~I

REMOTE SENSING OF HAWAIIAN COASTAL SPRINGS

USING MULTISPECTRAL AND INFRARED TECHNIQUES

by

Larry K. Lepley

Leonard A. Palmer

Technical Report No. 18

August 1967

No.5 of 5 Reports in Completion

of

GEOPHYSICAL EXPLORATION FOR HAWAIIAN GROUNDWATER, PHASE I

OWRR Project No. B-005-HI, Grant Agreement No. 14-01-0001-1011

Principal Investigators: Doak C. Cox, William M. Adams, Leonard A. Palmer

Project Period: July 1, 1966 to June 30, 1967

The programs and activities described he rein were supported i n par t by fundsprovi ded by the United States Department of the I nt er i or as aut hor i ze d undert he Water Resources Act of 1964, Public Law 88-379 .

Ii

ABSTRACT

Two remote sens~ng techniques of measuring offshore ground-water

spring discharge that have bee n studied are : (iJ mul tispect ral photo­

graphy~ and (iiJ infrared thermometry .

During the summer of 1 96 7~ expos ure factors f or filter-film­

camera combinations extending outside the visible spectra were experi ­

mentally determined and tabulat ed . One spectral series of simulated

aer ial photography was obtained . Darkroom procedures and equipment

for multispectral enhancement of the sus pect ed col or di f f erence s

between sea water and offshore spring water in these photographs have

not yet been precisely de fi ned .

Helicop ter f l i ght s with an infrared thermometer have demonstrated

that offshore spr ings can be i dentified by their temperatures . Due

t o t he irregu lar shape ~ size~ and location of these fres h wate r out­

flows~ a synoptic t echnique is needed . An infrared scanner i s desira­

ble but the present costs ar e prohibitive .

i ii

CONTENTS

LIST OF TABLES •.••• ••.•...••.• ••.•.••.• ••••••••..••.••••..••••••••••vi

LIST OF FIGURES ..••••••••.•.••.•••••••. •.•.•••..•••••.••••••••••••••vi

INTRODUCTION •...•• . ••.....•.••.•.•••.• ; •• •.•••••.•••. •••••••••••.•••. 1

MULTISPECTRAL PHOTOGRAPHY

OBJECTIVE •...•.••.....••.•...•....•...••••.•.•.•.•••.••.••.••.••••••• 7

APPLICATIONS OF MULTISPECTRAL PHOTOGRAPHY •••.•.••••.•..•••••••••••••• 7

INSTRUMENTATION AND FIELD EXPERIMENTS ••••••.••••.••..•..•••••• •••• ••• 8

SUMMARY AND CONCLUSIONS 11

INFRARED THERMOMETRY

OBJECTIVES •••.•.••••.•••••..••••..•.•••••••.•••••••••••.••.••••••• ••17

PRELIMINARY COASTAL SPRING EXAMINATIONS • ••••••••••••• •.•••••••••.•••17

INSTRUMENTATION .••..•• •.••.•• ••••••..•.•.••.••••••••• ••••••..•.•••..19

Laboratory Testing of Instrument Response 21Field Application of Infrared Thermometer 26Temperature Data Recording ~ 29

FIELD TESTING ••.••• •• •••.•••.••••• •• •••••••.••.•••••.••••• •..•.•••••.29

Electric Power and Helicopter Platform 29

RESULTS AND ANALYSIS OF PRELIMINARY FIELD FLIGHTS •.••..•••••••••••••29

SUMMARY AND CONCLUSIONS • • . . . . • . . • • . • • • • ; ••••• ••••. ; ••••••.•••..•••••36

ACKNOWLEDGEMENT ••••••••••.•.••.•.•.•.•.....•••••.••••••••..•••••••.•36

BIBLIOGRAPHY •••.••.••••.•••••••••.• ••••••••••••••••.•••.••••••. ••••• 37

v

LIST OF TABLES

Table1 Filtered ASA Rating: Visible and Ultraviolet Light 102 Filtered ASA Rating: Infrared 103 Contrast in Infrared Image Density and Temperature and

Salinity of Selected Coastal Springs on the Island of Hawaii.19

LIST OF FIGURES

Figure1

2

3

4

5

6

7

8

9

1011

121314

15

16

Spectral Locations of Filters 9Unenhanced Pictures of the Same Scene Photographed; n Different Bands ,. I ,' •••••• • 12

