djy. 7 'S 70 cO MP - /V- 7 b •/& r U.S. DEPARTMENT OF COMMERCE National Technical Information Service AD-A025 333 FEASIBILITY OF MONITORING FLOW PATTERNS AND SEDIMENT AND POLLUTANT DISPERSION OF WATER BODIES WITH 24-CHANNEL SPECTRAL DATA Army Engineer Waterways Experiment Station Digitized by the Internet Archive in 2018 with funding from University of Illinois Urbana-Champaign Alternates https://archive.org/details/feasibilityofmonOOsmit VSXQeo /<- r r. ’ a ‘‘ " vr:*'«o Dim Knl.rmd) REPORT DOCUMENTATION PAGE READ INSTRUCTIONS BEFORE COMPLETING FORM r REPORT RUMbER Miscellaneous Paper M-76-10 Z. GOVT ACCESSION NO. 3 recipient’s Catalog number A. TITLE < and Subtitle) FEASIBILITY OF MONITORINC FLOW PATTERNS AND SEDIMENT AND POLLUTANT DISPERSION OF WATER BODIES WITH 24-CHANNEL SPECTRALDATA 1 Type OF report ft PERIOD COVERED Final report f PERFORMING ORG. REPORT NUMBER 7. AuThOR(*) Margaret H. Smith 8 CONTRACT OR GRANT NUMBER'*, » PERFORMING ORGANIZATION NAME ANO ADDRESS U. S. Army Engineer Waterways Experiment Station Mobility and Environmental Systems Laboratory P. 0. Box 631, Vicksburg, Miss. 39180 10. program element project, task AREA ft WORK UNIT NUMBERS Project No. 6.11.01A, 4A061101A91D 11. CONTROLLING OFFICE NAME AND ADOPESS Office, Chief of Engineers, U. S. Army Washington, D. C. 20314 12. REPORT DATE May 1976 13 NUMBER OF P AGES U MONITORING AGENCY NAME ft AOORESSffl dll/«r«nl /rotn Controlling Otllca) is SECURITY CLASS (ol tbta raporf) Unclassified 15* DECLASSIFICATION 'DOWNGRADING SCHEDULE »6 DISTRIBUTION S T A T f U t N T (ot (hia Haport) Approved for pub]ic release; distribution unlimited. 17. DISTRIBUTION STATEMENT (ot tha aba trail on farad In Block 20, // d'tla-eni from Hoport) 10 supplementary notes D n c |~T^ i r~ - - —. . 19 KEY WORDS ( Con:Inua on irv«'«« alda // nacaaaary rand idont'.ty by block numbar) ' Chesapeake Bay Remote sensing 1976 Data processing Sediment Pollutant dispersion Sensors Rappahannock River Water flow 20 APSTR»C t rCocfh»j« cxi rvr*r«* • - iA // ’t*v k m ry •rj-f UnnJ i fy fc» fcJotk number) The priiirary objective of this research effort was to develop data-hand i ing procedures to transform digital data collected by a Bendix 24-channel airborne sensor into radiat.cJ^ values and to produce images free of skew and reflectance ’ ’ 1 disc r t ion. Data collected over the Chesapeake Bay trea n 26 I l' 1 /.' and 2.’ April 197 3 at approximately 3200 m and later recorded on ccmputor- compat ihlo taper. (C.t » wore studied. Special attention was focused or. data from Rappahannock Kiver. fhe scanner system, including the mechanics, optics, : (Continued) DO 1473 UITIOM OF I MOV AS IS OBSOLETE i j.l£ij - ifieu SLCURITV classification of This race ; Data Fntatatff i Uric 1 ,i y s j I ted SECj«|Ty CLASSIFICATION Of This P»li[,»li«n 0*1* In:.n<; 20. ABSTRACT (Continued). and electronics, is described with an explanation of data formatting on the NASA-generated CCT and the formatting required by existing 0. S. Army Engineer Waterways Experiment Station (WES) compucer software and programs to handle re¬ motely sensed data, specifically CRTS or LANDSAT CCT data. Two critical correction problems were encountered, scanning geometry and aircraft attitude. Correction procedures are presented. Those were incorpo¬ rated into a system for 24-channel CCT data conversion by a small computer, FDP 15, and image preparation on an Optronics film writer. \ procedure for converting CCT data to radiance at the earth surface was developed, maintain¬ ing the relation of radiance to pixel source. Performance problems in the Bendix 24-channel sensor system prevented the completion of an effort to use radiance at the earth surface as a tool to identify varying conditions in or on tie water. However, the procedure was developed and has been automated. It is included as an option in the 24-channel data conversion system using the small computer and film writer at WES. U I’nc l.i 'si i i od >.CC U»I T ~ C_ AySif Ol"S 1> P* of .0«fa [ + J THE CONTENTS OF THIS USED EOF ADVEHTISI NG , PROMOTIONAL PURPOSES. NAM . . 1 NOI :0N3T] DORSEMENT OR APPROVAL COMMERCIAL REPORT ARE NOT TO BE Publication, or CITATION OF TRADE irrn f A5TTOT ' ’ V• • - U l ... /--I of i ^ ^ . / vi^ : OF THE USE OF SUCH PRODUCTS. / PREFACE KRT S Project 281/282- "Hydrod T W±th data collected for y ^ riciiTi i. C Act i fine n _1 n 1 Sediment Concentrations, Chesapeake Bay R e gW -. under the In-House Laboratory Independent Research aUth ° rlZed lNo - 6.11.01.A, 4A061101A91D P *o 8 ram as Project P6U1 °1 S22 ° 7 ^ s P°nsored by the'omT^I.T was performed during the period A ’ ‘ hn 8 ln ecrs. The work ° e period Aucust 1071 t the Environmental Character^ *-• ^one 7 j by personnel of laracterization Br anch (fc n , Division (ESO), Mobility and Ervi ' ^ EnVlronment «l Systems l! * S * Arny Engineer Waterways SyStenJS Moratory (MESL) , general supervision of Messrs. ^ Grabau Special Research A- i ^ MESL ’ and W * E. , ^search Assistant and former Chief ren direct supervision of Messrs T r ’ ? ’ and und '”‘ Cne former Chief, ECS. y iss j, H ’ ’ " CL>U ’ Chle ‘’ ECB • and R - R. Fries*, report was prepared by ^ ^th "" ^ Pr ° jCCt ^ COL G.H.Hilf" rr r,, ’ CL, was Director of the utc a report preparation. Mr F R w Urin g thls study and 1 " BTOm »« Technical Director. I CONTENTS PREFACE P age 2 CONVERSION FACTORS, METRIC (SI) TO U. S. CUSTOMARY AND U. S. CUSTOMARY TO METRIC (SI) UNITS OF MEASUREMENT . 4 PART I: INTRODUCTION.5 Background. 5 Objectives and Scope. 9 PART LI: DESCRIPTION OF 24-CHANNEL SCANNER SYSTEM . 10 Mechanics, Optics, and Electronics.10 Data Tape Format.12 PART III: RESEARCH PLAN.14 Introduction.14 Anticipated Problem Areas . 15 PART IV: DEVELOPMENT OF DATA HANDLING PROCEDURES.18 Data Conversion Software.18 Test Site Selection.18 Data Transformation.19 Water-Land Discrimination . 19 Voltage-to-Radiance Transformation.20 Image Geometry Correction . 21 PART V: CONCLUSIONS AND RECOMMENDATION'S.24 Conclusions.24 Recommendations . 