Currents and Pollutants - IEEE Xplore

97

IEEE TRANSACTIONS ON GEOSCIENCE ELECTRONICS, VOL. GE-15, NO. 2, APRIL 1977

Satellite, Aircraft, and Drogue Studies Currents and Pollutants

of

Coastal

VYTAUTAS KLEMAS, MEMBER, IEEE, G. DAVIS, J. LACKIE, W. WHELAN, AND G. TORNATORE

Abstract-The mounting interest in extracting oil and other resources from the continental shelf and continuing use of shelf and estuarine waters for waste disposal is creating a need for synoptic means of determining currents and monitoring pollutants in this area. A satelliteaircraft-drogue approach is described which employs remotely tracked expendable drogues together with satellite and aircraft observations of waste plumes and current tracers such as dyes or suspended sediment. Tests conducted on the continental shelf and in Delaware Bay indicate that the approach provides a cost-effective means of studying current circulation, oil-slick movement, and ocean waste dispersion under a wide range of environmental conditions.

INTRODUCTION

THERE EXISTS an urgent need to better understand estuarine and continental-shelf environments because of economic pressures to extract oil and other iesources; to increase the harvest of food; to continue using these environments for waste disposal; and to route ships effectively. The increased utilization caused by the large concentration of population in the coastal zone is likely to have deleterious effects on the shelf regions. The offshore-onshore transport rates of pollutants, sediments, and nutrients strongly influence the ecology of the coastal zone. In order to keep the environmental impact within acceptable levels, it is important to understand the circulation and exchange processes on the shelf and in the estuaries. The Eulerian method of simultaneously measuring the current direction and speed at preselected points in the water column requires expensive current-meter arrays and considerable ship time when synoptic measurements over large coastal areas are to be made. This is particularly true in the coastal zone, where the gradients of the flow field and other properties become large and vary rapidly. Considerable cost effectiveness can be attained if Eulerian techniques are complemented by Langrangian methods employing expendable drogues in combination with aircraft and satellite observations. Moreover, although in theory the Eulerian- and Langrangian-type data contain the same information and can be interchanged, Langrangian results are more suitable for interpreting the path taken by contaminants subjected to water movement. Therefore, a satellite-aircraft-drogue approach will be described which is based on the Langrangian technique and employs remotely tracked drogues and dyes together with satellite observations of natural tracers, such as suspended sediment, or Manuscript received March 24, 1976; revised December 15, 1976. This work was supported in part by NASA-ERTS-1 under Contract NAS5-21937 and in part by E. I. DuPont de Nemours and Company. V. Klemas, G. Davis, and J. Lackie are with College of Marine Studies, University of Delaware, Newark, DE 19711. W. Whelan and G. Tornatore are with ITT Avionics Division, ElectroPhysics Laboratories, Inc., Columbia, MD 21045.

