EARTHSCEM,‘ . W1 assorZr’siasuHIaB THE SAN JOAQUIN RIVER AND ITS IBUTARY STREAMS, CALIFORNIA, 1985 This report was prepared by the U.S. Geological Survey in cooperation with the San Joaquin Valley Drainage Program. The San Joaquin Valley Drainage Program was established in mid-1984 and is a cooperative effort of the U.S. Bureau of Reclamation, U.S. Fish and Wildlife Service, U.S. Geological Survey, California Department of Fish and Game, and California Department of Water Resources. The purposes of the Program are to investigate the problems associated with the drainage of agricultural lands in the San Joaquin Valley and to develop solutions to those problems. Consistent with these purposes, program objectives address the following key areas: (1) Public health, (2) surface- and ground-water resources, (3) agricultural productivity, and (4) fish and wildlife resources. Inquiries concerning the San Joaquin Valley Drainage Program may be directed to: San Joaquin Valley Drainage Program Federal—State Interagency Study Team 2800 Cottage Way, Room W-2143 Sacramento, California 95825—1898 TRACE ELEMENTS IN BED SEDIMENTS OF THE SAN JOAQUIN RIVER AND ITS TRIBUTARY STREAMS, CALIFORNIA, 1985 By Daphne G. Clifton and Robert J. Gilliom U.S. GEOLOGICAL SURVEY Water—Resources Investigations Report 88—4169 Prepared in cooperation with the SAN JOAQUIN VALLEY DRAINAGE PROGRAM 6439-34 Sacramento , California 1989 51/938883 gal/a7" DEPARTMENT OF THE INTERIOR MANUEL LUJAN, JR., Secretary U.S. GEOLOGICAL SURVEY Dallas L. Peck, Director For additional information Copies of this report can write to: be purchased from: District Chief U.S. Geological Survey U.S. Geological Survey Books and Open—File Reports Federal Building, Room W-2234 Federal Center, Box 25425 2800 Cottage Way Building 810 Sacramento, CA 95825 Denver, CO 80225 Abstr Intro Study ”7’ ~~%/ x...“- , an JV: { grim ’9! i j 432% %? 6M7" CONTENTS act........................................................................ duction.................................................................... area...................................................................... Study design and methods........................................................ Conce Compa Envir Summa Refer Figur Table Sample collection.......................................................... Laboratory analyses........................................................ ntrations and distribution of trace elements............................... Effects of particle size and organic carbon................................ Abundance of trace elements................................................ Interrelations among trace elements and their areal distribution........... rison of elements in bed sediments and valley soils........................ onmental significance...................................................... ry......................................................................... ences cited................................................................ ILLUSTRATIONS e 1. Map showing location of study area and sampling sites................ 2. Graph showing distribution of <62—um bed sediments and concentrations of selected elements in <62—um bed sediments by site groups........................................... 3. Graph showing principal component analysis of elements in <62-um bed sediments .............................................. TABLES 1. Selected characteristics of site groups used for data analysis....... 2. Distribution of primary and duplicate analyses among types of bed-sediment samples...................................... 3. Summary of concentrations of elements in <62—um bed sediments from the 22 sites sampled................................ 4. Comparison of median element concentrations in <62-um and >62-um size fractions, and in whole bed sediments from six sites.......... 5. Partitioning of elements in San Joaquin River water compared to concentrations in <62—um bed sediments.......................... 6. Correlation of trace elements with the first four principal components......................................................... Contents Page . l . 2 . 2 . 4 . 6 . 6 . 8 . 12 . l3 . 14 . l8 . 20 . 21 . 22 Page . 3 . 9 . 17 Page . 5 . 7 . 8 . 12 . l4 . 16 III Page Table 7. Element concentrations in San Joaquin Valley soils compared to bed sediments in the San Joaquin River basin .................... 19 8. Comparison between trace—element concentrations in bed sediments from the San Joaquin River basin and other river systems ........... 20 9. Trace-element concentrations in bed sediments compared to hazardous waste criteria ........................................... 21 10. Analyses of duplicate samples for major and trace elements in bed sediments ...................................................... 25 ll. Analyses of duplicate samples for carbon in bed sediments ............ 27 12. Trace elements, major elements, organic carbon, particle size, and moisture content of <62—um bed sediments ................. 28 13. Trace elements, major elements, organic carbon, particle size, and moisture content of >62-um bed sediments ................. 31 14. Trace elements, major elements, organic carbon, particle size, and moisture content of whole bed sediments ........................ 32 CONVERSKHJFACTORS For use of readers who prefer to use inch-pound units, rather than the International System (SI) units used in this report, the following conversion factors may be used: Multiply Ex To obtain cm (centimeter) 0.394 inches ha (hectare) 2.471 acre km (kilometer) 0.6214 mile m (meter) 3.281 foot Concentrations of elements in bed sediments are given in micrograms per gram (pg/g). Micrograms per gram is equal to parts per million. ABBREVIATIONS USED h hour um micrometer <4 um less than 4 micrometers <62 um less than 62 micrometers >62 um greater than 62 micrometers TRADE NAMES The use of brand or trade names in this report is for identification purposes only and does not constitute endorsement by the U.S. Geological Survey. IV Contents TRACE ELEMENTS IN BED SEDIMENTS OF THE SAN JOAQUIN RIVER AND ITS TRIBUTARY STREAMS, CALIFORNIA, 1985 By Daphne G. Clifton and Robert J. Gilliam ABSTRACT The occurrence and distribution of trace elements in bed sediments of the San Joaquin River, California, were assessed to determine whether some ele- ments are concentrated in the sediments which may affect water quality adversely. Bed sediments were sampled