24 REFERENCES TABLES 1-3 FIGURES 1-24 PLATES 1-2 26 w-' CONVERSION FACTORS, METRIC (SI) TO l’. S. CUSTOMARY AND U. S. CUSTOMARY TO METRIC (SI) UNITS OF MEASUREMENT Units of measurement used in this report can he converted as follows: _M ultiply _ By _ __ To Obta in_ Metric (SI) to U. S. Customary nilimicrometres micrometres mill met res met res square metres cubic metres knots (international) mi 11iradians radians per second Kelvins incites feet degrees (angular) 3.280839 x 10 3.280839 x 10 0.03937007 3.280839 10.78391 6.102376 x 10 1.151543 0.05729578 57.29578 1.8 -6 -3 feet f eet inches feet square feet cubic inches miles (U. S. statute) per hour degrees degrees per second Fahrenheit degrees* U. S. C ustom ary to Metric (SI) 25.4 0.3048 0.01745329 millimetres metres radians btuin Fahrenheit reading: fro- Kelvins, u.- F = 1.3(K - 273.15) *4 FEASIBILITY OF MONITORING FLOW PAT TERNS AND SEDIMENT and pollutant dispersion oi- water bodies" WITH 24-CHANNEL SPECTRAL DATA PART I: INTRODUCTION Background 1. Within the la9t several years, new iiationa^ priorities have made necessary the acquisition and analysis of environmental data on a scale not contemplated heretofore. The regions being considered are so large and the environmental data requirements are so universal that conventional data acquisition systems are no longer adequate to meet the demands. Even systems based on aerial photography and manual interpre¬ tation are inadequate in many situations, because processing is too time-consuming and costly. 2. In an attempt to meet this challenge, personnel of the U. S. Army Engineer Waterways Experiment Station (WES), sponsored jointly by the National Aeronautics and Space Administration (NASA) and the Army Corps of Engineers, developed largely automated procedures for using the digital image data acquired by the Earth Resources Technology Satellite (ERTS-1) to obtain certain kinds of environmental data on a regional scale. In general, the procedures based on ERTS-1 data made it possible to very rapidly produce accurate and reliable maps of open-water surfaces and of the concentrations of suspended materials in the surface and 1 2 near-surface waters. ’ Since the analytical procedure for mapping the concentrations of suspended materials depended upon the use of the spectral data defined by the four ERTS-1 spectral bands, it was believed that similar procedures could be used to map the distributions of any other features of the landscape that exhibited unique spectral "signatures. Such features night be expected to include land uses, crop inventories, a limited amount of plant species discrimination (e.g. forest types, aquatic plant species, etc.), certain kinds of topographic expressions, and so on. Thus, a general analytical procedure for exploiting 5 spectral information would be capable of supplying a significant part of the environmental data required for Corps planning purposes. 3. Despite the fact that the ERTS-1 data proved to be extraordi¬ narily useful in many contexts, it was clear from the beginning thac there were situations in which the data would be inadequate. There are 3 two primary reasons. First, the ERTS (and LANBSAT*) pixel is so large (approximately 56.8 by 78.7 m**) that small terrain features are obscured. Since many items of interest to the Corps of Engineers, such as small changes in the positions of shorelines, are much smaller than the ERTS-1 pixel, the satellite view would not meet all Corps data requirements. 4. Second, the ERTS-1 nultispectral scanner divides the visible and near-infrared spectrum into only four relatively broad bands (i.e. 0.5-0.6, 0.6-0.7, 0.7-0.8, and 0.8-1.1 pm). This means that the ERTS-1 spectral data cannot be used to discriminate between two features that exhibit spectral reflectances that differ only within one band. For example, two tree species might exhibit identical spectra as sensed by ERTS-1, but a close examination of each individual spectrum might reveal that the energy from species A was concentrated in the 0.5-0.6-pra range at 0.51 pm, while that from species B was concentrated at 0.56 pm. S5r.ce this kind of situation is common, it seemed obvious that a multispectral scanner that broke the visible and near-infrared spectrum into narrower wavelength bands would provide additional discriminatory power., 5. In view of the factors discussed above, the conclusion was reached that great long-range benefit would be achieved if the concepts developed during the ERTS-based research (see paragraph 2) could be modified and expanded tc use digital data obtained with instruments such as the aircraft-mounted, NASA-operated, 24-cbanneI Bendix scanner system. In the development of such a capability, three general problem areas were anticipated. * Formerly ERTS. ** A table of factors for converting metric (SI) units of measurement to U. S. customary units and U. S. customary units to metric (SI) units is given on page 4. 6 a. The ERTS-1 (and LANDSAT) and Bendix 24-channel scanner* digital computer-compatible tapes (CCT) use quite different data storage format^, and thus new translation algorithms would be required in order to read the 24-channel data, b^. Because of differences in sensor operating altitudes and mechanical system arrangements and scanning geometry, the distortions inherent in the recorded data are of different kinds and degrees. Thus, algorithms developed to rectify ERTS-1 data would have to be modified, and in some cases, entirely new algorithms would have to be developed, c. The spectral bands recorded by the BMSS are quite different from those recorded bv the EMSS, and the calibration arrangements by which sensor output voltages can be trans¬ formed into radiance values are quite different from those in the EMSS; thus, new analytical algorithms would have to be written, even though the general principles would almost certainly be similar. 6. Once it became possible to reliably manipulate the BMSS data so that geometrically acceptable images using radiance values could be oroduced on a routine basis, the way would be open for the development of spectral analysis procedures. Even though difficulties encountered in manipulating EMSS data suggested that time and funds for the BMSS Study would not be adequate to attempc an actual experiment in spectral analysis, attempts were made to anticipate difficulties that were likely to occur in that area, in the unlikely event that the data manipulation problems required less time than expected. 