the actual pollutant plume. This technique is currently being employed to study estuarine and shelf circulation, particularly at critical points such as oil transfer and waste dump sites where disturbances of the environment are anticipated. SATELLITE OBSERVATIONS OF SURFACE CURRENT AND SEDIMENT PATTERNS Current-circulation patterns have been studied remotely by time-lapse photography of current drogues, tracer dyes, natural tracers such as suspended sediment, or thermal gradients [31 [5]. Surface water movement studies utilizing fluorescent tracer dyes have been performed on most of the major surface water bodies and near coastal waters of the United States. These studies have been conducted to determine the dynamic characteristics of the water bodies with the primary objective being to trace the current flow rate and direction, and the rate of dispersion [6]. Rhodamine B and rhodamine WT are two of the more commonly used dyes, having specific gravities at room temperature of 1.12 and 1.19 and maximum emission wavelengths of 0.579 and 0.582 ,u, for solutions of 40 and 20 percent, respectively. Rhodamine WT dye can be tracked for several hours by low-altitude aircraft carrying color cameras. After several hours, however, the dye is diluted to concentrations that can be discemed only by aircraft cameras with special optical filters that are optimized for the spectrum of each dye used. This is particularly true for dye experiments in turbid coastal waters [7] . Using suspended sedimnent as a natural tracer, it is possible to study current circulation in the surface layers of turbid estuaries and coastal waters by analyzing satellite imagery with the help of a small amount of ground truth data [6], [8], [9]. We have analyzed imagery and digital tapes from 36 passes of the Earth Resources Technology Satellite (ERTS-1) over Delaware Bay. The ERTS-1 imagery used in our work was produced by the four-channel multispectral scanner (MSS) having the following bands: band 4 0.5-0.6 ,; band 5 = 0.6-0.7 ji; band 6 = 0.7-0.8 ,u; band 7 - 0.8-1.1 ,. From an altitude of 920 km, each frame covered an area of 185 X 185 km. In addition to nine-track 800-bpi magnetic tapes, reconstructed negative and positive transparencies on 70-mm format and prints in 9-in format were obtained from NASA. Before visual interpretation, some of the imagery was enhanced optically, using density slicing and color additive techniques. Annotated thematic maps were prepared by computer analysis of digital tapes and by direct photo interpretation of the transparencies reconstructed by NASA [101. Figs. 1-3 contain ERTS-1 pictures and NOAA-NOS tidal current charts for Delaware Bay. Only MSS band 5 images

98

IEEE TRANSACTIONS ON GEOSCIENCE

TIDAL CURRENT CHART DELAWARE BAY AND RIVER Red tt

s

ket th

o

-t

en

p*). 7tb ttE fp.-kentithe Ct at Trek AtM-.t en.trt -f te lea ledn bta tAt A-Th ai Gt;e* SIDlst.

V

NO.\t\lN

'\t

\DE

2.4

1.~~~~~~~~~~~~~~~t,

\0.3\

\0.3A

A>ARE

1

9

t

A

4

Was~~~~~~~~~~~~~a 0.5

HUSATRMXMMFOD 0WO

0.53

,_

.,

,

ATlW20.K%%~~~1.4 DLAW~ARE EAYETNC

(b) Fig. 1. (a) Predicted tidal currents. (b) An ERTS-1 MSS band 5 image of Delaware Bay obtained on October 10, 1972. (I.D. No.'s 107915133.)

ELECTRONICS, APRIL 1977

KLEMAS et al.: STUDIES OF COASTAL CURRENTS AND POLLUTANTS

TWO HOURS BEFORE MAXIMUM FLOOD AT DELAWARE BAY ENTRANCE

(a)

(b) Fig. 2. (a) Predicted tidal currents. (b) An ERTS-1 MSS band 5 image of Delaware Bay obtained on January 26, 1973. (I.D. No.'s 118715140.)

99

loo

IEEE TRANSACTIONS ON GEOSCIENCE ELECTRONICS, APRIL 1977

ON1E HOtUR AFlER MAXIMUM EHE

l

1 AVAh^RI HAY IN lIAN,

(a)

(b) Fig. 3. (a) Predicted tidal currents. (b) An ERTS-1 MSS band 5 image of Delaware Bay taken on February 13, 1973. (I.D. No.'s 120515141.)

KLEMAS et al.: STUDIES OF COASTAL CURRENTS AND POLLUTANTS

101

* Salem

39s0

I

* Brd...

.I .I

I

.I .I

i

D.a, *

.I

i

KEY Concentrations from non-

I

linear regression. mg/liter

.I

i

* 211

m

5.6

* 104

m

77

* 39

j]

unclassified

D

unclassified

i El 16 I

i

X s

(i

SEDIMENT DISTRIBUTION MAP

I

.I i

IL.-.-._____ El

7 .4

5, II, 7530%

"

Co.