Schematic of Laboratory Enhancement Procedure 13Configuration of Fresh and Salt Water Interface 18A Small Underwater Spring Churning Sand in theShallow Nearshore Waters at Kawai10a Beach, Oahu 20Block Diagram of the Infrared Thermometer. . . . . • . . . . . . . . . . . . . . 22Infrared Passbands in the Transmission of RadientEnergy Through Atmosphere 23Spectral Distribution of Energy for Perfect Emitters 24Laboratory Tests of Repeatability, Response Timeand Accuracy 25

Response Rate of Infrared Thermometer 27Diagram of Operating Velocities and Altitudes for aHe 1i copter ' ' _. . 28

Shock-mounted Instrumentation for Field Testing . •. . . . . . . . . . . .30Flight Path of Preliminary Test F1ights 31Infrared Temperature Profile of Residential andForest Area in Nuuanu Valley 33Infrared Temperature Profile of Sea, Beach, Park,and Urban Areas -' 34

Correlation of the Terrain to Infrared TemperatureProf; 1e 35

vi

INTRODUCTION

Ground-water springs near the shoreline of the Hawaiian Islands

discharge a large unmeasured port ion of the Ghyben-Herzberg fresh-water

lens. Irregular distribution and size of openings in the volcanic rock

through which most Hawaiian ground water flows make accurate flow mea­

surements difficult. There are no established methods for studying

coastal springs. Although they have been recognized in the Mediterra­

nean since before the time of Christ, and were also known to the ancient

Hawaiians, there is almost no scientific literature about them (Kohout,

1966).

In this project, spectral photography and infrared thermometry

are being evaluated as sensors to develop inexpensive and reliable

methods of detecting and measuring Hawaiian coastal ground-water springs.

These techniques show promise of being useful to coastal ground-water

spring detection.

Photography and photogrammetry has been used for documentation

of hydrologic projects for some time. A relatively new technique uses

multiple photographs in different portions of the light spectrum to

make multiple spectral photographic comparisons of the same objects.

It is possible to recogni ze and differentiate apparently similar ob­

jects. The recognition of plant diseases and crop identification as

well as military applications in photo reconaissance and camouflage

detection have made spectral photography valuable (Colwell, 1961; Loh­

man and Robinove, 1964).

MULTISPECTRAL PHOTOGRAPHY. Multispectral photography has been care­

fully analyzed on various types of ground and plant surfaces but less

experimental data is available on its response to various water sur­

faces (Colwell, et aZ., 1966). Spectral photographic experiments to

test multiple passband photography were organized to determine their

usefulness in detection of fresh-water coastal springs. Color filters

were applied to isolate bands of color in photographing the sea surface.

The resulting pictuyes are developed in both positive and negative

transparencies and projected in their various combinations upon a screen

for interpretation. False-color light is used to aid identification

of the effect of each color passband. For example, an obj ect is photo-

2

graphed through blue, green, light-red, and dark-red filters. Both

negative and positive projection transparencies are prepared from each

photograph, providing 8 slides of the object for comparative projection.

A positive black and white transparency photographed through the red

filter, for instance, may be projected onto a screen using any color

of monochromatic light. Other positive arid negative transparencies

photographed through other filters are compared by superposi~g 'them

on the screen in a contrasting monochromatic light. The resulting 'two­

color picture on the screen will emphasize only differences between

slides by the dominance of either the false color used or the projec~

tion.

The object is to find a combination which may be useful in en-. : . ' .

hancing color differences that will help to detect an object of interest.

In the present study, a coastal spring may have different organic and

other coloring matter which may be detectable by color enhancement.

INFRARED THERMOMETRY. Infrared thermometry has a clearly defined tech- '

nique of application but a more difficult instrumentation. The theory

of infrared temperature sensing is simple. Gates (1962) described many

of the instruments and their problems. ' Recent development of operational

instrumentation has been stimulated by research directed 't o military

applications of "imaging" radiometers and by research applied to remote

sensing of objects for use in space science applications (Harris and

Woodbridge, 1966). (For a definition of terms imagery and photography~

see Robinove, 1963.)

A military-type imaging radiometer was used by the U. S. Geolo­

gical Survey in a study of the temperature distribution of volcanoes

on the island of Hawaii. As an adjunct to those studies, a number of

aerial traverses were made along the coastal areas of the island of

Hawaii. Two hundred nineteen springs were detected by infrared imagery

which registered both warmer and colder temperature anomalies in the

ocean where volcanic warmed and normal ground-water springs occurred

(Fischer, et aZ.~ 1964 and 1966).