7. As an example, it was anticipated that the very great differences in the sensor altitudes and scanning geometries of the EMSS and the BMSS would make it impractical to use the same eorrelc' ion relations between radiance and suspended materials concentrations t r * To avoid repetition, the following acronyms will be used: EMSS = ERTS-1 roultispectral scanner; BMSS = Bendix aircraft-mounted multispectral scanner. 7 - the BMSS data as were used for the EMSS data. Three major factors could be expected to contribute to the differences. a^. The pixel sizer, of the two systems are quite different. 2 The EMSS pixel has an area of about 4470 whereas, the BMSS, flown at an altitude of about 3000 r, has a 2 pixel covering about 36 n along tne ground track of the flight line. Thus, the BMSS would "see" much smaller features than the EMSS. For example, spatially complex features like eddy systems in water may not be detected ir. EMSS images but may be detected easily in BMSS Images. jo. The BMSS scans from -40 to +40 deg with respect to the perpendicular from the scanner to earth, -o that the angle of the optical axis of the pixels varies through a range of 30 deg; the range for the EMSS is less than 10 deg. Thus, the reflectance geometries (i.e. the angular relation, among light source, terrain surface, and sensor) of the two systems would be quite different. The amount of energy reflected by the same suspended material c'-noen- trations in different parts of the BMSS scan swath could well be different, while that of the EMSS could be treated as a constant. There was, therefore, rer..-w table doubt that simple correlation relations, as used for EMSS wata, would be useful for BMSS data. Because of the differences in scanning geometries ana sensor altitudes, the optical paths taken by energy cuanta in the two system- are quite different. The effects of this were, difficult to anticipate. 3. Tr. view of all of these differences and uncertainties, there • as considerable :c ibt that an automated interpretation based on EMSS .1 v •eld close 1- resemble one made with EMSS !nta. n i Objectives and Scope Objectives 9. Primary objective . The primary objective of this research effort was to develop data-handling procedures such that BMSS digital data could be transformed into radiance values and used to produce images free of skew and reflectance geometry distortions. As a practical natter, the objective was to produce the best image possible, from the BMSS digital data using information regarding flight and sensor conditions recorded at the beginning of the NASA-prcvided CCT. 10. Secondary objective . The secondary objective, to be pursued only if time and funds permitted, was to develop computer software that would make possible the mapping of the distributions of spectrally definitive landscape features, such as concentrations of suspended materials in water, with BMSS data. Scope 11. Tire research effort v;as confined to the use of existing BMSS data and to the use of a PDP-15 computer for basic data manipulation. Essential peripheral equipment -was restricted to an Optronics film writer and conventional input-output devices, such as line printers, teletype terminals, and tape and disc driver. 12. Test areas were restricted to the estuaries of the Choptank River in Maryland and the Rappahannock River in Virginia. Data from only one flight line for each estuary were used in the development of computer procedures. PART 11: DESCRIPTION OF 24-CHANNEL SCANNER SYSTEM Mechanics , Optics, and Electronics 13. The NASA-operated BMSS”* consists of an aircraft-mounted nultispectral scanner and a ground-based data analysis system. The BMSS is an imaging spectrometer that is mounted in a NC-130B aircraft (Figure 1). The Data Acquisition System (DAS) is a ground-based system that processes and rerecords the spectral data on computer tapes and also provides an imagery display of the collected data on a viewing cathode-ray tube (Figure 2). 14. The BMSS scanner collects energy in the spectral range 0.34 to 13.0 um, which includes the near-ultraviolet, visible, and near- infraivd regions of the spectrum. The spectral energy is separated into 24 channels (or wavelength bands) by two grating spectrometers sharing a common scan mechanism. The scanning mechanism is a flat mirror that scans the earth through a slit in the ‘nitton of the scanner and an open door in the bettor, of the aircraft, lie mirror rotates counterclockwise with respect to the aircraft heading providing left-to-right pixel scanning, while the forward motion of aircraft provides a sequential line-by-line scan oi the earth below. The scanning mirror views the eart:■* through 80 deg and completes a 360-deg rotation, during which it senses four calibration sources (Figures 3 and -). During che earth, scan the reflectance from each of 700 pixels is processed, and the response of each of 24 sensors to a pixel reflectance is recorded on r magnetic tape. According to the NASA training manual, the calibration sources are expected to permit calibration oi the spectral data to a high degree of accuracy. A precise vitlo of aircraft velocity (V) to height (H) , V/l!, must b< tai tain e non < relation, i.e. adjacent scan lines just touching each other along the ground track 1 nter if t 1 ;< • wath . i pi Ld< thi normal an, the scan .mot or speed i s .. scablished in terms of Y/h. 10 \ 15. The scan mirror is set at a 45-deg angle with respect to the aiicraft longitudinal axis. The scan mirror reflects radiant energy from the ground and from the calibration sources to a Dall-Kirkham tele¬ scope. A roll-rate gyro is used to maintain the scanner field-of-view (FOV) constant regardless of aircraft roll (within +3 deg). The Dall- Kirkham telescope transmits the reflected energy through a system of mirrors and a 0.2286-cm aperture onto a series of dichroics and grating spectrometers, which separate energy of wavelengths less than 2 pm from those of 2 pm and greater (Figure 5). The aperture is dimensioned to define within the center of the received energy field a FOV of 2 miad. Figure 6 shows the path of the short wave reflected energy to reach the detectors. 