'I

._ City,

,5w-L

Fig. 4. Suspended-sediment distribution obtained by correlating water sample analysis with the ERTS MSS band 5 image radiance. are shown, since this "red" band was found to give the best contrast in delineating the suspended-sediment concentration in the upper 1 m of the water column [101. Each ERTS1 picture is matched to the nearest predicted tidal current chart within ±30 min. The current charts indicate the hourly directions by arrows, and the velocities of the tidal currents in knots. NOAA-NOS made observations of the current from the surface to a maximum depth of 20 ft in compiling these charts

[111, [12]. The satellite picture in Fig. 1 was taken on October 10, 1973, 2 h after maximum flood at the entrance of Delaware Bay. Masses of highly turbid water are visible around the shoals near the mouth of the bay and in the shallow areas on both sides of the bay. Since at that time strong currents and considerable waves were prevailing in most of the bay, some of the sedi-

#-

0

,C4.5' a

ment in suspension seems to be locally generated over shoals and shallow areas of the bay, resulting in a higher degree of backscatter from shallower waters. During the flood tide at the mouth of the bay, considerable correlation was found between the depth profile and image radiance, even though the water depth exceeded the Secchi depth by at least a factor of three in all areas. No such correlation was found during the

ebb tide when finer silt and clay are carried down the rivers into the bay. The flood currents are moving at maximum velocity in the upper portion of the bay, where strong boundaries (density fronts) parallel to the edges of the deep channel can also be seen. Fig. 2 represents the tidal conditions 2 h before maximum flood at the mouth of the bay observed by ERTS-1 on January 26, 1973. Hiigh water slack is occurring in the upper portion

102

of the bay, accompanied by less pronounced boundaries there as compared to Fig. 1. The shelf tidal water is not rushing along the deep channel upstream anymore as in Fig. 1, but is caught between the incipient ebb flow coming down the upper portion of the river and the last phase of the flood currents still entering the bay. The satellite overpass on February 13, 1973 occurred about 1 h after maximum ebb at the mouth of Delaware Bay. The corresponding ERTS-1 image and predicted tidal currents are shown in Fig. 3. Strong sediment transport out of the bay in the upper portion of the water column is clearly visible, with some of the plumes extending up to 20 mi out of the bay. Several eddies are visible, including the large clockwise one on the lee side of Cape Henlopen. Small sediment plumes along New Jersey's coast indicate that the direction of the near-shore current at that time was towards the north. In addition to current-circulation studies, the ERTS-1 image radiance of band 5 was correlated with suspended-sedimentconcentration and Secchi-depth data obtained from boats and helicopters during the satelite overpasses [101. A suspendedconcentration map based on ERTS-1 image-radiance correlation with water-sample analyses is shown in Fig. 4 for tidal conditions identical to those in Fig. 2. SUBSURFACE CURRENT MEASUREMENT WITH REMOTELY TRACKED DROGUES Two important shortcomings of satellite investigations of coastal currents have been their inability to determine current magnitudes and penetrate beyond the upper few meters of the water column. These deficiencies have been overcome by supplementing satellite observations with drogues tracking currents at various selected depths. Two types of drogues have proven useful. The first are small compact units which can be dropped and tracked from lowflying circraft. Their basic design does not differ significantly fronm that of drogues used by various investigators during the past few decades [131. These small drogues are deployed whenever a detailed charting of the current circulation over a relatively small area, such as 4 mi2, is desired. As shown in Fig. 5, the drogues consist of a styrofoam float and a line to which is attached a current trap consisting of a galvanized steel biplane. The length of the line determines at what depth currents will be imonitored. Vertical shear limits the use of this drogue design to approximately the upper 6 ft of the water column. The floats are color-coded to distinguish their inovement and mark the depth of the biplanes. Packs with dyes of two different colors can be attached to the float and the biplane. The movement of the dye and drogues is tracked by sequential aerial photography, using fixed markers on shore or on buoys as reference points to calibrate the scale and direction of the drogue movement. The second type of drogue is a "radio sonde," which can be released to track currents over large areas such as an entire segment of the continental shelf. The unit was developed by the Electro-Physics Laboratories (ITT Avionics Division) with the intention of providing a free-drifting drogue which would be sufficiently economical to render it unnecessary to recover the expended hulk. The sea sonde employs a low-powered

