineer 9—52568 SPECIAL TESTS ON CLAY CAP MATERIALS RELATIVE TO INVESTIGATIONAL WORK FOR PREVENTION AND CONTROL OF SEA WATER INTRUSION SUMMARY This Appendix is a report describing several tests undertaken bj^ the Testing Division in investigating the possible causes for the observed instances of de- struction of the clay cap, in the West Coast Basin Bar- rier Project, which is a field test of ground-water recharge by wells. The possible causes were considered in two groups, as follows : 1. Chemical causes, due to the nature of the re- charge water used. (a) A change in the plasticity of the clay cap, resulting in flow of the clay by failure in the plastic state. (b) An increase in the erodability of the clay cap, due to dispersion, so that dispersed particles are carried through voids in the surrounding granular materials. 2. Physical causes, arising from the type of drilling operations used, and the methods of development of the wells. (a) Erosion of the clay cap by lateral flow from the well. (b) Erosion of the clay cap by vertical perco- lation. (c) Erosion of the clay cap by flow through voids adjacent to the well easing. (d) Physical collapse of the clay cap into adja- cent large voids due to gravity alone. (e) Collapse of the clay cap due to sudden pres- sure changes associated with well operation. The tests made involved six different waters of varjring chemical composition. The chemical causes were investigated by comparative tests of settling ve- locity, plasticity, and permeability of the clay with each water. These tests indicated the chemical effects to be relatively unimportant. The ph.vsical causes were investigated by means of hydraulic models of the recharge wells, using "undis- turbed" samples of the clay cap. By controlling the conditions of these models, it was found possible to reproduce the physical failures listed under 2(a), (c), and (e) above, but only the latter two were considered serious. Recommendations are made for drilling any future wells so as not to create large voids adjacent to the casing, and for sealing the clay cap to the well casing by grouting. Infroduciion The West Coast Basin Barrier Project is essentially a field trial of a proposed method of halting the intru- sion of sea water into the fresh water aquifer along the coast. The method consists of creating a barrier or ridge of fresh water, by pumping (recharging) fresh water into the aquifer through a line of recharge wells approximately parallel to the sea coast. The success of the operation depends in part on the existence of a relatively impermeable stratum, gener- ally called the clay cap, wliieh is seven to twenty feet thick. This material lies immediately above the aqui- fer. In the vicinity of the recharge wells it is seven to twenty feet thick and lies at a depth of eighty to one hundred feet. This clay cap, being quite impermeable prevents upward movement of the recharged water and thus forces it to spread laterally and form a continuous body rather than a series of disconnected water "mounds" at the wells. Shortly after the start of the recharge operations, it was found that the clay cap had been greatly dis- turbed in the vicinity of two of the recharge wells. This disturbance was evidenced by two phenomena: (1) the appearance of recharge water in the normally dry dune sands lying above the clay cap, and (2) subsidence of the ground surface around the well. The Chemical Section of the Testing Division was asked to investigate the possible causes of destruction of the clay cap, and if possible, to recommend methods of restoring some impermeable continuous layer in the clay cap region. This report summarizes the investiga- tions and results thereof. Hypotheses of the Destruction Mechanisms Several possible causes for the clay cap failure have been proposed. Generally speaking, they can be divided into two groups: (1) Chemical causes, due to the chemical composition of the recharge water, and (2) physical causes, related to the hydraulic conditions which exist at the wells. A more detailed listing and description follows: (115) 116 SEA WATER INTRUSION IN CALIFORNIA 1. Chemical Causes The recharge water is obtained from the Met- ropolitan Water District. Being zeolite-treated and chemically softened water from the Colorado River, it is high in both total dissolved solids and percent of sodium. It is known that sodium tends to disperse clays and change their physical prop- erties. Two such changes considered were : (a) A change in the plasticity of the clay cap resulting in an increased tendency for the material to flow in the direction of the hydraulic gradient by failure of the mass in the plastic state. (b) An increase in the erodability of the clay cap due to the dispersion by sodium ions. It has been shown that dispersed material, acting as individual particles rather than aggregations of particles, may be more easily eroded by flowing water. This hy- l)otliesis visualized dispersed particles be- ing carried through voids in the dune sand above or the "Silverado zone" aquifer below, though the voids were too small to permit pas-sage of particle aggregations. 