at 24 sites on the San Joaquin River and its tributary streams in October 1985. Samples showed that the percentage of sediment less than 62 micrometers in size correlated with total organic carbon. Concentrations of elements are higher and less variable in bed sediment of less than 62—micrometer size compared to sedi- ment samples with all size fractions. Bed sediments are coarser in tributaries originating in the Sierra Nevada compared to tributaries draining the western valley, and thus trace-element composi- tion is different between the two groups. Interrelations among trace elements from different stream-site groups were examined using principal component analy— sis. Together, the first and second principal components account for 57 per- cent of the variance, and show a distinct separation between sites dominated by Coast Range and Sierra Nevada sediments. The third and fourth components together account for 21 percent of the variance and distinguish the mixed-source sedi— ments of the intermittent upper San Joa- quin River from other parts of the river system. Elements in bed sediments of the San Joaquin River basin were similar in con- centration to elements in San Joaquin Valley soils, and were well below hazar— dous waste criteria. Concentrations were lower than in sediments from some pol— luted urban rivers and were comparable to rivers in other rural agricultural areas. The results indicate that selenium and other trace elements from subsurface agricultural drains have not concentrated to hazardous levels in bed sediments of the San Joaquin River. Abstract 1 INTRODUCTION High concentrations of selenium in sub- surface agricultural drain water from the western San Joaquin Valley, California, have impaired waterfowl reproduction where water was impounded at Kesterson Reservoir (Ohlendorf and others, 1986). Similar concentrations, in the range of hundreds of micrograms per liter, have reduced growth and survival of fish in laboratory experiments (Hamilton and others, 1986). Previous studies have described the areal distribution of selenium and other trace elements in drain water and in shallow and deep ground water of the western San Joaquin Valley (Deverel and others, 1984; Presser and Barnes, 1984, 1985; Neil, 1986; Deverel and Millard, 1988). Tidball and others (1986a, 1986b) studied the distri- bution of selenium and other elements in valley soils. Subsurface drain water from about 31,000 ha of western valley farmland, some of which contains high concentrations of selenium and other dis— solved elements, flows to the San Joaquin River or its tributaries (California State Water Resources Control Board, 1987). There is concern that, even after dilution in the river, selenium or other trace elements may adversely affect the water quality of the river or concentrate in the bed sediments. The purpose of this study is to assess the occurrence and distribution of trace elements in bed sediments of the San Joa- quin River and its tributaries. Trace- element concentrations in bed sediments can be indicators of potential water- quality problems. Bed—sediment samples for this study were collected from 24 sites on the San Joaquin River and its tributaries during October 7-11, 1985. Sample analyses included trace elements, major elements, organic carbon, and par— ticle size. This study is part of a com— prehensive investigation of the hydrology and geochemistry of the western San Joa- quin Valley by the U.S. Geological Survey in cooperation with the San Joaquin Valley Drainage Program. STUDY AREA Below the headwaters of the San Joa— quin River in the Sierra Nevada, the river extends 309 km from Friant Dam in the foothills, to Vernalis (fig. 1) just upstream from the tidal backwater influ— ence of the Sacramento—San Joaquin Delta. For the first 105 km between Friant Dam and Mendota, the river generally has intermittent flow and often river water does not reach Mendota Pool near Mendota. In the next 108 km between Mendota and Stevinson, the river has perennial flow in the upper 33 km because of Delta Men— dota Canal inflows, and intermittent flow along the remaining 75 km as the result of irrigation. diversions. Flow in 'the remaining 96 km between Stevinson and Vernalis is perennial and increases down- stream as tributaries, irrigation—return flow, and ground water enter the river. This study focuses on the San Joaquin River between Mendota and Vernalis and on the tributaries to the river within that reach (fig. 1). About 0.8 million ha of irrigated agricultural land drain direc- tly or indirectly to this reach of river (California State Water Resources Control Board, 1987). During low—flow condi- tions, water in the river consists mainly of irrigation—return flows (Hunter and others, 1987), surface runoff, and sub- surface drain water. Surface irrigation- return flow is typically high in sus— pended sediments from soil erosion (California Regional Water Quality Con— trol Board, 1977). Drain water from sub- surface drainage systems in the western San Joaquin Valley is typically high in dissolved solids, boron, and commonly selenium, and low in suspended sediments (California Regional Water Quality Con- trol Board, 1979; Deverel and others, 1984). During high-flow periods, most San Joaquin River water is runoff from the Sierra Nevada. Valley soils consist of weathering products from either the Sierra Nevada to the east, the Coast Range to the west, or a combination of both. The Sierra Nevada is composed of igneous and metamorphic 2 Trace Elements in Bed Sediments, San Joaquin River and Tributaries, California 121|°00’ 'uer @ Stanislaus Rl Ripon . Je/usale o 4’2 Waslew. -— ' Verali 1 Hospital/"' "’— . crelek‘ «70 STUDY A - n am a , cgek--\,W¢~elzsrry (’41, 120 30 37030,___as ewa ' VALLEY DelPuertoS Creek ,_./ OrestimbaJs' N‘ Creek—=' . Newman/ F I 00 Wasleway Gusting 7 us Garzas, , A Creek ‘5’ . 2‘ A. Manposa Slough ‘9“ 0” Los Banos .5 99 37°oo'—|Sw):§7/ __ FriantDam—> (32 kilometers) EXPLANATION 1 Mendota _, ~/ M-‘aanQI..,(d 712 SAMPLING SITE AND NUMBER Mendotao 12 " M PERENNIAL FLOW N fum— INTERMITTENT FLOW "32 ‘3 0 “99.191: "-x 10 20 MILES 10 20 KILOMETERS FIGURE 1.