16. Scanning in a left-to-right rotation, the mirror first scans the earth surface, then the calibration source elements in this order: 16 calibration elements from sensor exposure to the calibration instru¬ mentation housing (not used), 16 from the UV-VIS-IR (ultraviolet-visible infrared) integrating sphere, 16 from the sky radiance tube, 16 from the high-temperature blackbody, and finally 16 from the low-temperature blackbody. Figure 7 is a simplified block diagram of the complete system. Figure 6 shows the calibration grating and tining sequence in the scan rotation. The UV-VIS-IR integrating sphere establishes the maximum radiance, and the low-temperature blackbody the minimum railiar.ee for the scan. 17. The same detection procedures are applied to radiance from calibration sources as fro:; earth elements; therefore, the calibration radiance value fer a scan and the recorded pixel radiance from each channel are on the same basis. Figure 9 illustrates the reflectance from the scan pixels as the earth is scanned through 80 deg and the corresponding reflectance for each channel in the calibration sources as tne 360-dog rotation is completed. 18. The UV-VIS-IR sphere (Figure 10) is a 16-in.-diam aluminu" - sphere painted with high-reflectance paint and illuminated from the side bv a 2C0-w tungsten halogen lamp. The scan mirror views the illuminated sphere through a large, plain optical glass window and "sees" a 11 diffused surface of supposedly known spectral radiance. The radiance emitted from the sphere covers the range 0.3 to 2.5 urn and ia calibrated prior to flight to a standard lamp (Q3 #11) of known radiance per wavelength band (Table 1) . 19. The sunlight-matching part of the UV-VIS-IR calibration contains a filter wheel with eight different: neutral-density attenuation screens, which is located between the lamp and the UV-VIS-IR sphere and provides a means of changing the spectral intensity to natch the sunlight conditions during a data run. This equipment was not operational during the data runs investigated in this study. Data Tape Format 20. The 24-channel data are in the universal format as defined by NASA. The 2A channels of data are digitized into 3-bit words and recoiled on 12 tracks of a 14-track tape recorder. Two consecutive •hansels of scanner data are interleaved onto one tape track; the first • referred to as the odd channel, and the second is referred to as • avn. Calibration sources, aircraft information, such is altitude and h’c it’.on, and other housekeeping information, are multiplexed together on' recorded do the 13th track. The 14th. tra_k contain, the time code is. lata .: • lirect-ri rd< Li anehestei -phase-L) code. Nine-track CCT containing these data were requested lro::. record contains the information in the house- The CCT non and to- auxiliary data annotation set (ADAS/ASQ) recorded on the ori ;:unl tapes (figures 11 and 12). A program was written to present there data in a rare usable form, (see paragraph 31). An example is •.haven Ln Table 2. Note the channel order of data requested by the user is al odd ehinnuls in sequence followed by all even channels in sequence (hr i tetel group in Table 2). This is the order of data on CCT received - r->r NASA. Ft j 1 owing the header record are the requested data scans. ntaii 7QC : imples, foilove3 by BO calibration samples ^ *> for each channel (Figure 33). Data are interleaved by bands in each sample, two channels to a data word. 22. For seven channels, each scan on the CCT contains 5460 8-bit bytes of data (43,680 bits). The 80 calibration samples are from the four calibration sources (see paragraphs 14-16) plus a spare (16 samples each). Data from two of the calibration sources, the integrating sphere and the low-temperature blackbody, are used to establish the range of reflectance reception by each channel at the time of the scan. 13 PART 111: RESEARCH PLAN Introduction 23. When solar energy strikes the earth's surface, it is reflected, absorbed, and combined with energy emitted by the earth (Figure 14). As it passes through the atmosphere, solar energy is subjected to scattering and absorption, thus limiting the amount of radiation incident upon the earth. The set of curves in Figure 15 shows the varying amounts of absorption at different wavelengths of the spectrum. The top curve, m = 0, is the spec'ral curve of irradiance before it enters the earth's atmosphere. The m = 1 curve describes the solar irradiance at sea level when the sun is directly overhead. The other curves illustrate the situation at various sun angles. The figure shows that the larger the sun angle, the greater is the attenuation of energy. This results from the longer paths through the atmosphere; the closer the sun is to the hori-on, the greater is the amount of atmosphere that the rays must penotr ite. Figure 16 shows radiance transmission through the atmosphere, fhc dips show the areas in which absorption has taken place due to water vapci , carbon divide, ozone, molecular oxygen, and Rayleigh and aerosol •: nattering. J The modal wavelengths of each of the 24 channels (or wave1engtl bands] ir« indicated in th< figure and are iefin< in Table 3. Vote thac 5 of the 24 channels fail within the visible spectrum, and that : 1 cover the entire range of visible and near-visible wavelengths f 0.35 to 1.1 'r..). Ibis study :s : ders channels 4-10 which closely parallel the run.e of wavelength band a received by the ERTS sensors (Tabic a) . 25. Clear water that is free from suspended sediments reflects -ost of the blue-green and reen portions of the spectrum and absorbs the other visible radiant energy. In the near-infrared portion of the ♦ectrum (0.8-1.1 cm), air energy is • rbed by water. A record o r • i me ror the earth suriice in this portion of the spectrum can define approximately -he land water :r.te-fare of a water body. The \ resolution of the interface is determined by the size of the ground element (pixel) defined by the field of view of the scanner; and since the total radiance from a pixel may be partly from water and partly from nonwater or from wetted soil, a clear definition of the water's edge is not possible. Although absolute definition is impossible, good approxi¬ mation can be made within the dimensions of two pixels in length and width. 