2. Physical Causes The well holes were drilled by a cable tool rig rather than a rotarj' drill rig. It is known that the holes thus created were somewhat larger in diameter with irregular voids outside of the casing because of the type of action of the drill tool. Prior to the observed destruction of the clay cap, no measures were taken to seal any gaps which may have existed at the clay cap. Necessary pumping to develop the wells and remove fines from the aquifer immediately adja- cent to the well created some voids, of unknown size and distribution. Unavoidable interruptions of the pumping or recharge operations resulted in rapid changes of pressure conditions within the system. Hypotheses of primarily physical destruction included : (a) Erosion of clay cap material by water flowing horizontally outwards from the well along the bottom of the clay cap. (b) Erosion by water flowing upwards through the clay cap mass under the un- naturally high hydraulic gradient pro- duced by high rates of recharge. (c) Erosion by water flowing upwards through the voids in the clay cap adjacent to the well casing, removing particles from the exposed clay surface. (d) (e) Physical collapse due to the action of gravity alone, of the clay cap into adja- cent large voids in the aquifer, which were created during development of the wells. Physical collapse of the clay cap into adjacent large voids due to sudden pres- sure changes resulting from rapid fluctu- ation in rates of recharge, or shut-down. It is here emphasized that the existence of any one condition tending to cause failure does not rule out the existence of other conditions. Two or more may conceivably occur simultaneously. Tests and Investigations The first tests performed were intended to test the chemical hypotheses. To a large degree the tests were improvised, as previous standards for such procedures were largely lacking. The chemical tests involved the use of several dif- ferent waters, as follows: (1) Distilled water (2) MWD treated water (used in the recharge op- erations at the West Basin) (3) MWD water before treatment — a high calcium water, high total dissolved solids (4) San Gabriel River water — low in total dis- solved solids, but high in percentage of cal- cium (5) Sea water — very high total dissolved solids, high percent sodium (6) MWD treated water saturated with calcium sulfate so as to make it a high calcium water again Data on these waters is presented in Table I. TABLE I CHEMICAL COMPOSITION OF TEST WATERS Water Total Dis- solved SoHds, epm Total Hard- ness, epm Na&K epm % Ca epm % Na+K Ca MWD treated (at WeU G).. MWD untreated 674 636 297 36,360 2,783 134.3 301.0 174.0 6,539.9 1,554.9 37.12 20.95 8.79 38.88 9.68 9.48 18.87 30.76 1.72 39.19 3.91 1.11 San Gabriel River 0.29 Sea Water 22.6 MWD treated Ca S0» Saturated- - 0.25 A number of disturbed samples of clay cap ma- terial were available for the chemical tests. All were classified by means of the Atterberg tests, (ASTM Designations D-423-39 and D-424-39) and found to fall into three general classes according to the BPR system — A4, A6, and A7. Samples belonging to each class were then mixed together to give three com- posite samples which were used for further tests. The Atterberg tests were then performed on com- posite sample A7, using distiUed water and also the SEA WATER INTRUSION IN CALIFORNIA FIGURE I EFFECTS OF WATER ON PLASTICITY 117 300 100 - - X UJ o 5 o 30-- 10 PURE US.P BENTONITE WEST BASIN CLAY CAP COMPOSITE A-7 MATERIAL CALCIUM SULFATE SATURATED (MWD TREATED) HIGH- CALCIUM WATERS / MWD O-^UNTREATEO DISTILLED CALCIUM SULFATE" 1 / SATURATED (MWD^q MWD TREATED / TREATED) i ^ // MWD UNTREATED SEA WATER O /SAN O-^GABRIEL O RIVER A MWD TREATED LOW-CALCIUM WATERS DISTILLED SEA WATER SAN GABRIEL RIVER I 10 40 100 400 LIQUID LIMIT- MOISTURE CONTENT IN PERCENT OF DRY WEIGHT AT 25 BLOWS waters listed in Table I. The purpose was to investi- gate possible changes in plasticity due to water com- position. The tests were made onlj' on the A7 sample because it contained a higher percentage of clay than the other two, and thus would be more sensitive to chemical effects. Test results are presented in Table II and Fig. I. Details of the test procedure are given in the sup- plement following this report. Pure bentonite was also tested, because it is known to be sensitive to chemical changes and thus was used as a standard. The "Flow Index" shown in the fourth column of Table II measures the slope of the Liquid Limit curve, and is defined as the change in moisture con- tent required to change the number of blows in the Liquid Limit Test from ten to one hundred. A low flow index indicates a rapid increase in fluidity with increasing moisture content; a high flow index indi- cates a slow increase. TABLE 11 EFFECTS OF WATER ON ATTERBERG LIMITS Water Li- quid Limit Plas- tic Limit Plas- ticity Index Flow Index (a) West Basin A7 Clay Cap Distilled .. 52.0 49.8 49.6 50.7 49.2 48.8 25.9 27.3 27.4 27.0 25.3 25.2 26.1 22.5 22.2 23.7 23.9 23.6 12.2 MWD treated _ 17.2 MWD untreated . - 12.2 11.2 Sea Water . . 