--Location of study area and sampling sites. Study Area 3 rocks. Sediments derived from these rocks form the alluvial fans originating in Sierra Nevada foothills and are the main component of flood-plain deposits in the San Joaquin River. River channels of Sierra Nevada tributaries river deposits of gravel, sand, silt, and small amounts of clay. These sediments are coarser and more permeable than those from the Coast Range. The Coast Range is composed of gypsiferous marine shale, sandstone, and volcanic fragments, and soils of Coast Range alluvial fans gen- erally are fine grained. Coast Range streams drain continental rocks and deposits, and the streambeds are a heterogeneous mixture of poorly sorted clay, silt, sand, and gravel (Miller and others, 1971; Page, 1983, 1986). Bed sediments in the San Joaquin River channel are a mixture from both sources, but the coarse-grained Sierra Nevada sediments predominate because of the much higher streamflow of the tributaries originating in the Sierra Nevada. Study sites on Bear Creek, Los Banos Creek, Salt Slough, and Mud Slough (fig. 1) are in flood—basin deposits consisting of clay, silt, and some sand (Page, 1986). STUDY DESHHJAND METHODS Sampling sites were selected at 24 loca- tions, which represent the intermittent— flow and perennial-flow parts of the San Joaquin River, tributaries from the Sierra Nevada and Coast Range, and canals and sloughs that carry irrigation—return flow and subsurface agricultural drain water to the San Joaquin River (fig. 1). Site numbers, names, and groups used for data analysis are given in table 1. Site groups were based on geography, primary sources of water, geology, and hydrology. The upper San Joaquin River (sites 12, 14, 16, 18, 21, 22, and 25) contained mixed sources of water and sediments during the study period, primarily water from the Delta Mendota Canal and irrigation—return flow. Bed—sediment material contained from 2 to 94 percent consist of. of the less than 62-micrometer (<62—um) size fraction. The finest grained sedi- ments occurred in pooled areas and at the intermittent—flow sites, which contained predominantly irrigation-return flow. Streamflow in the San Joaquin River below Mendota Pool (where Delta Mendota Canal water is stored) was equivalent to 15 to 16 percent of flow in the San Joaquin River near Vernalis (site 11), the far- thest downstream site. Streamflow down- stream of Dos Palos (site 18), after all the canal water from Mendota Pool was removed for irrigation, was equivalent to less than 0.1 percent of flow at site 11. At the time of the study, water released to the river from Mendota Pool did not reach the perennial—flow part of the San Joaquin River at Stevinson (site 1). Although Salt and Mud Sloughs (sites 2’ and 4) enter the San Joaquin River from the west, they’ were grouped separately because they receive more subsurface agricultural drain water than other west- side tributaries. The sloughs also have coarser bed sediments (8 and 9 percent <62 um) compared to westside tributary streams. Streamflows in Salt and Mud Sloughs were about '7 and 2 percent of flow at Vernalis during the study period. Streams classified as westside tribu— taries (sites 29, 32, 34, 35, 36, 39, 40, and 42) primarily originate from small valleys at the foot of the Coast Range. Except during high runoff events, these streams consist mainly of intermittent irrigation-return flow, overflow water from the Delta Mendota Canal, and, in some cases, a small quantity of subsur— face drain water. Streamflow during the sampling period in individual streams of this group ranged from less than 0.1 to 3.8 percent of streamflow at Vernalis. Bed sediments consisted of 14 to 72 percent in the <62-um size fraction. The four eastside tributaries (sites 5, 8, 10, and 27) originate in the Sierra Nevada and contain primarily coarse bed sediments (less than 1 percent <62 um). These tributaries have the highest flows compared to other tributaries-—each 4 Trace Elements in Bed Sediments, San Joaquin River and Tributaries, California TABLE 1.-—Selected characteristics of site groups used for data analysis [Streamflow data collected September 23-27, 1985; bed-sediment data collected October 7-11, 1985; do., Do., ditto; <, less than; um, micrometer] Stream- <62-um flow size (as fraction Geologic Site Site name percent Type of of bed sources Primary source No. of flow streamflow sediments of bed of water at (as percent sediments at the time Vernalis) of sample) of the study Upper San Joaquin River and Tributaries (Site group 1) 12 Fresno Slough near 62-um size fraction, the whole sample, as well as sieve wash water were analyzed in addition to the <62-um size fraction. Five samples, including three of <62-um sediments and two whole samples, were split for duplicate element analysis, and five samples were split for duplicate carbon analyses. Laboratory Analyses Samples for element determination were analyzed in the U.S. Geological Survey laboratory in Denver by Paul Briggs and David Fey. Sample aliquots were measured for percent moisture, air-dried, ground, and homogenized. Analysis of most of the elements involved digestion in strong‘ acids (HF, HCl, HNO3, HClOn) prior to inductively coupled plasma (ICP) analysis coupled with atomic emission spectroscopy (Crock and others, 1983). 6 Trace Elements in Bed Sediments, San Joaquin River and Tributaries, California TABLE 2.——Distribution of primary and duplicate analyses among types of bed-sediment samples Site Primary analyses of elements No. and total organic carbon <62 um >62 um Whole Duplicate analyses Total Elements organic carbon <62 um Whole <62 um whole xxxxxx I I H H H o I I l I I I I I N 0 xxxxxxxxxxxxxxxx l I I I Major elements analyzed were aluminum, calcium, iron, magnesium, phosphorus, potassium, sodium, sulfur, and titanium. Trace elements analyzed were barium, beryllium, bismuth, cadmium, cerium, chromium, cobalt, copper, europium, gal- lium, gold, holmium, lanthanum, lead, lithium, manganese, molybdenum, neo- dymium, nickel, scandium, silver, stron— tium, tantalum,/ thorium, tin, uranium, vanadium, ytterbium, yttrium, and zinc. Selenium and arsenic were determined by hydride generation-atomic absorption spectrometry after digestion in strong acids (HF, HCl, HN03, HZSOu) (Crock and Lichte, 1982; Briggs and Crock, 1986). A cold vapor, atomic absorption procedure was used for mercury determination after digestion in a NaZCrOH/HNO3 solution (Crock and Kennedy, 1986). Total organic carbon (TOC) in bed sediment was deter- mined by the difference of total carbon (measured by oxidizing the sample in an induction furnace) and total inorganic carbon (TIC) (measured by treating a sam— ple with acid, heating it, and measuring the amount of carbon dioxide