The infrared imaging radiometer is well-suited to the detection

and measurement of the relative temperature distributions of coastal

springs, especially in its ability to present a synoptic view of flow

patterns and temperature distribution by registering their location

3

in relation to other known objects. However, imaging radiometers pre­

sent difficulties in their inability to give accurate temperature read­

ings, their relatively high cost of operation, and the need for highly

skilled technicians to operate the tempermental instruments. They must

be transported in a rather large aircraft capable of supplying the neces­

sary power sources and space.

The logistics and cost of infrared imaging are offset by its

efficiency in a cooperative multiple-use program. However, for limited

use to measure small areas or selected areas intermittantly, such as,

documentation of variations with time, the cost is prohibitive. Inex­

pensive infrared thermometers which measure the infrared radiant energy

for a small area are available at a more reasonable cost. These instru­

ments provide greater accuracy than imaging radiometers but do not pro­

vide an instantaneous synoptic view, nor do they provide geographical

orientation to the point of the temperature measurement.

MULTISPECTML PHOTCGMPHYby

LARRY K. LEPLEY

I

7

OBJECTIVE

The objective of the current experiments in multispectral photo­

graphy is to devis e a t echnique of enchancing an otherwise imperceptible

color difference, including those in the ultraviolet and infrared wave­

bands, for the detection and delineation of coastal spring water bound­

aries by their optical differences from the surrounding sea water.

APPLICATIONS OF MULTISPECTRAL PHOTOGRAPHY

Multispectral photography, or multiband spectral reconnaissance,

involves darkroom projection techniques of additive and subtractive

superposition of multiple images of the same scene photographed separately

in color bands between 3000 and 9000 AO. The object of this spectral

separation and opt ical counterbalancing is to produce a brilliant image

in visually contrasting arbitrary colors where natural color contrasts

are visually undetectable (Molineux , 1964).

Studies of human color perception (Land, 1959, Legrande, 1959)

have shown that in the human visual system, combinations of two or

more color bands wi l l produce the s ensation of a third color. For ex­

ample, red added to green is perceived as yellow. Optical filters

passing only narrow bands of color cannot be so deceived. This differ­

ence between human and instrumental color perception forms the basis for

multispectral color enhancement techniques.

Photographic spectral analysis has been applied to camouflage

detection, agronomy, and forestry (Colwell, 1961, 1965), but only limit­

ed data are available on its use on various water surfaces (Ewing, 1965).

Some of the suspected physical differences between spring water

and sea water supporting the search for instrumentally detectable optical

contrasts are:

1) Fresh water contains different dissolved and suspended organic

matter, i .e . ~ tannin dyes from land vegetation in fresh water

as oppose~ to yellow "gelbstoff" from chlorophyl-bearing phyto­

plankton in sea water (Kalle ,. 1960);

2) Spectral interference effects from thin organic films on either

water surface may be especially detectable in the near infrared

(visible sea slicks from such films are known to sometimes

8

align themselves with subsurface water structures, such as,

,internal waves);

3) Exsolving gasses bubbling from cold floating spring water warmed

by sea water may throw small droplets of water into the air.

This effervescent mist may be detectable by its scattering of

infrared light (an infrared "fog" layer);

4) Capillary surface waves on the sea water surface may be smoothed

by the spreading surface layer of fresh water. The consequent

modification of surface-reflected skylight caused by the smooth

surface can be enhanced by a comparison of images photographed

through polarizing filters at various orientations.

INSTRUMENTATION AND FIELD EXPERIMENTS

Kahana Bay, Oahu was chosen as the site of field experiments because

it is small, sheltered, and contains submarine springs. The high cliffs

overlooking the bay provide platforms for simulated aerial photography.

A Calumet view camera with wide-angle lens provided the advantages of a

large-size negative, ground-glass viewer, and easily changed components.

The angle of view to the water surface ranges from 8° to 45° from the

horizontal.

The first phase of operations empirically established the filter

factors (the number of times by which the exposure must be increased for

a given filter with a given film) for the filter~film-camera combinations

of more than 20 Wratten filters covering color bands of various widths

ranging from 3000 to 9000 AO. (Figure I shows the spectral locations of

most of these filters.) Films included Polaroid PN and infrared black

and white.