10.8 MWD treated. Ca S0» Saturated 19.8 Mean _ 50.0 3.2 39.3 341 332 351 129 283 26.4 2.2 48 45 46 45 48 46 23.7 3.9 345 296 286 306 81 237 13.9 9.0 (b) Bentonite, USP Distilled 83 MWD treated _ 59 MWD untreated 139 65 Sea Water 15.4 MWD treated. Ca SO4 Saturated 72 305 264 46.3 3 258 264 72 Range. _ 123.6 118 SEA WATER INTRUSION IN CALIFORNIA It •will be seen from this data that the bentonite is far more sensitive than the A7 material. For ex- ample, using Ca SO4 saturated water instead of dis- tilled water changes the liquid limit of bentonite by 28%, as compared with only 6.2% for the A7 ma- terial. This is to be expected from the lower clay con- tent of the A7 material (see mechanical analysis results below) as contrasted with the 100% clay content of the bentonite. However, on the absolute scale, the plasticity effects on the A7 material are small. It is therefore concluded that the "plastic flow" hypothesis is not supported by the evidence. Experimental hydrometer analyses were also made for three purposes: (a) To test the sensitivity of the clay to chemical treatment when in the form of a clay suspen- sion rather than a paste.^ (b) To evaluate the effect of each water, and (c) To obtain data which might indicate the rela- tive erosion characteristics. The general plan of this phase was taken from previous work by the writer (See Sherman, I., "Flocculent Structure of Sediment Suspended in Lake Mead," Transactions, Amer. Geophys. Union, Vol. 34, No. 3, pp. 394-406, June 1953). Again the A7 clay was used. Duplicate hydrometer analyses were made on the A7 composite samples using the following waters in the hydrometer jars : (1) Distilled water only (2) Distilled water with sodium silicate dispersing agent, (Standard laboratory procedure). (3) Distilled water with sodium hexa-metaphos- phate dispersing agent. (4) MWD treated water as received at Well G (5) MWD untreated water (6) San Gabriel River water (7) Sea water Details of the hydrometer technique are given in the supplement following this report. The test results are presented in Figures 2 and 3 and Table III : TABLE III EFFECT OF WATER ON CLAY BEHAVIOR Water Mean Settling Velocity, cm/sec Mean Floccule Density gm/cm* Mean Floccule Diameter mm MWD treated .0205 .0228 .0254 .0239 1.49 1.43 1.48 1.49 .0277 .0312 San Gabriel River .0312 Sea Water. .- .0319 1 The sensitivities may not be the same under both conditions. See R. E. Grim, in "Symposium on Exchange Phenomena in Soils," ASTM Spec. Pub. No. 142, 1953, p. 7. Floccule data are presented only on the natural waters. It was found that in all of the distilled-water treatments the samples were essentially dispersed, so that floccules did not exist to any great extent. The data on all treatments are presented graphically in Figures 2 and 3. The hydrometer results are consistent with chemical theory and lead to the following conclusions : (1) The clay cap material is sensitive to chemical dispersion treatments, and appears to be prob- ably a sodium clay in the original state. (2) Waters with low Na/Ca ratios and/or high total dissolved solids tend to flocculate the clay. (3) The quantitative differences between the di- ameters of floccules produced by the different waters are probably not large enough to have any significant effect on the phenomena at the wells. Computations were made of the pore diameters in the aquifer from the mechanical analyses of aquifer samples, as compared to the floccule diameters in the hydrometer test. It was found that even the largest floccules formed could individually pass through the aquifer pores, which naturally were of the order of several times the floccule diameters, even without the presence of large voids caused by well development. This indicates that further flocculating the claj- by a change ia water composition by calcium sulfate treat- ment would not have a large beneficial effect on pre- serving the clay cap. A third set of tests involved percolating each of the different waters through a specimen of claj^ cap ma- terial, to measure resulting differences in permeability and rates of erosion. The apparatus was patterned after that of J. E. Christiansen ("Some Permeability Characteristics of Saline and Alkali Soils," Agrie. Eng. Vol. 28, No. 4, pp. 147-1.50, April, 1947). How- ever, in order to get appreciable rates of flow it was necessary to "dilute" the clay cap material with Ottawa sand. The final mixture consisted of 15% of A7 clay cap composite sample, 85% of Ottawa sand, compacted by a modified miniature Proctor method between layers of pure Ottawa sand. In order of de- creasing rates of clay erosion, the different waters are : (1) Distilled (2) MWD treated as received at Well G (3) Ca SO4— Saturated MWD treated water (4) MWD untreated water (5) San Gabriel River water (No erosion) (6) Sea Water (No erosion) The eoeffieient of permeability was the same with all waters except the Ca SO4 saturated water, which appeared to flocculate the clay extremely and which produced a higher permeability. The increased volume of flow resulted in appreciable erosion, as contrasted with other flocculating waters. SEA WATER INTRUSION IN CALIFORNIA 119 aaNivxau XKasuad rr 1 iLl CO tr 1 rr < lO m in z CO i/i ^ in m a 1- o a (/) iij « t« >: g TE KE TE TE CATIO ea •5 < < a s oxissvd XKaoaaa 120 SEA WATER INTRUSION IN CALIFORNIA fO in I- cr yj tr cr < UJ CD (r U- CO < m en UJ o o o O o o O o cr> CD r-- lO in 'J- d3/wons iN30d3d 3AiivnnwnD I SEA WATER INTRUSION IN CALIFORNIA 121 Althon^'h the results of the Atterberf,', Hydrometer and Pereohition Tests are iiulieative, it is hazardous to exteud the results to the "West Basin problem be- cause the tests mentioned above do not actually dupli- cate field conditions. So a fourth set of tests was made in which ajiproximate hydraulic models of a "West Basin recharge well were subjected to varyin-^ condi- tions. A drawing