evolved) (Wershaw and others, 1987). Standard reference samples were used for quality control. Study Design and Methods 7 For all major and trace elements, analyses of duplicate field samples were not significantly different (a=0.05). Results of duplicate analyses are listed in tables 10 and 11 (at the back of report). The standard deviation of laboratory analyses of the inorganic con— stituents generally is less than 5 per— cent (Crock and Lichte, 1982; Crock and others, 1983; Briggs and Crock, 1986; and Crock and Kennedy, 1986). Dry—weight concentrations of elements in suspension in washwater remaining after sieving and settling for 12 h or more were five to seven orders of magni— tude lower than concentrations in bed sediments. This indicates that the time period (12 h or more) allowed for settling the <62—un1 particles from the slurry created during the sieving pro— cess was adequate, and loss from bed sediments in the remaining washwater was insignificant. Particle—size analyses were done by the U.S. Geological Survey sediment laboratory in Salinas, California. The particle—size distribution of the <62-um size fraction of bed sediments was deter- mined from the hydraulic properties of the particles and their fall Velocity using the visual accumulation tube- pipette method. The >62—um size fraction was determined using sieve analysis (Guy, 1977). CONCENTRATIONS AND DISTRIBUTION OF TRACE ELEMENTS Concentrations of trace elements, major elements, and total carbon in the <62—um bed sediments are summarized in table 3 for the 22 sites at which it was possible to collect enough <62—um sediments for chemical analysis. Concentrations of selected elements are summarized by site groups in figure 2. Tables 12 to 14 (at the back of report) list bed—sediment data for all, sites and sample types. TABLE 3.--—Summary of concentrations of elements in <62—um bed sediments from the 22 sites sampled [other elements were analyzed for but not detected and are given at the end of this table. <, less than; um, micrometer; ug/g, micrograms per gram] Constituent Minimum Median Maximum Physical characteristics (percent) <62-um <1 15.5 94 grain size. Major elements (ug/g) Aluminum..... 66,000 79,500 89,000 Ca1cium...... 10,000 19,000 32,000 Iron......... 18,000 41,500 50,000 Magnesium.... 5,400 14,500 25,000 Phosphorus... 5,000 800 1,300 Potassium.... 12,000 18,000 23,000 Sodium....... 3,000 14,500 23,000 Sulfur....... 100 300 6,600 Titanium..... 2,300 3,900 4,800 Trace elements (ug/g) Arsenic...... 5.4 9.15 16 Barium....... 640 755 990 Beryllium.... l 1 2 Cadmium...... <2 <2 15 Cerium....... 41 48.5 73 Chromium..... 22 97.5 180 Cobalt....... ll 16 30 Copper....... 8 38 560 Gallium...... 13 19 20 Lanthanum.... 23 29.5 38 Lead......... 15 20 790 Lithium...... 16 51 63 Manganese.... 620 920 4,400 Mercury...... .05 .08 .40 Molybdenum... <2 <2 3 Neodymium.... 19 25 30 Nickel....... 22 69 120 Scandium..... 5 13 17 Selenium..... <.2 .5 1.5 Strontium.... 190 290 440 Tin.......... <20 <20 90 Thorium...... 8 13 19 Vanadium..... 44 110 140 Ytterbium.... 1 2 3 Yttrium...... 14 17 22 Zinc......... 34 110 230 Carbon, 2,100 9,600 29,000 organic, total. Carbon, <100 250 8,000 inorganic, total. Other elements analyzed but not detected Detection Detection Element limit Element limit (ug/g) (ug/g) Bismuth..... <10 Silver...... <2 Europium.... <2 Tantalum.... <40 Gold........ <8 Uranium..... <100 Holmium..... <4 8 Trace Elements in Bed Sediments, San Joaquin River and Tributaries, California IRON, IN MICROGRAMS PER GRAM 100 N # m o o 8 o PERCENTAGE OF BED SEDIMENT < 62 um 0 50,000 40,000 30,000 20,000 10,000 90,000 E < D: (D II Lu 0. 2 80.000— E _ _ < II o E O [I 9 — 2 Z 70,000 — o _ 5 D ._ Z (2) E (7) a (8) (3) (2) 3 (7) (2) (8) (3) (2) E < —Q— 1 2 3 4 5 60'000 1 2 3 4 5 SITE GROUP NUMBER SITE GROUP NUMBER 7,000 2 _ I?! E 6000— — O Q . —9— __ n: Medlan E 5,000— '— . 2 0 g 4,000— — _ o _ 2 g 3,000— — 2 Z 2,000 — — __ of ° 2 _I 1,000 — -— (7) (2) (8) (3) <2) 8 (2) ‘fi’ (0’ “I 0 1 2 3 4 5 1 2 3 4 5 SITE GROUP NUMBER SITE GROUP NUMBER EXPLANATION — Median Boxes represent the middle 50 percent of data. Lines extending from boxes represent range of data, excluding outliers 0 Values are more than 1.5 times the interquartile range from the top or bottom of the rectangle. 0 Values are more than 1.0 times the interquartile range from the top or bottom of the rectangle. (4) Numbers in parentheses indicate number of observations in each site group. SITE GROUP NAME 1 Upper San Joaquin River 2 Salt and Mud Sloughs 3 Westside tributatries 4 Eastside tributaries 5 Lower San Joaquin River FIGURE 2.--Distrlbution of < 62-um bed sediments and concentrations of selected elements in < 62—um bed sediments by site groups. Concentrations and Distribution of Trace Elements 9 6,000 16 3 E (I SE 6 14— __ o E 0: 1:. E ° 2 12— — (0 Ian 5 4.000 —— Q Ca E _ 8 10—— E _ 8 E 5 g 2 8— _ E Z 2 a e~ — _ (3) (D g (7) <2) . <3) (2) g (7) (2) (8) (3) ‘2) .— — 4 '— 2’000 1 2 3 4 5 1 2 3 4 5 SITE GROUP NUMBER SITE GROUP NUMBER 200 600 E E 175 E 0 55 500 — — D: (I E 150— “ g g m 400— _ < 125— — 3 2F, u: g 100— _ 8 300— .— 2 E E Z 75 ~ — i 200 _ _ 2' _. _ __ 2 50 E (2) 2 <8) & 100 — (2) <8) (3) — O 251—— . — O (7) g (7) <2) <3) (2) o q: E a — O O o o 1 2 3 4 5 1 2 a 4 5 SITE GROUP NUMBER SITE GROUP NUMBER 800 70 z 3 700— _ é £3 60— _ (5 E 600— _ E 3: 500 g 50— _ 3 e 55 400—— ~ 8 40— E — O n: 5 300— — 9 S i 30 — — §_ 200— _ 2” D D 4 (7) (a) E 5 (7) (2) o (3) (2) .0. _ 0 1 3 4 5 1° 1 2 3 4 5 SITE GROUP NUMBER SITE GROUP NUMBER FIGURE 2.--Distrlbution of < 62-pin bed sediments and concentrations of selected elements in < 62- um bed sediments by site groups --Continued. lO Trace Elements in Bed Sediments, San Joaquin River and Tributaries, California 0.40 91 O O O .1; b o O l 0.35 0.30 (7) (2) —o— (8) [:l (1) (1) E < 2 I < 0 O: 0: o E 35 (D D. E a) g 3 8 3,000— — g 0.25 6 ° 0 S €15 0.20 2 2,000 — — 2 ‘- z 5 E - s 0.15 g 1 000 — g E -— g {(0 E: E 0.10 E (7) (2) (8) (3) (2) 2 0.05 0 1 2 a 4 5 SITE GROUP NUMBER 120 1.6 5 3 1: 5 1.4 g 100 — —- a: LIJ E 3) 1.2 m —‘ 2 2 < 80— g - E‘ 1.0 E 0 8 2 5 0 g 0.8 _ 6 —- -— 2 L. i g —. 0.6 .1“ :2) UJ 40 _ _ 5 r g 0.4 E (7) (2) (3) (3) (2) g C 0.2 20 1 2 3 4 5 SITE GROUP NUMBER 250 30,000 E 25 000 g 200 — — z- 5 ' I O 0: Lu 0: (D n. 1: E20000 U) < LLI 2 150 — — 0 O. < o 9 w 5 § 315,000 g 100 _Mediano E Q _ § § 3 _. {110,000 2 ,5 $2 —' 50 — -— '9 2 2 o 3 5.000 N (7) (2) (8) (3) (2) 0 0 1 2 3 4 5 FIGURE 2.