26 . Suspended material in water causes changes in the amount of reflectance and in the variety of reflected spectral energy, i.e. the amount of energy at various wavelengths. Theoretically, these changes are associated with th* amount and kind of suspended material. This theory' was supported by WES investigation of data obtained from ERTS-1 2 overpass in the Chesapeake Bay. 27. Considering these facts and theories, a research plan was developed which fell naturally into two parts which are illustrated schematically in Plates i and 2. Plate ] describes the plan to convert 24-chanr.el data recorded on NASA CCT to WES computer program formats and to write, analyze, and correct geometric distortion when images are written from that data using the WES film writer. Plate 2 describes the plan to develop the analytical procedures required to map spectrally defined regions. In this context, a spectrally defined region (i.e. a patch of landscape that displays a unique combination of radiance values in seme selected set of wavelength bands) is one that has a specified concentration of materials suspended in water. In the plates, tne blocks tracing the various stages are numbered sequentially; the subsequent discussions are keyed to the relevant block numbers in the plates. Anticipated Problem Areas 28. The BMSS scanning geometry is quite different from that of the EMSS and, in consequence, quite different kinds of geometric distortions were expected to occur (Block 12, Plate 1). The anticipated distortions are of two major kinds. 15 £. Scanning geometry . The aircraft was flying at an approxi¬ mate altitude of 3050 m during the missions used in this study. Since the scanner swings through an arc of 80 deg, the ground swath covered by the scanner is about 5118 m wide. Since the solid view angle of the scanner is fixed, the pixels at the edges of the scan swath will be much larger than the pixels along the center of the scan swath. The Optronics film writer, which is used at WES to write images from digital data, writes a square pixel of fixed size, and thus a direct transcription of the BMSS data onto fi'ji with the film writer will result in a narrower swath than the true scanned swath. It was clear that this distortion would have to be removed, b. Aircraft attitude . While aircraft roll was automatically compensated for by on-board platform stabilization, aircraft yaw, if any, could be expected to introduce skew into the image. Since it was almost certain that there would be some wind from the side, the aircraft heading would diverge to some degree from the line of flight. A procedure would be needed for transforming the BMSS tape data arrays to correct for this effect. 29. The calibration relations by which transducer voltage values (digital form on (XT) can be transformed into radiance values (Block 10, Plate 1) in the BMSS system are different from the relations used for the EMSS system. Thus, new calibration transforms would be required. 30. One major problem was that there was r.c possibility of deter¬ mining the actual suspended materials concentrations in the waters of the test areas during the time of the aircraft overflights. This would have required teams in hosts collecting water samples, and this was far too costly to be practical. As a result, there was no possibility of establishing an empirical relation between suspended materials concentration and radiance (Rlock 18, Place 2) as had been done previously for the 2 studies using the EMSS. The best that could be hoped for was a 16 V \ . v. / relation that described relative differences in suspended materials concentrations, based on the fact chat low concentrations of suspended 2 materials appear to produce direct linear correlations with radiance. 17 ■ -y~ -■ \ PART IV: DEVELOPMENT OF DATA HANDLING PROCEDURES Data Conversion Software Blocks 1 and 2, Plate I 31. With the help of Reference 6, a computer program was written that decoded the BMSS CCT and reassembled the contained data into forms more readily usable by WES computer hardware and available software. Several modules were involved. For example, the NASA tapes contain housekeeping data (Figure 11), aircraft location and attitude data (Figure 12), radiant energy data from terrain, and calibration data. These data were all reformatted so that they could be printed out in a more readable form, as illustrated in Table 2. 32. Another module was designed to transform the scanner transducer data (i.e. the output of the sensors that measure the amount of energy received by each BMSS wavelength band) into the format required by the WES Optronics film writer. This format is illustrated in Figure 17. The general principle is that a new tape is devoted to four wavelength bands or dummies so that, each wavelength band can be processed independently with existing WES software. Test Site Selection Blocks 3 and 4, Place 1 33. In the prior work with ERTS-1 data, maps were produced of suspended materials concentrations in several estuaries leading into 2 Chesapeake Bay. Of the several available, the Rappahannock and Choptank estuaries were chosen. The P.a p pah anno ck River estuary was chosen because it has sources of relatively uniform suspended materials and a small tidal range. For contrast, the Choptank River estuary was chosen because its sources of suspended materials are quite complex, and its _idal range is somewhat greater than that of the Rappahannock. The locations of the selected study areas are shown in Figure 18. EMSS tapes covering these areas were ordered from NASA. 18 Data Transformation Blocks 5, 6, and 7, Plate 1 34. Once the programs to read and reformat the NASA BMSS tapes had been written, the NASA tapes covering the BMSS records for the two test areas were reformatted (Block 1) into WES Optronics film writer format (Figure 17). At this stage, the contained