of the apparatus is piven in Fig. 4. Four models were set up. The essential ditTerences were in the nature of the ' ' clay cap ' ' specimens which were as follows: Model No. 1— "Well G, Sample #2G, depth 101.0'- r 101.3', field description : very fine B sandy silt. Specimen placed with es- sentiallv no disturbance. Model No. 2— Well G, Sample #20, depth 92.2'- 92.5', field description : clayey silt. ♦ Specimen seriously disturbed at pe- riphery durinji placement. (G-20-0.) Model No. 3 — Same as No. 2, but specimen less seri- ously disturbed. (G-20-1.) Model No. 4 — Same as Nos. 2 and 3, but specimen al- lowed to swell by access to distilled water before placement. Specimen placed with no obvious disturbance. Table TV summarizes some of the soil data for the hvdraulic models: Model No. 1 was abandoned as being too permeable ; No. 2 was abandoned because of the excessive periph- eral disturbance. The peripheral voids in the specimen in Model No. 3 enlarged progressively until a continuous channel througli the specimen had been created. At that time the flow of water into the "Well" was shut off, thus reversing the direction of the hydraulic gradient. This caused the "clay cap" to collapse and sand from the overlying stratum fell into the voids thus created. No large peripheral voids existed in Model No. 4. Some small voids originally present persisted without enlargement. The permeability increased, but surge tests failed to collapse or visibly damage the specimen. Six such tests were performed. During the tests of the model, "clay cap" material in measurable quantities was deposited in the overflow beaker, and more "clay cap" material was subsequently found in the "Sil- verado Zone" after disassembly of the model. The only possible source for this material seemed to be the lower surface of the "clay cap." The mechanism of removal could have been erosion by lateral flow. The process of erosion was not directly observable however. TABLE V Model No'a. 12 3 4 101.0-101.3 92.2-92.5 Very fine Sandy Silt 92.4 44.3% 92.4 High but not measured 1 Not tried Clayey Silt 98.1 26.9% Dry Density in experiment, p.c.f 98.1 2.8X10-' 1 Not tried 98.1 1.8X10-' 1.4X10' 13 Spec, failure 93.4 5.2X10' (Initial) 3.0X10-' (Final) 32 Result of surge teota No failure 122 SEA WATER INTRUSION IN CALIFORNIA SEA WATER INTRUSION IN CALIFORNIA The following is the material balance for the destroyed "clay cap" specimens of Models Nos. 3 and 4: TABLE V CLAY CAP MATERIAL BALANCE 123 Model No. 3 Model No. 4 Before destruction 29.23 46.0 5 43.7 29.25 43.7 1 43.2 Percent Grams Percent Grams After destruction or dis-assembly 81.8 4.1 7.8 5.3 35.72 1.80 3.41 2.32 91.9 0.2 3.0 1.1 39.68 0.10 1.29 0.48 99.0 1.0 42.25 0.4S 96.2 0.4 41.55 Estimated weicht which passed through overflow beakers, based on sieve ana- 0.20 Kiftimn*^ t/itjal 100 43.70 96.6 41.75 Conclusions and Recommendafions 1. The quality or chemical composition of the water used probably has relatively little effect on what has happened at the West Basin. 2. Any deleterious effect of MWD-treated water would probably not be reversed by using Ca SO4 — saturated water. 3. If the clay cap is not disturbed by large un- sealed voids, it will not be destroyed by well opera- tion. But this excludes large void spaces around the easing such as are produced by cable drilling tools. 4. Any future wells should preferably be drilled by equipment which does not create the disturbances, impacts, and large voids produced by the cable tool, and the clay cap sealed to the casing by grout. 5. There may be a certain amount of erosion of the lower surface of the clay cap due to horizontal flow through the aquifer during recharge. The model tests indicate that such erosion will not be serious if there are no large voids in the clay cap where turbu- lence could become a factor. SUPPLEMENT DESCRIPTION AND DISCUSSION OF TECHNIQUES I. Hydrometer Analysis The standard procedure used at the Testing Divi- sion was basically that of AST:\r Designation D-422-39 (see "Procedures for Testing Soils" 1950). Procedures used in this experiment follow that standard method except as noted below. The paragraph numbers fol- low the numbering scheme of Method D-422-39. ASTM methods were followed where paragraph num- bers are omitted. The notation (FCTDS) indicates a procedure which was standard at the Testing Division. 2. Apparatus (f) Sieves. Those used were as follows: Tyler No. 14 (1.168 mm) U.S. No. 30 (0.59 mm) Tyler No. 40 (0.38 mm) Tyler No. 48 (0.295 mm) Tjder No. 100 (0.147 mm) Tyler No. 200 (0.074 mm) (FCTDS) 3. Sample All portions for different treatments were ob- tained by use of a splitter for small samples. The weight of each portion was intended to be approximately 40 grams air-dry, in order to con- serve the small amount of sample available. Actual weights of the portions varied between 37.13 gm and 45.70 gm. One portion was oven-dried and then weighed. (FCTDS) All portions were taken from material passing the Tyler No. 14 sieve. 4. Hygroscopic Moisture The determination was made in duplicate on portions of approximately 14 grams each, ob- tained by splitting. 5. Dispersion of Soil Sample All water was filtered. The seven treatments used were as follows : (a) ASTM Standard with sodium silicate. This was used on the portion which had been oven-dried. 