--Distribution of < 62-um bed sediments and concentrations of selected elements in < 62-um bed sediments by site groups-—C0ntinued. SITE GROUP NUMBER Concentrations and Distribution of Trace Elements 1 2 3 4 5 SITE GROUP NUMBER (7) (8) (2) —e— (2) (1) 1 2 3 4 5 SITE GROUP NUMBER (3) a (7) (2) (3) (2) 1 2 3 4 5 SITE GROUP NUMBER 11 Effects of Particle Size and Organic Carbon This study focused on the <62-um size fraction of bed sediments for element analyses and assessment of areal distri- bution because the greatest concentration of most trace elements are usually asso- ciated with the clay— and silt-sized par- ticles, which have the greatest surface area (Salomons and Forstner, 1984; Horo— witz, 1986). Analysis of the <62-um size fraction is a means of standardizing for site comparisons. The abundance of organic matter in bed sediments is another key factor that affects trace-element concentrations. Many elements, such as cadmium, chromium, copper, iron, lead, manganese, molyb— denum, nickel, and zinc, tend to concen- trate in organic coatings on sediment particles (Horowitz, 1984; Salomons and Forstner, 1984). However, a combination of low organic carbon content (0.02 to 2.77 percent) and a high correlation (r2=0.91) between organic carbon and the <62—um size fraction made it difficult to distinguish the effect of organic matter content from the effect of particle-size on element distribution in San Joaquin River bed sediments. If the organic mat— ter is present as coatings on particles, a high positive correlation would be expected between organic carbon and the increasing proportion of material with the greatest surface area. Trace-element concentrations generally were higher and less variable in the <62-um size fraction than_in larger size fractions. At six sites where trace elements were determined in <62-um and >62-um size fractions and whole sediments (table 2), median concentrations of most elements were highest in the <62-um frac— tion (table 4). Results for these six sites, plus eight additional sites for which whole-sediment samples were analyzed (table 2), showed that aluminum, arsenic, copper, iron, lithium, magne- sium, mercury, phosphorus, selenium, TABLE 4.——Compar'ison of median element concentrations in <62—um and >62—um size fractions, and in whole bed sediments from six sites [The six sites are 2, 5, 14, 22, 25, and 36; see table 1 for site names. <, less than; um, micrometer; ug/g, micrograms per gram] Median concentration Element <62 um >62 um Whole Major elements (ug/g) A1uminum........ 76,700 76,500 77,500 Ca1cium......... 21,300 20,700 21,500 Iron............ 37,300 28,800 33,000 Magnesium....... 11,900 9,600 12,000 Phosphorus...... 8,170 7,170 6,670 Potassium....... 17,200 20,000 17,700 Sodium.......... 17,000 21,200 19,000 Sulfur.......... 300 270 <100 Titanium........ 3,720 2,930 3,480 Trace elements (Hg/g) Arsenic......... 8.25 5.23 5.05 Barium.......... 778 780 762 Beryllium....... 1.2 <1 1.2 Cadmium......... <2 <2 <2 Cerium.......... 48.2 38.5 44.5 Chromium........ 72.7 44.8 64.2 Cobalt.......... 14.5 11.5 14.5 Copper.......... 34 20.5 27.8 Gallium......... 17 17.2 17 Lanthanum....... 29.5 24.2 27.3 Lead............ 22 20.2 19.3 Lithium......... 41.2 31.5 36.3 Manganese....... 1,610 943 857 Mercury......... .08 .03 0.02 Molybdenum...... <2 <2 <2 Neodymium....... 24.7 21.7 24.8 Nickel.......... 47.7 30.0 58.7 Scandium........ 11.5 8.2 10.7 Selenium........ .3 .15 .1 Strontium....... 288 318 290 Thorium......... 13.8 10 12.3 Tin............. <20 <20 <20 Vanadium........ 97.5 72.3 88.3 Ytterbium....... 1.8 1.6 1.8 Yttrium......... 15.3 12.5 15.2 Zinc............ 113 70.7 82.5 Carbon, organic, 10,900 9,150 7,220 total. 12 Trace Elements in Bed Sediments, San Joaquin River and Tributaries, California sulfur, titanium, total organic carbon, and the percent <4-um size fraction, were correlated with the fraction of sample material <62 pm. For 12 of these 14 sites (table 2), whole and <62-um sedi- ments were analyzed. Concentrations of 28 of the 33 elements analyzed in whole sediments showed greater variability com— pared to concentrations in <62-um sedi— ments from the same 12 sites, as deter- mined using the coefficient of variation (geometric deviation/geometric mean). Only barium, copper, lead, manganese, and sodium concentrations were more variable in the <62-um size fraction. The median coefficient of variation for elements in whole sediments (0.468 ug/g) was nearly double that for the same elements in <62-um sediments (0.238 ug/g), indicating the effectiveness of using the <62—um size fraction to reduce variability of element concentrations between sites. The median percentage of the <62-um size fraction in bed sediments was much lower in eastside tributaries (less than 1 percent), the lower San Joaquin River (1 percent), and Salt and Mud Sloughs (8.5 percent), when compared to westside tributaries (41 percent) and the upper San Joaquin River (67 percent) (table 1, fig. 2). Analysis of variance showed that the proportion of <62-um particles was significantly different between the eastside and westside tributaries (P<0.05); the Mann-Whitney test also 'showed a. significant difference between medians (P<0.05). But there was no sig— nificant difference between either of these site groups and the sites on the upper and lower San Joaquin River, and Salt and Mud Sloughs, which contain a variable mixture of sediment derived from eastside and westside sources. Standardizing trace—element concentra- tions by analyzing only the <62-um par— ticles decreases the effect of these physical differences between bed sedi- ments so that effects of the differences in geology, hydrology, and possibly land use on trace-element composition can better be examined. In addition, trace- element loads to the San Joaquin River and transport in the river are mainly in the <62-um fraction of suspended sedi— ments; between 80 and 90 percent of sus- pended sediments in the eastside tribu— taries, and 90 and 100 percent in the lower San Joaquin River and Sloughs are <62 um during most of the year (Shelton and Miller, 1988). Abundance of Trace Elements Most elements that were analyzed for in this study occurred at detectable concen— trations, although there is considerable variability in abundance among elements (table 3). Because of their low concen- trations, high detection limits, or both, bismuth, europium, gold, holmium, silver, tantalum, and uranium were not detected in any of the