data have not been transformed in any way. The data relevant, to scanner pixels are still raw digital values defining transducer output voltages, and the array is a simple and rigid raster array. If used to drive the Optronics fiSm writer, the resulting image would include all of the distortions intro¬ duced by scanning geometry, aircraft yaw, and so on. 35. In addition, all housekeeping data (Block 6), calibration data (Block 7), and so on, were extracted and assembled into tables as illustrated in fable 2 . Among the critical information are those, items describing flight parameters. Of particular importance are the tine, location, radar altitude (which is a measure of absolute height above the terrain), true heading, drift, and date. The significance of these parameters will be discussed later. Water-Land Discrimination Blocks 3 and 9, P late 1 36. Because the secondary objective (see paragraph 10) involves the distributions of suspended materials in water, it was obviously advantageous to delete .ill nenwater areas fren the reformatted BMSS tapes, since this would result in a significant saving of processing time. To achieve this, the water-land discrimination program developed 2 tor use of EMSS data was used without modification. When used with EMSS data, the discrimination between water and land is made with the rear-infrared band (ERTS-1 channel 7; 0.8-1.1 un) , since ail energy in tuts wavelength band is nearly totally absorbed by water, while it reflects strongly from vegetation, soil, and other "land" materials. 19 I X • \ \ 37. The properties of F.RTS-1 channel 7 are most closely approximated by those of BMSS channel 10 (0.981-1.045 yin. Table 3), and this wavelength band was therefore used to define the pixels relating to open water surfaces. This information was then used as a "digital mask" (see procedure described in Reference 2) to edit out all "land" areas in all of the wavelength bands used in this study (Block 8) (3MSS channels 4- 10 ) . 38. Despite the fact that the imagery resulting from writing the raw tapes is badly distorted, it is helpful to write an image of each wavelength band (Block 9), simply to confirm that all decoding and reformatting procedures have operated correctly (Figure 19). Even an uncritical comparison of the image with a map of the corresponding area (Figure 20) ..-ill reveal that the geometry is badly distorted. Voltage-to-Radiance Transformation Blocks 10 and 11 , Plate 1 39. Using the reformatted data and the calibration data contained on the EMSS tapes, a program was written that transformed the values recorded by the BMSS into radiance values. Since calibration data varied considerably, no attempt wan made to develop a conversion equa¬ tion in which constants were applicable to the entire data set. Instead, a conversion based on the data from calibration sources from each scan was derived and applied to that scan only. D — p ix i1 - S5 1 1 (RSL.) (B ) r rH 1_ cos SA “ij S2j - Pj T where i * pixel in a scan i J = channel, i.e . channel *"*■ 9 5, 6, 7, S, 9 , or 10 20 R Pix ij i.1 S5 S2 j j RSL. B . J T SA = radiance at earths surface of i pixel in channel j = recorded digital value of reflectance from i^ pixel in channel j = average of 16 samples from the 1ov-temperature blackbody for channel j for a scan *= average of 16 samples from the integrating sphere for channel j for a scan = percent reflectance with which the integrating sphere compares to the standard lamp for channel j = spectral radiance evaluation of the standard lamp per micrometer for channel j = channel width of j channel = atmospheric transmittance factor - sun angle from nadir (perpendicular from sensor to earth) For each data collection run. (o^) ((RSL^)(B^)) (I/O (cos SA) for each channel is constant. • Image Geometry Correction B locks 12, 13, 14, and 15, Plate _1 40. Two major kinds of geometric distortions occur in the BMSS data; namely, that caused by aircraft crabbing or yaw (i.e. the angular relation between aircraft heading and the ground track of the aircraft) and that caused by the scanning geometry (see paragraph 28a). 41. The procedure used to correct for aircraft crabbing is illustrated in Figure 21. Line A'B' in Figure 21a indicates the actual scan relation to the center line (i.e. the true course of the ground track of the aircraft) of the scene. The orthogonal grid array of data on the BMSS tape places the scan line at 90 deg to the scene center line (line AB in Figure 21a and b), which causes a misplacement of features as indicated in Figure 21c. The displacement at each end of the scan can be computed from the difference in the true heading and 21 true course. By staggering blocks of pixels across the scar, in a •diagonal array down scans (Figure 21d and e) , features aie adjusted to their normal relation to each other and to the center line of the scene. The geometries illustrated by Figure 21n-e were reduced to a computer program that rearranges the pixels in an orthogonal grid such that, when written with the film writer, the skew created by aircraft crabbing is eliminated. An example of this type of correction is given in Figure 22. Note that alter the correction, the ends of the views are not at right angles to the center line. 42. A quite different procedure is required to compensate for the variation in pixel width ns the instantaneous viewing angle is removed from ti.° center J ine (Figure 2’i). At nadir the instantaneous FOV (2 nrad) is 6.1 m (20 ft) wide when the aircraft is flying at an altitude of 30 4.6 a (10,000 ft). The 24-channel Forth observations Aircraft Program flight over the Rappahannock on 22 April 1973 vas flown at -A O ~ (id,550 ft), which defines the oixe width at nadir as 6.6/ ft), bach .successive \ • .-.el is .:jder as th*. scanning angle from a r. an no :nereises pixel viewed at the scan extreme is of -0 deg the. center line and is 11.31 -. vice, ■ re ' .: the at fix ■ taj lata tc Lnd ti the patia] ri itions between th< :arth pixels from which data ;. ,o re drawn, some Lve proced >plied tii rrays was necessaiy. To s: r<.