124 SEA WATER INTRUSION IN CALIFORNIA (b) 50 ec of 5% Calgon solution (essentially sodium metaphosphate, buffered to pH 9 with sodium carbonate) instead of sodium silicate; added at the start of the soaking period instead of at the end. This is essen- tially the new ASTM specification not yet published. (c) No dispersing agent. Distilled water. (d) No dispersing agent. Treated MWU water as received at Well G used instead of distilled water. (e) No dispersing agent. Untreated MWD water used instead of distilled water. (f) No dispersing agent. San Gabriel River water (from the reservoir behind San Gabriel Dam) used instead of distilled water. 6. Hydrometer Test — time of readings (a) For the portion treated with sodium sili- cate, hydrometer readings were taken at the following times: 4, 1, 1^, 2, 3, 4, 5, 10, 15, 30, 60, 240, and 1440 minutes. Be- cause of the close spacing of the first readings it was not possible to remove the hj'drometer after each reading until after the five minute reading. (FCTDS). (b) For all other portions, hydrometer read- ings were taken at the following times: 1, 4, 16, 64, 256, and 1260 minutes. The main purpose of taking fewer readings was to save time and speed the work. (c) The portion treated with Calgon, because it settled most slowly, was also read at 93 hours in an effort to approach the "ten percent finer" point. (d) The portions treated wtih sea water, San Gabriel River water and distilled water only, were also read at 8 minutes. 7. Sieve Analysis After the last hydrometer reading, each portion was treated as follows: (a) The portions treated with San Gabriel River and sea water were washed on the No. 40 sieve, using their respective waters. This was done with the intention of col- lecting the material passing the No. 40 sieve for Atterberg tests. The material retained was then oven-dried and weighed. (b) All other portions were washed on the No. 200 sieve, oven-dried, and weighed. They were then dry-sieved tlirough the sieves listed in paragraph 2 (f) above. (FCTDS except for the Tyler No. 40). The Tyler No. 40 sieve was included for comparison with the San Gabriel River and sea water portions. Calculations 10. Percentage of Soil in Suspension (a) Temperature corrections were all ob- tained from a graph similar to Fig. 6a of D-422-39, but reading in degrees centi- grade. (FCTDS). (c) The percentage of soil in suspension rep- resented by a hydrometer reading was calculated from the formula: W% = (^h + Cm + M. + C.)100 ^^^^^^^ W g in which We = original weight of sample, oven-dry basis, grams W% = percentage of soil in suspension Eh = hydrometer reading at the top of the meniscus, gm/1 Cm = meniscus correction il/j. = temperature correction Cn = correction for the density of the sus- pension fluid, as compared with dis- tilled water. The values of Cp obtained for the treatments, in terms of Type A hydrometer readings, were (a) Standard sodium silicate : (b) Calgon: (e) Distilled water only: (d) MWD treated water (Well G) : Cd= 1.0 gm/1 (e) MWD untreated water: Cd= 1.3 gm/1 (f) San Gabriel River water : Cd = 0.1 gm/1 (g) Sea water : Co = 40.4 gm/1 11. Diameter of Soil Particles in Suspension (a) The maximum diameter of particles in suspension, corresponding to the percentage indicated by a given hydrometer reading, was calculated from Stokes Law ^ as given in ASTM method D-422-39. The values in- serted in the equation were as follows: n = coefficient of viscosity of the suspending medium, taken from the nomogram as a function of temperature (FCTDS) except for sea water ; the value of i] for sea water was the nomogram value multiplied by 1.08, which is the vis- cosity of sea water in centipoises at 20° C. L = distance in centimeters through which soil particles settle. Taken as the effec- tive depth of the hydrometer at the reading in question. Effective depths Cd= 0.9 gm/1 Cd= 3.9 gm/1 Cd= 0.0 '' stokes Law '-J' 30 L n 980 (G-GO T SEA WATER INTRUSION IN CALIFORNIA 125 were determined by ealibratiiiir the hydrometer aecordiii": to a modification of the method proposed by Edward E. Bauer (see p. 148, "Procedures for Testing Soils," 1950). The modifica- tion consisted of actually deterininiurr the center of volume of the hydrometer bulb, which was asymmetrical, instead of assuming it to be at the midpoint of the length of the bulb, (see p. 79, "Pro- cedures for Testing Soils," 19.jO). G = specific gravity of soil particles. As- sumed as 2.70. The actual value was not determined because of lack of sample. Gi := specific gravity of the suspending liquid. Assumed as l.O on the nomo- gram (FCTDS). Corrected to 1.025 for the sea water treatment. T = settling time in minutes. D = particle diameter in millimeters. It should be noted that the assumption of 2.70 as the value of G was made only for the three portions treated with distilled water, wliich appeared dispersed or nearly so in suspension. The four portions treated with "natural" waters were obviously flocculated, and it has been shown that the effective density of soil floccules in suspen- sion is less than the specific gravity of the soil itself. The effective density is unknown except as an average value for the entire flocculated sample, as explained below. For that reason, diameters of soil floccules in suspension in the natural waters were cal- culated only for mean floccule diameters corresponding to the "50% slower" point on the curve relating settling velocity to percent of soil in suspension. 