bed-sediment samples. Chromium, copper, lithium, manganese, nickel, and zinc are among the more abun- dant trace elements that also have been measured in river water. Except for lithium, these elements tend to be asso- ciated mainly with particulate matter in water (table 5). In contrast, arsenic, molybdenum, and selenium are among the least abundant elements in bed sediments and occur mainly in dissolved forms in river water (table 5). The relative similarity of elemental concentrations in bed sediments in dif— ferent parts of the river system, the overall variability in concentrations, and the small number of sites in each site group, result in few clear distinc- tions between site groups based on concentrations of individual elements (fig. 2). Comparison of site groups 1 through 4, which represent tributaries from differ- ent parts of the drainage system to the lower river, shows clear distinctions between site groups only for lithium, manganese, selenium, and zinc. Lithium is notably lower in <62-um sediments from eastside tributaries compared to the other three groups, and manganese and Concentrations and Distribution of Trace Elements l3 TABLE 5.-—Partitioning of elements in San Joaquin River water compared to concentrations in <62-um bed sediments [The proportion of elements in dissolved forms are medians of the ratios of dissolved to total recoverable concentrations at the 11 main sites in the San Joaquin River Study, June 1985 - January 1986. <, less than; um, micrometer; ug/g, micrograms per gram] Proportion Median <62 um in bed-sediment Element dissolved forms concentration (percent) (ug/g) (22 sites) Elements that occur mainly in dissolved phase Arsenic....... 75 9.15 Lithium....... 85 51 Molybdenum.... 100 <2 Selenium...... 100 0.5 Elements that occur mainly in particulate phase Aluminum...... 0.6 79,500 Chromium...... 4.2 97.5 Copper........ 25 38 Iron.......... 1.0 41,500 Manganese..... 24 920 Nickel........ 21 69 Zinc.......... 37 110 zinc are higher. Thus, the eastside tributaries, which have bed sediments derived solely from Sierra Nevada sources, typically have low lithium and high manganese and zinc in the <62—um size fraction. Selenium is distinctly higher in <62—um bed sediments from Salt and Mud Sloughs than it is for any other site group. This corresponds to the higher levels of selenium measured in the water at those sites as well. Interrelations Among Trace Elements and Their Areal Distribution Interrelations among trace elements were examined using principal component analysis. Principal component analysis expresses the total variance for a group of variables in terms of principal components, which are linear combinations of the original variables. Each prin- cipal component explains a part of the total variance and each original variable is related to each principal component to different degrees. Generally, the first few principal components explain most of the total variance and, sometimes, a com- bination of only a few of the total number of variables dominates each com— ponent. In this study, the variables are concentrations of selected major and trace elements and selected additional data on organic carbon and particle size. The combinations of elements and other properties that dominate the total vari- ance may result from factors (such as chemical, geological, or biological) that determine the abundance of certain combinations of elements. Because concentrations of some ele- ments varied over several orders of mag- nitude, the principal component analysis was based on a correlation matrix of standardized variables that were log transformed (Davis, 1973). Principal component analysis requires available data for every variable for each site. A few values for mercury (sites 1, 5, and 27), selenium (sites 1 and 5), and sulfur (sites 1, 5, 8, and 27) were missing. To avoid eliminating these sites from the analysis, concentrations were estimated using the median of the appropriate site group. This procedure did not distort the analysis for sites without missing data. The principal component analysis examines relations among concentrations of selected elements in the <62—um particle-size fraction of bed sediments collected from 22 sites in the different parts of the river system represented by the site groups (table 12). Trace ele- ments included in the analysis were arsenic, chromium, copper, lead, lithium, mercury, nickel, selenium, and zinc. All of these elements, except for lithium, can have toxic effects on aquatic life when present in high concentrations (U.S. Environmental Protection Agency, 1986). Little is known about the effects of lithium on aquatic plants and animals. Iron and manganese were included because iron and manganese oxides form coatings on particles which are sites for trace- metal sorption. Aluminum and titanium were included because they are regarded as conservative elements, with a uniform contribution to bed sediments over a long 14 Trace Elements in Bed Sediments, San Joaquin River and Tributaries, California period of time from crustal rock sources (Horowitz, 1984). Organic carbon was included because of the tendency for trace elements to become concentrated in organic material (Horowitz, 1984). The clay—size fraction of bed sediments (<4 um) was included because smaller par— ticles have more surface area for adsorp- tion of trace metals. Sulfur was added because of its association with selenium in westside ground water (Deverel and others, 1984) and soils (Tidball and others, 1986b). Data were not standard- ized to a conservative element (alumi- num), to the clay fraction (percent <4 um), or to total organic carbon, because correlations with elements in the <62—um bed sediments were low and princi- pal component analysis results were not improved. The first four principal components account for a total of 78 percent of the variance in the data. The first and second components combined account for 57 percent of the variance, whereas the third and fourth account for only 21 per- cent. The remaining principal components each account for 6 percent or less of the total variance. The first principal component accounts for 31 percent of the total variance (table 6). Positive variable loadings are greater than or equal to 0.20 for iron, chromium, lithium, nickel, zinc, aluminum, and copper, in order of domi- nance. Most of these elements occur mainly in the particulate phase in water (table 5). High negative loadings