-tch the scan >rrcy a. data, pixel radiance values ware repeated at hese points ! • ring the difference in distances from the nadir point when computed using th.o tangent of the scanning angle and the altitude of the sensor and when computed using the number of pixels from nadir and the width of the pixel at nadir. ..lie:' the difference in these distances exceeded the width cf the pixel . ' radii aJ le wa repeated expanding the number of pixels in the scan. This expanded scan pushes the scan extremes to more nearly coincide with the earth location sensed by the extreme instantaneous r, . v '. In Figure 2J, pixel 75 has been repeated. The farther, away from the renter : in.:, th? more often pixels are repeated. In the Rappahannock. 2? I digital nap, 70 pixels were of this situation were also of a scene before and after Figure 24. Note that after than before correction. It to write an example of the the programs have operated added to each half of each scan. The geometries reduced to a computer program. An example correction for scanning geometry iB given in correction the scene is significantly wider is usually helpful, as illustrated in Figure 24, scene in each wavelength band to verify that properly. 23 PART V: CONCLUSIONS AND RECOMMENDATIONS Conclusions obj ecti ve •'*3. The primary objective was achieved. A set of computer programs is now available that transform. the NASA-provided BMSS CCT into an image formed of radiance values and free of both skew and scanning geometry errors. All programs are run « n a ILU-13 computer, and images are produced directly with an a routes film v L'lSS tapus can be used effectively by agencies computer facilities to create pictures of orco¬ lter . Thus, tile NASA having only very modest of the earth's surface and its radiance.. d ar> object!ve • - . • * : *; i ■ . ' { . o - j i li ce, but the rh has n>: Tiie re a;Its of this .ffoi t vi paper . Some sign if imut bse* •. seamed data that have l wen ma .a live need further study. There (-cue r o changing slope anile due re the property of the vote fact' rs, such, as bottom reflect that reflector ess h m. bo:en ma hr toward the s b: '.A' to i ■ rssfu] •. • ■ t« ' . hopeful •, in : iti • - about s rortions ini' in t . • pursuit ' •hi secend is the problem of increased r t ho ■ r ice'' , which over!ays v itself. There are other dis r.nce and the d.pths to T h-: bo: ecor.dc. ry n 1 usion. *N ■» ■' ' t .* mutely ary ofcjec- adian.ee the radiance tor tion torn at • uk a: • 1 • It is recommendei st Lht in-hand 2^-channel data inti rue in an effort : dev€ 05 prer lures to reduce distortion caused by c-xtranevtis radiance, which ...his to the radi .nee due only to the .■■report ies ■ tive as aigt equip'' c-nt >.53- . • - • . i I The Bendix 24-channel sensor has been abandoned In favor of the Bendlx 11 -c.hannel sensor because of the nore consistent performance of components less sensor "noise", and better control of the internal flight environment The channels investigated in this study are included in the Bendix 11- channel sensor equipment with the same channel wavelength ranges and same type receptors. It is recommended that this study be repeated with existing data collected by the Bendix 11-channel sensor over water bodies. 25 REFERENCES Williamson, A. N. and Grabau, W. E., "Sediment Concentration Mapping in Tidal Estuaries," Paper, Third NASA Seminar on ERTS-1 Studies, Dec 1973. Williamson, A. N., "Movement of Suspended Particles and Solute Concentrations uith Inflow and Tidal Action," Technical Report (in preparation), U. S. Army Engineer Waterways Experiment Station, CE, Vicksburg, Miss. Grabau, W. E. , "Pixel Problems," Miscellaneous Paper (in preparation), U. S. Army Engineer Waterways Experiment Station, CE, Vicksburg, Miss. NASA/Goddard Space Flight Center, "ERTS Data Psers Handbook," 1972, Greenbelt, Md. National Aeronautics and Space Administration, "NASA 24-Channel Multispectral Scanner System Training Course," Information Systems Division, Manned Space Flight Center, Houston, Tex. National Aeronautics and 5 : -nace Administration, "Earth Resources Data Format Control," PHO-TR543, Vol I, May 1973, Manned Space Flight CenterHouston, Tex. Table Calibration Data for Standard Lamp (QB ’( 11) Wave length Spectral Radiano um w/sr/pm/cm- (1) 0.325 (1) 2.65 X 1 o r—i (2) 0.350 (2) 6.48 X io" 4 (3) 0.375 (3) 1.22 X io" 3 (4) 0.400 (4) 1.89 X I o rH (5) 0.425 (5) 2.87 X IQ' 3 (6) 0.450 (6) 3.95 X 10~ 3 (7) 0.475 (7) 5.07 X io" 3 (8) 0.500 (8) 6.25 X I o r—1 (9) 0.525 (9) 7.35 X io' 3 (10) 0.550 (10) 8.16 X io" 3 (11) 0.575 (ID 9.38 X 1 O r—\ (12) 0.575 (12) 1.01 X io" 2 (13) 0.6375 (13) 1.11 X 10 " 2 (14) 0.675 (14) 1.19 X io -2 (13) 0.700 (15) 1.22 X 10 (16) 0.725 (16) 1.27 X io" 2 (17) 0.750 (17) 1.29 X io" 2 (18) 0.775 (18) 1. 30 X io -2 (19) 0.800 (19) 1.33 X io" 2 (20) 0.900 (20) 1.25 X io" 2 (21) 1.0 (21) 1.22 X io" 2 (22) 1.1 (22) 1.13 X CM 1 O H (23) 1.2 (23) 1.02 >( in" 2 (24) 1.3 (24) 8.91 X io" 3 (25) 1.6 (25) 5.33 X io” J (26) 1.9 (26) 2.78 X io" 3 (27) 2.2 (27) 1.28 X io" 3 (26) 2.5 (28) 5.41 X io" 4 s Table 2 Information extracted From Rappahannock River Tapis Record Sire 156 Nc. of Elements 700 Channel No. 5 Channel No. 7 Channel No. 9 Channel No. 4 * Channel No. 6 Channel No. 8 Channel No. 10 _ Scan StatuG 255 Starting Scan Line 66351 Time 16:46:15.3 Latitude N37 deg 37.3 rdn Longitude W 76 deg 28.5 min Radar Altitude 10950 Barometric Altitude 36800 True Heading Deg. 285.5 Drift Deg. L Pnan Course 1.4 Roil Rt. Wing D Deg. 0.7 Pitch Nose D Deg. 0.2 Ground Speed 271 kts Late 4-22-73 Mission 230 Site 244 Line Run 1 Volta Chan 3 ) 1 Figure 7. Simplified basic multispectral ser.sor function diagram V\C 0 VP k NCO0C* ' • T $ IUIM rCAL< GATES BACKG ../JND GATES (NOT USEOi CAL R(AD CVGS JAG ppr SPATE f . V'VlS ca'fs J S-vV Rif 'fS < - JL JL ;• . * \ '. n (!'.\0 i i (— SOT \(>S J L -o+ ^ ^ »■* *• • 3*} Figure 9. Composite spectral input scanning optics 5 38 < to J^rating sphere, UV-VIS-IR (ultraviolet-visible- i :'.i rated) in the Bendix 24-cbannel nultispectral sensor systen;-’ t'RAMK SV.C SCAN Si AH’S STAKE •«— SCAN LINE COLKI —*•-«-A DAS / AS‘j- '*0 DATA FORMAT 101011_ 10010000 > [WD l I WD 2 j WO 3 I WD A J WD 5 »N oc J eT O' rx ec *_3 ►J cT I 1 °i to o I l°i CO O’ I* I i cx ^ O I I ! — co O n !f a: O - O' *-f [ ef j c* iS 00 S 3 L,. Ol 1 - r gT £ • c e "i co i G~ o-l « 5 J g" -o .