13. Plotting For purposes of comparison between treat- ments, the results of all seven treatments were plotted as cumulative percent of soil in suspen- sion versus the settling velocity (cm/sec = L/60T), uncorrected for temperature, viscocity, or specific gravity of the suspending liquid. The original plot was made on logarithmic probability paper, which has the effect of tend- ing to straighten out the usual S-shaped cumu- lative curve and facilitate drawing a smooth curve. The curves for "natural" water treat- ments would have been other-wise almost im- possible to draw with any degree of accuracy or reproducibility. The curves were then carefully transferred to an ordinary semi-logarithmic graph by noting their intersections with each percentage line. Tlie results for the three treatments in dis- tilled water were also plotted on semi-logarith- mic graph paper in the usiml manner, showing particle diameter vs. cumulative percent finer. The Calgon results were plotted on log prob- al)ility jiaitcr and extrapolated to the 10% finer point. 14. Report The results for the three distilled water treat- ments were reported as follows (FCTDS) : (a) percent clay = percentage by weight of particles finer than .005 mm (b) effective diameter = Dio (diameter cor- responding to 10% finer) (c) uniformity coefficient = D«o/-Dio (d) texture classification — taken from a tex- tiire triangle based on pereents of clay, silt (.005 mm to .05 mm) and sand ( > .05 mm) The results for the four treatments with "natu- ral ' ' waters were reported as follows : (a) iledian settling velocity (b) Mean floeeule density in suspension (c) Mean floccule diameter, mm //. Calculations for Flocculated Sediments in "Natural" Waters This entire experiment, using different waters for the hydrometer test, was patterned after some similar work previously done by the writer (see Sherman, I — "Flocculent Structure of Sediment Suspended in Lake Mead," Transactions, American Geophysical Union, June, 1953). The methods of calculation here are taken directly from that work. Tlie paper referred to shows that when sediment settles in the flocculated state, the floccules are both porous and impermeable. Therefore, the effective den- sity of the sediment floccules in suspension depends on the floccule porosity. The porosity of anj' individual floccule is unlvnown, but the average porosity of the floccules comprising a sample can be calculated from the porosity of the deposit formed by the floccules as they settle out. The technique used in this experiment involved a careful measurement of the volume of the deposit at the bottom of the hydrometer jar after 24 hours. This volume included, in the case of the four flocculated portions, the entire weight of sample originally placed in the jar. The equations for the calculations which are made are given below : (1) e„=TF,/F, (2) e, = 1 — Oa/Qr (3) 0,= {Qa + et — ei) /{1 — ei) in which Qa is the specific weight (or apparent density) of the deposit in the hydrometer jar gm/cm* 126 SEA WATER INTRUSION IN CALIFORNIA Ws is the weight of soil in the deposit, gms Vs is the volnme of the deposit, cm^ Ct is the decimal porosity of the deposit Qr is the specific gravity of the soil (assumed as 2.70 in this experiment) Qf is the mean effective density of the flocenles in suspension, gm/cm^ ea is the decimal porosity around the floceules (as contrasted to within the floceules) in the deposit in the hydrometer jar. Cd was assumed as 0.400 for all deposits. Those interested in the assumptions and derivations of the above are referred to the paper previously mentioned. The value of Q/ is then inserted in the equation of Stokes Law instead of the value of G (see paragraph 11, under "Hydrometer Analyses" above) in conjunc- tion with the median settling velocity (50% slower point). This gives a mean floccule diameter for the entire sample. ///. Atterberg Limiis A. Liquid Limit of Soils (ASTM Designation D- 423-39) This test was performed as specified in the ASTM method except as listed below. The para- graph numbers follow those in the ASTM speci- fications. 7. Procedure The mechanical liquid limit device was used throughout. (a) The water used in the determination was was varied, a different water being used with each representative portion. The waters used were : (1) Distilled (2) MWD before treatment (3) MWD after treatment, as deliv- ered at Well G (4) MWD after treatment, saturated with Ca SO4 (5) San Gabriel River water (6) Sea water (b) In order to ensure equilibrium in any chemical reaction between soil and water, an excess of the water was added so as to produce a highly saturated paste, which was then allowed to stand over- night. The resulting mixture was above the liquid limit. The moisture content was reduced during the test by allowing evaporation to occur until determina- tions both above and below the li(|uid limit had been made. 8. Preparation of Flow Curve In such cases where considerable "scatter" of points made the drawing of the flow curve unduly subject to human error or personal judgment, the flow curve was calculated as a semi-logarithmic least-squares line of regres- sion by the usual statistical procedure. (B) Plastic Limit and Plasticity Index of Soils (ASTM Designation D-424-39) Except as noted below, the ASTM procedure was used throughout: 3. Sample The sample consisted of a portion of the material remaining from the liquid limit test described above. 