occur for mercury and titanium. The variance in scores computed for the first princi- pal component from variable loadings and their values at each site relate partly to differences in element composition (table 6) among site groups (fig. 3). With the exception of site 36, the clearest distinction among site groups based on the first-component scores is between westside tributaries, which are dominated by Coast Range sediments, and the eastside tributaries and lower San Joaquin River, both of which are domi- nated by Sierra Nevada sediments. With the exception of zinc and iron, westside tributaries generally have higher concen- trations of the dominant variables of the first component (fig. 2). This distinc— tion may be related to two main factors: the greater abundance of these elements in Coast Range formations, and the greater dominance of fine-grained par— ticles in Coast Range derived sediments. Aluminum, copper, iron, and zinc are often removed from solution by binding to fine particles (Johnson, 1986). Mercury and titanium concentrations generally were lowest in westside tributaries. The second principal component is almost as dominant as the first component and accounts for 26 percent of the vari- ance. The second component is dominated by high positive loadings for manganese, lead, copper, zinc, and total organic carbon in order of dominance, and high negative loadings for aluminum and <4—um particle size (table 6). The second com— ponent indicates an association of the elements manganese, lead, copper, and zinc, primarily with organic matter in <62—um sediments in which fine-grained and aluminum—rich particles of clay size are a relatively small proportion. Thus, component scores are highest for samples with high concentrations of trace ele- ments and carbon and low concentrations of aluminum and <4—um particles. Together, the first and second compo— nents show a distinct separation between the site groups dominated by Sierra Nevada sediments—-Mud and Salt Sloughs, the Eastside tributaries, and the lower San Joaquin River--and the westside tri— butaries that are dominated by Coast Range sediments (fig. 3). Results for site group 1, the upper San Joaquin River and tributaries, reasonably indicate a mixture of sediment sources (table 1). The distribution of component scores by site group (fig. 3) shows the differ- ence in bed—sediment composition of the eastside tributaries and Salt and Mud Concentrations and Distribution of Trace Elements 15 TABLE 6.—‘—Corr-elations of trace elements with the first four principal components [Only variable loadings percentage of total variance. 20.20 are reported. < Number in parentheses is , less than; um, micrometer] Variable loadings Principal Principal Principal Principal component I component II component III component IV (31) (26) (12) (9) Iron....... 0.41 Manganese.... 0.39 Sulfur....... 0.53 Titanium..... 0.49 Chromium... .38 Lead......... .38 Selenium..... .47 Nickel....... .44 Lithium.... .38 Copper....... .36 Arsenic...... .39 Arsenic...... .30 Nickel..... .34 Zinc......... .34 Carbon, .31 Chromium..... .30 Zinc....... .28 Carbon, .24 organic, Selenium..... .20 Aluminum... .26 organic, total. Carbon, -.42 Copper..... .20 total. Lead......... -.29 organic, Mercury.... «.30 A1uminum..... —.34 Copper....... —.25 total. Titanium... —.28 <4-um -.33 Sulfur....... -.26 particle Zinc......... -.23 size. Sloughs compared to the other site shales account for large sulfate concen- groups, which generally have low or nega- tive scores. The eastside tributaries have the highest median concentrations of lead, and the Sloughs have the highest median concentrations of manganese and zinc; both groups have lower median con— centrations of aluminum compared to the other site groups (fig. 2). Samples from all sites in site groups 2 and 4 contain a very low percentage of fine-grained sediments, generally less than 9 percent <62 um, and organic matter seems to be a more dominant factor affecting the trace- element composition at these sites. The third principal component, which accounts for 12 percent of the variance, has positive loadings for sulfur, sele- nium, arsenic, and total organic carbon, and negative loadings for lead and copper, in order of dominance (table 6). The association of selenium and sulfur may result from the similar chemical behavior of selenium and sulfur and their common source. A variety of selenium minerals are found in certain sulfide deposits (Luttrell, 1959). Mudstones and trations in ground water near Los Banos, south of Kettleman City (Miller and others, 1971); and selenium and sulfur concentrations are correlated in shallow ground water in wells in the western San Joaquin Valley (Deverel and others, 1984). Arsenic has been used as a cot— ton defoliant in the western San Joaquin Valley. The association of arsenic, selenium, and sulfur with total organic carbon (r2=0.34 to 0.61; a=0.05 for log- transformed data) suggests that one way in which these elements are enriched in bed sediments in all parts of the river system is through biological uptake and sedimentation of organic detritus. The distribution of component scores by site groups (fig. 3) shows little distinction among site groups. The fourth principal component accounts for only 9 percent of the variance, but shows some clear distinctions between site groups. High positive loadings occur for titanium, nickel, arsenic, chromium, and selenium, suggesting a Coast Range element assemblage, and high 16 Trace Elements in Bed Sediments, San Joaquin River and Tributaries, California 8 T I I I I I ._ S _ '26 ‘ LU Z 0 [L 5 o 4*' — j A E O E gz— 9 _ O z #0 o 0 $0 U. U 0 I E? o I O I 0)_2__ _ I .. | | I | | | -4 -12 —1o -8‘ -6 —4 —2 o 2 4 SCORE OF FIRST PRINCIPAL COMPONENT 2'5IITIII 2.0 — — SCORE OF FOURTH PRINCIPAL COMPONENT o —2.0 — O '— -2.5|II III -4—3-2-101234 SCORE OF THIRD PRINCIPAL COMPONENT EXPLANATION SITE GROUP NAME OI>I<> Upper San Joaquin River Salt and Mud Sloughs 1 2 3 Westside tributaries 4 Eastside tributaries 5 Lower San Joaquin River HWfiaMMMmmmeW$MMmmm62—um bed sediments [See table 1 for group and site names. <, less than; um, micrometer; ug/g, micrograms per gram] Concentrations are on a dry weight basis. Site group 1 Site group 2 Site group 3 Site group 4 Constituent 14 22 25 2 36 5 Physical characteristics (percent) Moisture........... 19 57 41 18 18 21 <62 um............. 39 93 94 9 l6 <1 Major elements (ug/g) Aluminum........... 