IEEE TRANSACTIONS ON GEOSCIENCE ELECTRONICS, APRIL 1977 3/8 THREADED STEEL ROD

MARINE STYROFOAM

1/4" PAINTED PLYWOOD

I

l WASHER SWIVEL, SNAP HOOK

------ADJUSTABLE NYLON ROPE

WOOD

-----GALVANIZED STEEL BIPLANE

PACKET

DROGUE AND DYE EXPERIMENTAL PACKAGE

Fig. 5. Compact current drogue with dye packs. Clusters of these drogues are dropped and tracked from aircraft to chart the current in the upper few meters of the water column.

(solid-state) HF radio transmitter feeding a matched antenna system. The antenna radiates 10-50 mW in the 6-MHz region to telemeter its position and any of four other optional variables of interest. The particular sonde used in the studies described here consists of a 2-ft length of plastic pipe, somewhat less than 2 in in diameter (Fig. 6). Buoyancy is provided by a pair of floatation chambers so attached that when properly ballasted the antenna portion of the sonde projects approximately 75 in above a still-sea datum plane [14] .

All necessary electronics are carried inside the pipe below the water line. Power is provided by four standard D-sized flashlight cells including heavy-duty zinc chloride, alkaline, or lithium batteries. The transmitter exerts a drain of about 20 mA average on the power supply. A light-sensing phototransistor can be used to shut down the system during the hours of darkness and conserve battery life. The sonde's operating life is selectable from one to almost six weeks through the choice of battery type and control of the transmitter power. The tracking range is not particularly sensitive to a planned operating life. Current sensing is achieved by the use of a current trap or biplane which is suspended at a controllable depth beneath the sonde hull. The current intercept area is from 4 to 16 ft2 and is isotropic. A ballast weight of approximately 5 lb is attached to the bottom of the current trap by a 3-ft-long slack line. The ballast provides a righting moment about the system's metacenter, resisting heelover under strong wind conditions and righting the sonde if it is upset by the actions of large waves.

Position finding is normally accomplished by the use of triangulation from two (or more) shore radio direction-finding

103

KLEMAS et al.: STUDIES OF COASTAL CURRENTS AND POLLUTANTS

chosen to permit corrosive leaks to flood the chambers and let the units sink after the mission is completed. HF Antenna

6/14 Ft

APPLICATIONS

Influence of Frontal Systems on Oil-Slick Drift Electronics

/

DEEP CURRENT DROGUE WITH LOW PARASITIC DRAG (Model 3)

L

A-10

Flototaon Chamber

Low - drog LIne

Variable; to 300 Ft

|.

4Ff -

1

4Ft.

-*

-

4-*-m

B

Current Trap (Bip/ane)

allast

Fig. 6. Deep current drogue with low parasitic drag (Model 3). These drogues are tracked from shore stations by means of radio signals over ranges of 100 mi.

stations using surface-wave or ground-wave signals. The ground-wave triangulation routine provides an accuracy of about ± 1.0° if the system is recalibrated before each measurement on a moored sonde at a known position. Even with a long baseline, however, the position error of drogues moving to the outer portion of the shelf becomes unacceptable. Therefore for studies covering distant segments of the shelf, more expensive drogue designs will have to be considered, e.g., drogues which contain Loran or Omega repeaters. Since the buoy portion of the drogue is nearly awash and has only a thin radio antenna protruding above the water surface, wind drag on the drogue is usually not significant. A ratio in excess of 20: 1 is maintained between the projected area of the buoy exposed to surface currents and the projected area of the current trap exposed to currents at its depth. Careful hydrodynamic design of the buoy hull structure further reduces the effect of surface currents. Compared to the cost of current meters, the drogues are not expensive. Each radio-tracked drogue costs about $175; each aircraft-tracked drogue, about $20. The low cost makes the drogues expendable, i.e., usually it is less expensive to leave them in the water than to recover them. To remove drogues from the ocean's surface, the coating of the flotation cans is