4. Procedure The water used in the determina- tion was the same as was used in the previous liquid limit determination. IV. Analysis of Clay Cap Disiniegration in Hydraulic Models The model was dis-assembled after the test and the contents divided into f oiir portions : (1) The "dune sand" region, including any clay cap material eroded upwards. (2) The "clay cap" region, including any "dune sand ' ' which had fallen into voids created dur- ing the test. (3) The "Silverado" region including any clay cap material eroded into it. (4) The solids carried through the "Silverado Zone" into the overflow beaker. Each portion was dried and weighed, then wet- screened on the No. 325 screen, re-weighed, and dry- screened through a nest of sieves. It was found that all of the clay cap material passed the Tyler No. 40 sieve, and all of the Ottawa sand was retained on the U. S. No. 50 sieve, with only a small fraction of one percent in the overlap range. This made it possible to separate out the two materials in each portion, and calculate the weight of clay cap material in each. The weights of material in the overflow beakers were very small, and the samples would have been "lost" on the usual 8" diameter sieves. Those very small samples were sieved on 3" diameter sieves made available through the courtesy of the Arcadia Soil Laboratory of the U. S. Forest Service. For comparison with the portions listed above a standard sieve analysis was made on an undisturbed portion of the clay cap which had not been placed in a hydraulic model. Comparison of the sieve analysis results of the undisturbed sample and of the portion retained in the overflow beakers made possible the computation of the weight of silt and clay washed out of the overflow beakers and thus not directly meas- urable. LOS ANGELES COUNTY FLOOD CONTROL DISTRICT HYDRAULIC DIVISION APPENDIX D LETTER REPORTS ON GROWTH OF MICROORGANISMS IN THE AQUIFER By DR. CARL WILSON SEA WATER INTRUSION IN CALIFORNIA 129 Juue ID, VJ'S^ June 29, 1953 Mb. H. E. IIedger, Chief Engineer, Mr. H. E. Hedger, Chief Engineer, Los Angeles County Flood Control District, Los Angeles County Flood Control District, Box 2418 Terminal Annex, Box 2418 Terminal Annex, Los Angeles 54, California. Los Angeles 54, California. Attention: Mr. Finley B. Laverty Attention: Mr. F. B. Laverty Dear Sir : Dear Sir : Fearing that lioavy ehlorination of water which Under date of June 19th 1953 you sent us six you are injeeting into the underground basin at Man- samples of water for examination in the hope of af- hattau Beaeh might be increasing the possibilities of foi-ding additional light upon the corrosive properties corrosion of pumps and well casings, you sent me of Colorado River water when heavily chlorinated for under date of June 15th two samples of the water in injection into the underground. This work was question for determination of the increase in hydrogen promptly completed and the results were telephoned ion concentration caused by the present chlorine dos- to ]\lr. Zielbauer. This memorandum will confirm that age of 15 parts per million. That work was completed report and provide copies for your files. All samples the day the samples were received and the results were designated : "West Basin B Test." Our findings were reported by telephone to I\lr. Zielbauer. were as follows: Methyl The first sample was taken from the Metropolitan Sample Phenol Orange Carbon District inlet line, and represented water before chlo- ^o- P^ "^^ «'*' dioxide rination. Its pll was found to be 8.30. \ :::::::::": 7;^ ^J 1^ 6 The second sample, taken five minutes later from ^ IIIIIIIIIIII tIto 135 ^6 the office tap. represented water after chlorination at 5 7.6O 130 7 the dosage of fifteen parts per million : its pll was 6 7.70 13.5 6 found to be (.55. Identification of the samples is as follows: _, „ , ,. ., i i J! ii J. No. 1, Metropolitan Water District water prior to chlorination The free carbon dioxide content of these waters, a taken ^j ogp^^ jung igji,^ at 9 .30 ^ jj Temperature constituent which may be actively corrosive under 18" C. certain conditions, was found to be 1.5 parts per mil- No. 2, Metropolitan Water District water obtained at Well E, lion in the first sample and 8 parts per million after ^, ^ J'"^<= l^'!"- ^^ If ^-^J;: Temperature 18= C .... . ■ -n 4. ■ No. <5, Metropolitan Water District water after chlorination, chlorination; not a very significant increase. taken at office, June 19th, 9:53 A.M. Temperature IS" C. ■\r I, !• « u J i-u « „„-™«i«„ ;^ +v,of nVilrt No. 4, Metropolitan Water District water from Well G, samp- My belief, based upon these samples, is that chlo- j.^^ ^^p ^^^^^ ^^ ^.