74,000 80,000 76,000 65,000 83,000 81,000 Calcium............ 24,000 19,000 24,000 18,000 18,000 21,000 Iron............... 23,000 32,000 22,000 15,000 45,000 36,000 Magnesium.......... 7,600 11,000 8,000 5,000 14,000 12,000 Phosphorus......... 300 600 900 300 1,200 1,000 Potassium.......... 22,000 21,000 20,000 23,000 15,000 19,000 Sodium............. 23,000 22,000 24,000 20,000 17,000 21,000 Sulfur............. <100 300 900 <100 100 <100 Titanium........... 2,800 3,200 2,500 1,500 4,000 3,600 Trace elements (ug/g) Arsenic............ 3.1 8.7 4.6 5.7 5.4 3.9 Barium............. 770 760 750 960 710 730 Beryllium.......... 1 l 1 <1 1 1 Cadmium............ <2 <2 <2 <2 <2 <2 Cerium............. 36 39 36 24 49 47 Chromium........... 23 41 32 31 8O 62 Cobalt............. 9 16 10 6 15 13 Copper............. 4 27 18 9 44 21 Gallium............ 17 18 16 14 20 18 Lanthanum.......... 21 25 22 15 33 29 Lead............... 17 20 19 17 28 20 Lithium............ 21 42 23 12 51 40 Manganese.......... 530 670 470 790 2,100 1,100 Mercury............ .02 .04 .04 .02 .05 <.02 Molybdenum......... <2 <2 <2 <2 <2 <2 Neodymum........... 21 20 20 15 27 27 'Nickel............. 12 34 25 14 56 39 Scandium........... 8 8 6 5 12 10 Selenium........... <.1 .2 .2 <.1 .2 <.1 Strontium.......... 320 310 360 320 280 320 Thorium............ 9 10 7 4 17 13 Tin................ <20 <20 <20 <20 <20 <20 Vanadium........... 55 78 71 47 100 83 Ytterbium.......... 2 2 1 <1 2 2 Yttrium............ 13 12 12 7 16 15 Zinc............... 48 94 58 29 110 85 Carbon, organic, 7,800 28,000 18,000 1,500 4,600 700 total. Carbon, inorganic, 300 300 1,100 700 700 <100 total. Table 13 31 TABLE 14.——Trace elements, major elements, organic carbon, particle size, and moisture content of whole bed sediments [See table 1 for group and site names. Concentrations are on a dry weight basis. <, less than; um, micrometers; ug/g, micrograms per gram] Site group 1 Site group 2 Constituent 14 18 21 22 25 2 4 Physical characteristics (percent) Moisture............ 19 —- -- 42 59 17 -— <62 Um.............. 39 2 67 93 94 9 8 Major elements (Hg/g) A1uminum............ 80,000 63,000 78,000 82,000 79,000 81,000 69,000 Calcium............. 21,000 18,000 29,000 19,000 29,000 11,000 24,000 Iron................ 30,000 11,000 25,000 35,000 41,000 44,000 11,000 Magnesium........... 9,800 3,300 9,000 11,000 17,000 19,000 4,500 Phosphorus.......... 400 200 500 700 1,000 700 300 Potassium........... 20,000 23,000 18,000 13,000 17,000 17,000 25,000 Sodium.............. 21,000 19,000 24,000 19,000 16,000 15,000 23,000 Sulfur.............. <100 <100 1,100 200 1,200 100 600 Titanium............ 3,600 1,100 2,700 3,900 3,800 3,800 1,400 Trace elements (Hg/g) Arsenic............. 4.5 6.2 8.0 7.1 8.0 8.0 2.7 Barium.............. 780 860 680 690 750 680 930 Beryllium........... 1 1 1 2 1 1 l Cadmium............. <2 <2 <2 <2 <2 <2 <2 Cerium.............. 39 22 34 51 52 44 34 Chromium............ 38 7 39 53 68 150 21 Cobalt.............. 11 5 10 13 15 26 6 Copper.............. 15 4 9 26 34 62 6 Gallium............. 17 11 16 19 19 17 12 Lanthanum........... 25 12 18 33 33 26 17 Lead................ 21 17 29 18 19 23 15 Lithium............. 37 14 24 45 52 5O 12 Manganese........... 610 500 920 720 810 690 740 Mercury............. .03 <.02 .03 .04 .05 <.02 .02 Molybdenum.......... <2 <2 <2 <2 <2 <2 <2 Neodymium........... 23 11 17 28 26 25 15 Nickel.............. 23 5 17 36 48 200 17 Scandium............ 10 4 10 10 11 16 4 Selenium............ <.1 <.1 .3 .2 .3 .1 .2 Strontium........... 290 310 360 290 330 180 380 Thorium............. 10 5 8 17 16 12 10 Tin................. <20 <10 <10 <20 <20 <20 <10 Vanadium............ 72 23 56 87 120 120 25 Ytterbium........... 2 <1 2 2 2 2 1 Yttrium............. 15 7 14 16 15 18 11 Zinc................ 74 22 61 94 110 110 20 Carbon, organic, 7,800 -- -- 12,000 17,000 2,000 ~- total. Carbon, inorganic, 300 -- -- 500 4,700 1,200 —— total. 32 Trace Elements in Bed Sediments, San Joaquin River and Tributaries, California TABLE 14.—-Trace elements, major elements, organic carbon, particle size, and moisture content of whole bed sediments—~Continued Site group 3 Site group 4 Site group 5 Constituent 36 5 8 10 27 l 11 Physical characteristics (percent) Moisture... ..... .... 13 21 15 22 4 12 -- <62 um.............. 16 <1 1 <1 <1 <1 <1 Major elements (Hg/g) Aluminum............ 72,000 71,000 59,000 69,000 63,000 66,000 64,000 Ca1cium............. 22,000 27,000 19,000 22,000 19,000 20,000 18,000 Iron................ 29,000 19,000 17,000 23,000 6,800 16,000 15,000 Magnesium........... 9,200 6,000 5,300 7,600 2,100 3,300 4,700 Phosphorus.......... 600 600 300 500 200 200 400 Potassium........... 18,000 21,000 23,000 23,000 26,000 24,000 23,000 Sodium.............. 19,000 24,000 17,000 21,000 21,000 23,000 20,000 Sulfur.............. <100 <100 <100 <100 <100 <100 <100 Titanium............ 3,000 2,800 1,700 2,700 900 2,60 1,500 Trace elements (Hg/g) Arsenic............. 5.8 5.6 2.7 2.5 3.0 1.3 2.8 Barium.............. 800 870 1,000 1,000 980 810 1,000 Beryllium........... 1 1 <1 1 1 1 <1 Cadmium............. <2 <2 <2 <2 <2 <2 <2 Cerium.............. 34 47 24 28 24 49 28 Chromium............ 52 24 48 51 11 18 45 Cobalt.............. 12 10 7 12 4 4 7 Copper.............. 23 7 8 9 3 2 5 Gallium............. 15 15 12 14 11 13 12 Lanthanum........... 22 25 15 17 14 26 16 Lead................ 17 18 14 14 16 14 13 Lithium............. 18 16 9 13 9 10 11 Manganese........... 910 1,400 490 630 350 340 690 Mercury............. .12 <.02 <.02 <.02 <.02 <.02 <.02 Molybdenum.......... <2 <2 <2 <2 <2 <2 <2 Neodymium........... 21 26 13 15 12 25 13 Nickel.............. 27 18 24 29 9 6 17 Scandium............ 11 6 5 6 1 4 5 Selenium............ .2 <.1 <.1 <.1 <.1 <.1 <.1 Strontium........... 300 350 350 440 350 340 370 Thorium............. 9 10 4 12 5 8 5 Tin................. <20 <20 <20 <20 <20 <20 <10 Vanadium............ 83 48 52 66 18 39 40 Ytterbium........... 1 2 <1 1 <1 2 <1 Yttrium............. 12 15 8 9 7 15 8 Zinc................ 67 40 27 4O 15 21 22 Carbon, organic, 4,000 600 800 200 200 300 -- total. Carbon inorganic, 3,800 <100 <100 800 <100 <100 -- total. Table 14 33 nwwmnoa was mwwwwos In diam mrmZmzam Hz wmu mmUHz—mzam Om. aim m>z ho>ocHz wH20 Ham HwHwSJZQ maxmgm. n>erowzH> I.. £3.» mmuawoo 7§2§f ‘ s/l A ‘ , )4 i1, @ , w / ‘ Z / // 7K \ / ‘ 4 § /, $4, \x/a /, o 74x £5 // x¢§ \\ fl ‘ \s 7/ \> my §@ @wfiw . ‘ ‘ ‘ ‘ I ‘ ,, ,‘ m§§ , / g” Q“