Infrared data were also determined as follows: since there is no '

consistent rating for infrared film used in visible light ,the data recov­

ered were based on exposure trials which in most cases go beyond the

the scales normally included on standard ASA light meters. The ASA speeds

listed have therefore been computed from the most useful exposures. Fil­

ters selected and factors established are noted in Tables land 2. The

spectral transmissions were determined from Kodak Publication B-3 and

are based on a transmission level of 10%.1

Tables 1 and 2 were compiled by Bruce Erickson.

-ULTRA-VIOLET ·1' VISIBLE UGHT '1- INFRA-RED •

I. PANCHROMATIC FLM (VP)

~JT~ol

~--':'-II I f,-;12B i1 I .rS011 sooo 40 II i:'%11 ' mL 70

L87C _83OOAo 0

o

:~

89 B : d7050 AO ~ u.

..o

LL87 ,ou.7600 AO : ~

o ,

0t-88 A : a7330 AO

L3 4 '(§)

24-

I L 29 -

L 25A -

, 21-

<a>

~_180_llUv_J ~0· ~J L6)~ \I . . . ~(§)

FIGURE 1. SPECTRAL LOCATIONS OF FILTERS.1.0

""'"

10

FI LTEROR

STOP ASA WITHFILTER NO. FILTER COLOR TRANSMISSION RANGE FACTOR I~REASE 50 SPEED FI LM

18 A (W) BLACK PURPLE 3200 AOD_ 3900 AD 400 X 8 2/3 . 125

47 b DARK BLUE 3950 AD - 4700 AD 8 X 3 6. 000

75 DARK BLUE-GREEN 4700 AD - 5050 AD 32 X 5 1. 600

74 DARK GREEN 5200 AD - 5450 AD 20 X 4 1/3 2. 500

21 ORANGE 5200 AD 2 X 25.000

74 + 21 5200 AD - 5450 AD 320 X 8 1/3 .160

11 YELLOW GREEN 4650 AD - 6400 AD 4 X 2 12.000

21 + 11 5200 AD - 6400 AD 10 X 3 1/3 5.000

24 LIGHT RED 5800 AD - 6 X 2 2/ 3 8. 000

11 + 24 5800 AD - 6400 AD 32 X 5 1.600

29 DARK RED 6100 AD 25 X 4 2/3 2. 000

11 + 29 6100 AD - 6400 AD 100 X 6 2/3 . 500

34 DARK VIOLET 3300 AD - 4700 AD & 6500 AD ~ 16 X 4 3. 000

11 + 34 4650 AD - 4700 AD 400 X 8 2/3 . 125

73 DARK YELLOW @ 1% : 5550 AD - 5950 AD !l00 X 8 2/3 .125

72 b DARK ORAI'GE @ 1%: 5900 AD - 6350 AD 500 X 9 .100

TABLE 2. FILTERED ASA RAT ING; INFRARED.

TRANSM ISSIONFILTER NO. RANGE @ 10% ASA AC TUAL EXPOSURE IN SECONDS

89 b FROM 7000 AD ON 10.0000 1 5 @f 36 (ASA 5 OVER WATER )

87 FROM 7600 AD ON 6.000 0 1/ 5 5 @ f 18 IN BRIGHT SUN

87 b FROM 8600 AD ON . 0320 60 5 @ f 22 IN BR IGHT SUN

87 a FROM 9500 AD ON . 0125 60 5 @ f 11 IN BR IGHT SUN

The next phase was the collection of imagery . Instead of using

multiple airborne cameras to obtain simultaneous images through differ­

ent filters, the top of the cliff was used as a steady platform to pro­

duce sequential imagery of the same s cene wi t h filt er and film changes

between frames. Changes i n the scene due to the fOrward travel of ocean

11

swells were negated by "blurring out" with long time exposures. . Exami­

nations of photographs resulting from these initial tests showed that

longer exposures (+ 12 sec) are needed to smooth out water waves and

that a polarizing filter is needed in even the longer wavelengths of

light because the infrared light from foilage on the opposite side of

the bay is reflected from the water surface (Figure 2c).

Figure 2 (a,b,c) shows the initial unenhanced differences in

an image photographed in different bands. Repeat photography and more

surface mapping is needed before the anomalies can be interpreted as

either natural "ground" phenomena or "noise" picked up in the photogra­

phic process. To date, only a limited amount of time has been spent on

field and laboratory work in multispectral photography. No enhanced

versions of photographs have yet been produced.