^ ^^^ j^^^ ^^^^ Temperature rination at a dosage of 15 parts per million has not 18° c. increased the aggressiveness of this Avater to any sig- No. 5, Metropolitan Water District water taken at Well I, at nificant degree. However, because the acceptance rate 10:03 A.M. June I9tii. Temperature IS" C. at the inieetion wells has not been dronDin" I SU"- ^°-^' Metropolitan Water District water from Well K, taken at ine injecnon weus nas noi ueen uroppin^, 1 su^ _^^ ^^^^^ ^^ j^^^ ^^^^ Temperature 18" C. gested that the chlorine dosage be dropped to 12 parts per million in the interest of economy. This change The least benign of these waters is that represented was made at once, but should the acceptance rate by Sample No. 3, but even here the pH of 7.45 and a start to drop it may be desirable to restore the chlo- free carbon dioxide content of 10 parts per million rine dosage to its previous value. seems, in my opinion, unlikely to be actively corrosive. Samples Nos. 2, 4, 5 and 6 are not aggressive, while I would further suggest that simUar samples be ^^^ ^a^^^ ^^ Sample No. 1, is moderately protective, taken to check the pH drop under the present dosage of 12 parts per million. Conclusion : Chlorination at a dosage of 12 parts per million appears to impose no especial hazard of Respectfully submitted, corrosion. Carl Wilson /s/ Recommendation : It seems desirable to repeat these samplings at in- tervals of two weeks, for a multiplicity of readings would give more dependable information than can be deduced from only one series. Respectfully submitted, Carl Wilson /s/ 10—52588 130 SEA WATER INTRUSION IN CALIFORNIA August 20, 1953 Mr. H. E. Hedger, Chief Engineer, Los Angeles County Flood Control District, Box 2418 Terminal Annex, Los Angeles 54, California. Attention : Mr. Fiuley B. Laverty Dear Sir : I have had the privilege and pleasure of reading the report by A. F. Bush and S. P. Mulford covering some of your problems, and now ask your permission to comment upon one remark which it seems to me might easily lead to confused ideas. I make reference to the second paragraph on page 9. There the statement is made; "No Myxomyces (slime-producing bacteria) were noted." Since we have done much talking about slime-form- ing bacteria this remark could be regarded as a refuta- tion of some of my own statements, but actually it is not, for it refers to a group of organisms (the Myxo- myces) which has no relation to our problem, and ig- nores the great class of slime forming saprophytic bacteria which are normal inhabitants of water and soils, and which, in my opinion, play an active part in retarding the continued introduction of water to an aquifer. As far as is known only one species of Myxo- myces is found in water, where it is parasitic on Clad- ophora, one of the larger attached algae. The others are confined to wood, dung and soils of high organic content. One of the best general discussions of the group is to be found in "Fundamentals of Bacteri- ology," by Mai-tin Frobisher, Jr., 4th Edition, pages 419 et seq., published by Saunders. The "slime-forming bacteria" with which we are concerned are those organisms, and their number is legion, covering many genera and species, which se- crete pectins (slimes) with which to anchor them- selves to the substrate, sand, gravel, and even well casings, pump bowls aud runners, etc. They have been present in all of your samples, and we have estimated their numbers relative to the total count in many in- stances. This we have been able to do by taking ad- vantage of their slime-forming proclivities. A clean and sterile micro slip, one inch by three inches, is im- mersed in the sample and left for twenty-four hours, during which period the organisms impinge upon the glass, where they attach themselves with a quantum of pectin. By appropriate staining methods we are able to recognize the pectin (slime). A large number of these culture slides from your earlier samples showed numbers of slime-forming bacteria nearly as great as the total count on standard nutrient agar incubated for 72 hours at 20° C. I will repeat this work on some of your current samples and furnish you with some actual figures. Perhaps I can make a photograph of a culture slip to show the slime, and if so I will send you a print. Trusting you will pardon this long and uninvited note, I am. Yours faithfully, Carl Wilson /s/ August 20, 1953 Mr. H. E. Hedger, Chief Engineer, Los Angeles Coiinty Flood Control District, Box 2418 Terminal Annex, Los Angeles 54, California. Attention : Mr. Finley B. Laverty Dear Sir : You recently sent us two samples of water from your Manhattan Beach well 8, taken August 12th, with a request for a special examination to determine the nature of the yellow color which this water ex- hibited after standing for a while. A report on the bacterial findings was sent to j^ou under date of Au- gust 19th. It remains now to report on the microscop- ical and chemical investigation. The microscope revealed the presence of iron oxide floe embedded in bacterial slime. We found no organ- isms of specific significance. The chemical examination confirmed the presence of iron oxide and showed manganese to be absent. Iron in water in insoluble forms, usually as oxides, but often as si^lphides, is found in most alluvial fills, and it is taken into solution as ferrous bicarbonate (colorless) by the carbon dioxide released by bacterial activity. When such water comes into contact with air the iron is oxidized to a basic ferric carbonate, which shortly becomes converted to hydroxide (insoluble in water) shortly to be precipitated as the red-brown ferric oxide (FcaOs). This has happened in your samples, giving rise to the phenomenon which at- tracted the attention of your engineers. Trusting this gives you the desired information, I am, i Yours faithfully, Carl Wilson /s/ SEA WATER INTRUSION IN CALIFORNIA 131 December 24, 1954 Mr. II. E. IIedger, Chief Eu