Fronts or boundaries are a major hydrographic feature in Delaware Bay and other estuaries. Horizontal salinity gradients of 4 0/00 in one meter have been observed and convergence velocities of the order of 0.1 m/s are typical. Often, fronts extend for tens of miles; generally parallel to the axis of the bay's channels. They are observed on both sides of the bay and along every channel where looked for. Fronts have also been observed near the mouth of the bay where they appear to be associated with the tidal interaction of the shelf and estuarine water [17]. Fronts such as the ones shown in the ERTS image of Fig. 1 have been investigated using STD sections, the small drogues shown in Fig. 5, and aerial photography. Aquatic fronts are similar to atmospheric fronts in that the denser fluid tends to under ride the lighter fluid giving rise to an inclined interface. This dynamical behavior produces a marked surface convergence. In the authors' investigations in Delaware Bay, it has been observed that these convergences result in the concentration of foam, surface films, and oil slicks at fronts. Surface slicks and foam collected at frontal convergence zones near boundaries were found to contain concentrations of Cr, Cu, Fe, Hg, Pb, and Zn higher by two to four orders of magnitude than concentrations in mean ocean water [15]. By capturing and holding oil slicks, frontal systems, such as the one shown in Fig. 7, significantly influence the dynamic behavior of oil slicks in Delaware Bay. Recent oil-slick tracking experiments conducted by the authors, in order to verify a predictive oil spreading and drift model, have shown that during certain parts of the tidal cycle the oil slicks tend to line up along the boundaries [ 16]. This effect was illustrated by an oil spill which occurred on January 10, 1975, as a result of a lightering operation in the anchorage area off Big Stone Beach in Delaware Bay. As shown in Fig. 8, at 0930 hours the spill consisted of four large slicks and numerous smaller ones almost randomly dispersed throughout the area. The wind was about 10 knots from the east-by-southeast. The flood-tide cycle was coming to an end with a near slack current, as measured with current meters and the air-tracked small drogues shown in Fig. 5. At this time no boundaries were observed. As shown in Fig. 9, by 1100 hours the current had turned to the ebb direction with a velocity of 0.6 knots. Two convergent boundaries were clearly visible on either side of the oil slicks and were starting to attract some of the slicks. Changing to a smaller scale (higher altitude photography), one can see from Fig. 10 that by 1500 hours most of the oil was aligned along the boundaries and stretched into two 5-mi long slicks. The predictive model is currently being modified to include the effect on oil-slick behavior due to those boundaries which form repeatedly at known locations and prevail over major portions of the tidal cycle. To deter-

IEEE TRANSACTIONS ON GEOSCIENCE ELECTRONICS, APRIL 1977

104

Fig. 7. Convergent frontal system with a strong turbidity gradient in Delaware Bay off Woodland Beach (scale 1:80 000).

'WN 8

0950 hrs.

R.V Worfield

CIII,

11-11

SW`C F`4F

I

"I.-,

One Nautical Mile Kilometers

Fig. 8. On January 10, 1975, the shown oil-slick pattern was encountered at the oil-lightering area in lower Delaware Bay. At 0950 hours flood-tide currents were slowing down and no boundaries were observed.

mine the dominant locations in the bay where boundaries tend to form, an analysis has been performed of 36 ERTS pictures, such as the ones shown in Figs. 1-3, obtained during all portions of the tidal cycle. Combining satellites and aircraft with air-dropped air-tracked drogues and boats proved to be the Mnost effective means of monitoring the currents, boundaries, and oil-slick behavior in order to improve and verify the predictions of the oil-slick movement model.

DRIFT

AND

DISPERSION OF OCEAN WASTE DISPOSAL PLUMES

Thirty-eight nautical miles southeast of Cape Henlopen, Delaware, is the disposal site for waste discharged from a plant producing titanium dioxide. The discharge is a greenish-brown liquid containing up to 10-percent acidity (expressed as HCI) and 4-percent iron as iron chloride salts. The barge which

105

KLEMAS et al.: STUDIES OF COASTAL CURRENTS AND POLLUTANTS .1

Current

X~~~~

.6 Knots

I..