-T I Lf O • >r -3* -3 I u or i u: o ; . 31 Ol < CJ o o . 1 - * CL O -C 'J c- o cr. O < . v< 7 \J l _ „ t— CO CO o o o UJ O -•* nc o I — < o CL. o CO o t: i _. c to o U. O < o -- o CO G •O 10 X cr. g • < ° 1 c - o I ZD l_ to Q Z. o X Ut fi - H O o o UJ O DC O < o a- o CO O z c rj CL - S _H O uJ C G OC O rg < O c- o CO O o i. O' — rL 5 -H- O o o o two — CL O - < O a, o __cr — ir g o to o DC O < o sC CL C_> — Cl C — o o ^ O ^ r o o o o to CD -i oc o — < G — C.. O CO CD coo' oh. ;■ > rtO I Figure 11. BMSS housekeeping frames \ ADAS Wl) NO | 7 | 8 - 9 10 In] 12 ] l l l 2_j_3 _L ,4 5 1b! 7 1 8 1 13 ] 14 ] 15 J lb J 17 1 18 { 1 9 _1 10 _L_.11 J. 12 _13 | 14_1 15_ _L_ib __J [ 19 | 20 21 22 f 23 | 1 17 | 18 | 19 | 20 L 71 I 22 _] 23 _ ! 24 J 25 -RADAR ALTITUDE ( ?>. _ J ?i J 21 J 28 I 31 | 1 33 | 34 lNDrl v -«a-RAKOM ALT - I 33 j ?- j 35 ] '6 i _37 > I j 38 "j J ' I 26 1 29 I 30 -5S*»- 1 Nil FTC ORIENTATION 1 NT>EX ?_I 30_I 31 ! 32 35 3 6- n l K ItEADINC -£to- INDEX -it*- ; Ni' L _\ -el - 8 I 39 _j 40 J “1 " 1 42“ ] - 1 _ri_ 1 U L..\ J .. ,-VL 1 43 j 4A j 45 | i 1 -«E- ROLL - ! 44 | 50 | 51 j 52 53 i'Rli T I 46 ; 47 ! 48 J 4- | -/ j 4b j INDEX-<-i ITCH - 44 j 5J 1 S1 } s 2 - i>~ . 'OEX -C-* k.N I !:•:> . : RT VET I 56 J - 1 57 ] 58 J 59 I 60 55 j 5b j 57 i •. >• - 7 I It 61 58 i ■Air ►5 i 6b •> 7 61 I 62 j 6 3 | 7‘- ! 7 5 | 7:. | t> 7 j £ j b V • -*<3 --— . i *K-’ M ■: ^5 - 7’. ('* SI H2 I ;3 l 'E ! 74 | 7!) 0C> s? S i i» 8 j Vi i -> ! 1 ] ] . -i*— : Figure 12. ADAS/ASQ-90 packing in hoi sel eepir.g voi ds 41 < I / Figure 14. Interaction of radiant energy with matter OX(m) WATTS M PER MILLIMICROMETER Figure 15. Solar spectral irradiance curves at sea level with varying optical air masses^ l 001 R O •m w cd X) e * c ^ c x a u X o *-» X w) a. C vi V O fi a *-» > c cd 3 a CO i-J CO r: ^ CQ O &o co c o O T-« B ^ each OF THE foo.-s. Turn ■stTi Aa>t> c.icrA,i\e.T FfttlAftfC. G. L\VT »>ocb Of t>«TA EK- Tt^AC ted VMTH VTA- TlS flCb 'tr! ftePtftT AiiCwt with "We 8 TEST ChPV^ElsI S,T A rut) 9 POP.^. CuA-iSlFtitA- Ti&fo, T A.?e: ft p/A PvKEL. (J-) Pi»et /v-. ClP-SS'^A->sj Cotm 1 t-r»f I j-DlTYELO-p ICLSSS jL TEST d HAMELS 4, (» A^J-D 6 R>S> CUASS\F \C A — T\ OAJ , TAPE 8 ' 0(l)Cf 5 \te STEP 4 - Apply 24-channel program to correct for crabbing (essentially a skewing technique in the direction of line of "flight). A new tape is produced. STEP S - Apply 24-channel program to expand scan image producing a new tape. L7- WZS procedure, showing required data t’onr.ats, to produce a picture with, the Opt-ronics :’ilrn wri I / - CD OH C.ET H)j /v\PLCS- u 1 zano/cvc. s' . 10 1 F«lO 3 Cf«. 'O COD lr - j EXTRACT EPiO-H CHA tVfUE L. U)\TH OTT'O'O TO kH'JO L>5I/V<2> CH^.vVEL 10 Fo^ k^mj/uATeR. IruTSfc FAC a.' ALSO »PTvoW TO COAJv'efc.T TT> IXAtlAlJCE 4T EAa.ru CURPAce. CHAINES 6 OAvvei. 7 CWAA/WEk. "5 tA^ivet. io (TAPe BJ \ wcme.: cvue cji**wn- OfM e ACT TAPE I REPEAT OPEX.ATto -0 d&Ov£ 10 l TH TCPE" B Cv.aiJn: CHAIVKSSU <• CjiAAHVCk. 8 CP AKJfCCtC 10 C.H Mnc s 7 1 cmav^s c A tsAVJj ?l_ 10 CCfAPAfce EACH D\C.> 1 -A(_ VALde(p-UiS) TO A SET OF 15L. C5FA1 SHAT5ES AAJD pA.cboes a PlCTpiAE CM rue OTT8.0MC3. ■frm OF lESlft.GX> CKAftjWei. Data CH 4- 6 ChAVWITL «o STEP 2 - a. b. c. Rewrite tape data in format for ERTS programs on hand, masking land if desired. Apply 24-channcl algorithm to convert digital tape data to radiance at earth surface as new tapes are vrritten (if desired). Produce desired digital pictures from any channel data by coding the digital data (0^255) to 256 shades of gray on the Optronics film writer. ct_ (see FlGoAG 2b) TAr j C \ IJ KITE A??R0- PRjf\Tt GAAV «HATstL PoR. EAd-H •P\7.EU OdllVd. eG , r*x>.v>\eS Fil ^rrV^O G?Oft)NL 4ELOOTCP d,,\AN 'SiAki! AAA.H.T OF < b«A\-SPL.Af lX**// 4 .’•' ; ; ••*»" • ;’ , 1 .5 .-J. * ' - 1 ■ • • 'V- - - * i CHANNEL 7 ! 5 I * CHANNEL 8 r • - %*& *-V. ••*. 1«V •■ • <>‘ V !'..,••. • . " .V . .5.. . ... . _ : 1 »> -Llw;Laa.-.__ _ ■-^ - r - ' . &<*&.;■ CHANNEL 10 Figure 19. Film written from digital CCT data (0-255), Rappahannock River, Mission 230, 22 April 1073 4S< °h/rcn c\j l Q. < 2 U I a < ct o o Cl. O _i w z z < I o o z to D a < tr- ^ LU a:^ 5.0 _J iIh to < ua z J o< OTH a oo Jm o •H a o cj .c: d S rt tX n ►3 O CVJ a Li & •H V r 41 i Figure 21. Data correction procedure for aircraft crabbing V- ■* • c o u a o u V-* a lj VM to c3 d •r-4 ♦ .O o X) a d > _J *H U UJ CZ 4J z M-4 < U d X o o c d a y-f eJ ■H rC d d CL iw CL o d L_i CM CM V v-< D to •r4 Ui 51 < Flight altitude (a) Earth surface represented by tape pixels Figure 23. Correction procedure for registering scan extremes \ - u <1) Uj O q or CD u. o Uj z z Uj o z < c u, c f UJ z Uj u in G Uj u. O Uj fc a: ' o G Uj : K- u, o r CD z o & /.' I / ] * 4*' fr . T3 C d -* C O cfl U~i QJ CO o c •H O CJ •H c > u c LJ to Ti • 1 i q: •H r: 1 * o 1 u u< CT3 a. B CO C . <* i » cr o •ih -T Ik <’ .] f ■ Li H L. < o u* O “r-4 p V i o . U-< y | v 1 LJ u CO o c U i y> z c o fCS | < a \ ; J X a ■*J \ j i O jc u> rJ o a. Ul 3 Cu u V ro o * CJ s using iter \ y 54 < PLATE 1 Secondary Objective. Plan to P evelon Corputer Programs to Map Suspended Materials Concentrations with SMSS Data 55 < PLATE 2 In eeccrdfjsse with ER 70 - 2 - 3 , persgrapb 6c(l)(b), datad 15 Fabruery 1573» e facsimile catalog card in Library of Coojresa format Is reproduced below. Smith, Margaret H feasibility of monitoring flow patterns and sediment and pollutant dispersion of water bodies with 24-channel spectral data, by Margaret H. Smith. Vicksburg, U. S. Army Engineer Waterways Experiment SHration, 1976. 1 v. (various pagings) illus. 27 cm. (U. S. Waterways Experiment Station. Miscellaneous paper M-76-10) Prepared for Office, Chief of Engineers, U. S. Army, Washington, D. C., under Project 6.11.01 a, 4A061101A91D. Includes bibliography. 1. Chesapeake Bay. 2. Data processing. 3. Pollutant dispersion. 4. Rappahannock River. 5. Remote sensing. 6. Sediment. 7. Sensors. 8. Water flow. I. U. S. Army. Corps of Engineers. (Series: 'J. S. Waterways Experiment Station, Vicksburg, Miss. Miscellaneous paper M-76-10) TA7.W34m no.M-76-10 56 <