~

~

\

~

~

~

~

WV'C" Fl

~

~

~

~

R.V.

v

1100 hrs

Warfield

~

4sec

0 l One Nautical Mile 0 Kilometers

I

2

Fig. 9. By 1100 hours the tidal currents had assumed an ebb direction, and oil slicks were attracted by convergent boundaries which had appeared around 1030 hours.

transports this waste is capable of releasing

1

000 000 gallons

of the liquid upon radio command from a towing tug. It

Area

makes several trips to the disposal site per week.

The frequency of this dumping made is possible for the

ERTS satellite to image the waste plume in various stages of degradation, ranging from minutes to days after dump initiation. The long visual persistence has to some extent been explained by the formation of a suspended ferric floc. Eighteen images were found which show water discoloration in the general vicinity of the waste dump site. The spectral characteristics and position of the discoloration, the dump pattern, and the time difference between the dump and pho-

tograph

gave

strong indications that the discolorations are the

acid-waste plume. For instance, careful examination of the picture of January 25, 1973 discloses a fishhook-shaped plume about 40 mi southeast of Cape Henlopen caused by a barge disposing acid wastes. As shown in Fig. 11, enlarged enhancements of the acid-waste patterns, prepared from ERTS MSS digital tapes, aided considerably in studies of the dispersion of the wastes. Digital radiometric printouts of the waste plumes are currently being correlated with concentrations of dissolved- and particulate-substance, such as iron, samples from boats at the time of the aircraft and satellite overpass. The change in concentration over cross sections of the plume are used to determine the dispersion properties of the waste. Spectrometers deployed from helicopters and boats have also been used to

0

\Qk

Fl

'wU,\

Fowler Beach c,

Fig. 10. A higher altitude view of

up

'the area shows most of the oil lined in 5-mi-long narrow slicl:ks along the two boundaries.

106

IEEE TRANSACTIONS ON GEOSCIENCE ELECTRONICS, APRIL 1977

Fig. 11. Enlarged digital enhancement of the acid-waste plume imaged by ERTS-1 on January 25,1973.

determine the spectral signatures of waste plumes such as the one in Fig. 12, imaged by ERTS on August 28, 1975, which has the new figure-eight dump configuration. Fig. 13 shows the dispersion of a similar dump pattern 3 h after the dump occurred. Spectrometric measurements indicate that upon combining with seawater the acid waste develops a strong reflectance peak in the 0.55-0.60-p region, resulting in a stronger contrast in the ERTS MSS band 4 than in the other bands. Once a month for a complete year, remotely tracked current drogues, such as the ones shown in Fig. 6, have been released at the acid-waste disposal site to monitor currents. Three to four drogues are usually released at three depths: surface, mid-depth, and near bottom. The surface drogue tracks currents in the upper 2 ft of the water column; the mid-depth drogue is set just above the summer thermocline at 45 ft; the near-bottom drogue, at depths between 75 and 85 ft. A preliminary analysis of the drogue tracking data agreed with previously published results [2] and provided a few new insights. 1) The circulation processes at the waste disposal site are indeed highly weather-event-dominated, with the majority of the water transport occurring during strong northeaster storms. In a test of severe weather survivability, several drogues were released at the waste disposal site in January 1975. For a period of ten days, tracking was conducted while severe winter sea conditions prevailed in the test-site area. Significant wave heights on several occasions exceeded 20 ft and averaged 8-14 ft during the tests. The air temperature ranged from the low 20's to the mid-40's (OF). The sea temperature ranged from 37 to 42°F. Wind gusts exceeded 50 knots over a substantial portion of three days in the period. 2) The predominant direction of movement of the waste plumes imaged by the LANDSAT satellite was to the southwest (Fig. 14). The maximum range of measurable wastes estimated as being about 10 mi from the discharge point was substantiated by the satellite imagery since only one out of 16 plumes was observed to be significantly beyond this range. 3) The average drift velocity for the surface drogues and waste plumes as observed by LANDSAT was about 05 knots. Drogues released at different depths frequently travelled along different paths and at different speeds, indicating the presence

Fig. 12. Acid-waste plume imaged about 38 mi southeast of Cape Henlopen by ERTS (LANDSAT) on August 28, 1975, with the MSS band 4 (0.5-0.6 ,u).

Fig. 13. Acid-waste plume imaged about 38 mi southeast of Cape Henlopen by ERTS (LANDSAT) on February 24, 1976, 3 h after the dump was completed.

of current shear. During stratified conditions, the near-bottom drogues showed very little movement. 4) The waters at the test site were highly stratified and stable in the summer and nearly homogeneous in the winter (Fig. 15). A distinct thermocline was observed from June through August, at depths ranging from 43 to 103 ft. During most dumps the acid plume was unable to penetrate the thermocline and reach the bottom.

107

KLEMAS et al.: STUDIES OF COASTAL CURRENTS AND POLLUTANTS N

w

Fig. 14. Distance from center of dump site to centroid of imaged

plume.

0

2

4

6

10

8

12

TEMPERATURE ( C) 14 16 18 20

22

24

26

28

30

0

5

10

primarily during storms, particularly northeasters. During such storms, however, the plume is rapidly dispersed and diluted. Therefore, the probability of an identifiable plume containing heavy concentrations of waste reaching the shore is quite low. CONCLUSIONS

a: w

I C

15

0-

25

30

L

I11

11

Fig. 15. Temperature profiles obtained with bathythermographs show water stratification and thermocline location during summer months.

The current-circulation data are being used to assess the movement and dispersion of the acid-waste plumes. In general, it appears that rapid movement toward shore can occur

The movement and dispersion of natural or man-made suspended matter, such as sediment or ocean-dumped wastes, can be observed synoptically over large coastal areas by satellites such as ERTS (LANDSAT). However, most satellites cannot provide short-term time-lapse photography of a circulation pattem since their orbits are not earth synchronous. Therefore, the rate of the water mass movement cannot be determined. Moreover, remote sensors cannot penetrate more than the upper few feet of turbid coastal waters. As a result, little subsurface data can be obtained from the satellite. Both limitations can be overcome by combining satellite and aircraft remote sensors with remotely tracked current drogues to devise a cost-effective system for monitoring the movement of coastal currents and pollutants over large areas, at various depths, and under severe environmental conditions.

108

IEEE TRANSACTIONS ON GEOSCIENCE ELECTRONICS, VOL. GE-15, NO. 2, APRIL 1977

REFERENCES [1] P. P. Niiler and C. N. K. Mooers, "Programs in shelf dynamics," presented at the Conf. Physical Oceanography of the Continental Shelves, Annapolis, MD, Apr. 3-6, 1974. [2] D. F. Bumpus, "A description of the circulation on the continental shelf of the east coast of the United States," in Progress in Oceanography, vol. 6. New York: Pergamon, 1973, pp. 111156. [3] M. Keller, "Tidal current surveys by photogrammetric methods," Photogrammetric Engineering, vol. 29, no. 5, pp. 824-832,

1963. [4] J. R. Eliason and H. P. Foote, "Surface water movement studies utilizing a tracer dye imaging system," in Proc. Seventh Int.

[5]

[61 [7] [8] [9]

Symp. Remote Sensing of Environment, pp. 731-749, 1971. F. L. Scarpace and T. Green, "Dynamics surface temperature structure of thermal plumes," Water Resources Research, vol. 9, no. 1, pp. 138-152,1973. P. Teleki, "Data acquisition methods of coastal currents," presented at the Ocean Engineering II Conf., Newark, DE, 1975. V. Klemas, D. Maurer, W. Leatham, P. Kinner, and W. Treasure, "Dye and drogue studies of spoil disposal and oil dispersion," J. Water Pollution Control Federation, vol. 46, no. 8, pp. 20262034, 1974. R. L. Mairs and D. K. Clark, "Remote sensing of estuarine circulation dynamics," Photogrammetric Engineering, vol. 39, no. 9, pp. 927-938, 1973. G. A. Maul, "Applications of ERTS data to oceanography and

marine environment," in Proc. COSPAR Symp. Earth Survey

Problems, pp. 335-347,1974.

[10] V. Klemas, D. Bartlett, W. Philpot, and R. Rogers, "Coastal and estuarine studies with ERTS-1 and Skylab," Remote Sensing of Environment, vol. 3, no. 3, pp. 153-174, 1974. [11] U. S. Department of Commerce, Tidal Current Charts-Delaware Bay River, Environmental Science Services Administration, Coast and Geodetic Survey, 2nd ed., 1960. [12] U. S. Department of Commerce, Tidal Current Tables, Atlantic Coast of North America, National Oceanic and Atmospheric Adminsitration, National Ocean Survey, 1972 and 1973. [13] E. C. Monahan and E. A. Monahan, "Trends in drogue design," Limnology and Oceanography, vol. 17, no. 6, pp. 981-985, Nov. 1973. [14] V. Klemas, G. Tornatore, and W. Whelan, "A new current drogue for monitoring shelf circulation," presented at the 56th Annu. Meeting American Geophysical Union, Washington, DC, June 16-19, 1975. [15] K. H. Szekielda, S. Kupferman, V. Klemas, and D. Polis, "Element enrichment in organic films and foam associated with aquatic frontal systems," J. Geophys. Res., vol. 77, no. 27, pp. 5278-5282, Sept. 1972. [16] M. A. Tayfun and H. Wang, "Monte Carlo simulation of oil slick movements," J. Waterways, Harbors, and Coastal Engineering Division, ASCE, voL 9, no. WW3, pp. 309-324, 1973. [17] S. L. Kupferman, V. Klemas, D. F. Polis, and K. H. Szekielda, "Dynamics of aquatic frontal systems in Delaware Bay (Abstract)," Trans AGU, vol. 54, pp. 302, 1973.

Electromagnetic Shielding of Sources Within Metal-Cased Bore Hole

a

JAMES R. WAIT, FELLOW, IEEE, AND DAVID A. HILL

Abstract-An abbreviated derivation is given for the electromagnetic fields produced by a two-dimensional source located within a metalcased bore hole surrounded by a homogeneous rock formation. Numerical results are presented for the fields in the external region. It is shown that the combined influence of the conductivity and permeability of the steel casing will severely attenuate the fields even at frequencies as low as 100 Hz.

INTRODUCTION IN VARIOUS applications [1] -[41 in communications and geophysics, an estimate of how the metal casing of a drill hole will attenuate signals is needed. Whether we are dealing with a subsurface communication scheme or geophysically

probing the adjacent formation, the presence of the casing cannot be ignored. It is the purpose of this paper to consider this question in a quantitative fashion. FORMULATION The model we choose is rudimentary as indicated in Fig. 1. With respect to a cylindrical coordinate system (p, ', z), the metal casing or tube is bounded by the surfaces p = a and p = b = a + d. Here d is the thickness of the tubing, a is its inner radius, and b is its outer radius, which is flush with the external rock formation. The tube is composed of homogeneous

metal with conductivity a, permittivity E, and permeability,. The internal region, p < a, is assumed to be the same as free with permittivity co and permeability Muo. The external space Manuscript received December 16, 1976. J. R. Wait is with the Cooperative Institute for Research in Environ- region, p > b, is assumed to be homogeneous with conducmental Sciences, University of Colorado/NOAA, Boulder, CO 80309. tivity a., permittivity se, and permeability go. D. A. Hill is with the Institute for Telecommunication Sciences, OfWhile it is possible to construct a general theory (51, [61 for fice of Telecommunications, U.S. Department of Commerce, Boulder, any bounded source configuration within the tube, we will CO 80302.