A schematic representation of the laboratory enhancementproce­

dure in Figure 3 represents A as two visually indistinguishable color

spots in a "noise" field, and the appearance of the resulting negative

images obtained through any given filters. Posi~ive transparencies,

B, are made from all negatives. A negative transparency from one color

band is then superimposed on a positive transparency from another color

band and visa-versa in C. These composi~es now have enhanced the spot

of one of the respective colors and have cancelled out the background

"noise" field. In D, arbitrarily selected brightly-colored light is

used to project the sandwiched images sequentially and in register upon

the single piece of color film. The resulting image delineates in

brightly contrasting false colors previously subtle color differences

and has "grayed out" areas of original color lying outside of the filter

passbands.

SUMMARY AND CONCLUSIONS

Multispectral photographic equipment has been calibrated for spec­

tral sensitivity to sea water. More surface survey is needed before

imagery can be interpreted. Current laboratory work to produce false­

color enhanced images will determine feasibility of the technique for

use in locating submarine springs by their optical effects.

('

12

a . Wratten #11 filter (dark green) with polarizer(vertical mode).

b. Wratten #29 filter (visible dark red) with polarizer(vertical mode).

c. Wratten #87b filter (infrared) without polarizer .

FIGURE 2. UNENHANCED PICTURES OF THE SAME SCENE PHOTO­GRAPHED IN DIFFERENT BANDS.

A. FIRST STAGE(FIELD WORK)

ORIGINAL SUBJECT

13

FILTERED NEGATIVE NO. I UNFILTERED FILM FILTERED NEGATIVE NO.2

~~~

\0 -7

B. SECOND STAGE(LAB WORK)

\C)

POSITIVE NO. I

C. THIRD STAGE(LAB WORK)

UNFILTERED POSITIVE

7 /~ ~

ARBITRARY COLOR FILTERSONTO ONE PIECE OF COLOR

FILM

/

POSITIVE NO.2

\ - 07

POSITIVE NO.2

COLOR ENHANCED VERSION OFORIGINAL SUBJECT

FIGURE 3. SCHEMATIC OF LABORATORY ENHANCEMENT PROCEDURE.

INFRARED THEPJVUv1ETRYby

LEONARD A, PALMER

17

OBJECTIVES

The infrared thermometry project encompasseQ: (i) study of known

coastal springs to determine approximate size and temperature contrasts,

(ii) laboratory test of instruments and construction of accessory

recording and power equipment, and (iii) aerial traverses to test the

ability of a thermistor bolometer to detect known springs.

PREL IMINARY COASTAL SPRING EXAMINATIONS

Preliminary studies of known areas of coastal springs were con­

ducted to establish ground data for later serial traverses . Extensive

traverses in a boat with thermistor temperature and salinity probes of

the Pearl Harbor area (the most promising of spring area of the island of

Oahu) failed to reveal the expected large discharging bodies of fresh

water. The only marked thermal and salinity contrast measured was de­

tected adjacent to an electric generating plant cooling water out f a l l.

Most of the springs discharge in swampy areas from which the water flows

in channels to the harbor. It may be that the temperature contrast

between the spring and harbor waters is reduced along the surface fl ow

north .

Efforts were made t o locate large springs on Oahu suitable f or

over-flight traverses with reasonable assurance of thermal contrast .

Various cold-water springs reported by scuba divers and skin divers

were sought and measured from boats and surfboards and with diving

gear. The most promising spring locat ion on Oahu is i n Kahana Bay

on the northshore of Oahu .

A rapid traverse of the coastline of the island of Hawaii was

made to measure the temperature and salinity contrasts of s ome of the

springs detected by Fischer, et aZ . ~ (1966), by infrared imaging aerial

traverses. The typical configuration of t he fresh and salt water in­

terface of the coastal springs of the Hawaiian Islands is shown in

Figure 4 . Fresh-water flows from the shoreline or from a spring on

the shallow sea floor. The density of fresh water is l ower than that

of sea water regardless of relative temperatures, theref ore a fresh ­

water surface layer forms over the more dense sea water. In the mixing

f

WARM SALT WATER

< WARM OR COLDFRESH WATER

FIGURE 4. CONFIGURATION OF FRESH AND SALT WATER INTERFACE. FRESH WATER FLOWS OVER SEA WATER OF GREATER DENSITY.A MIXING ZONE BETWEEN THE FRESH AND SEA WATER SHOWS TURBULENT REFRACTION OF LIGHT AND CAN BE RECOG­NIZED BY SKIN DIVERS REGARDLESS OF WATER TEMPERATU~E CONTRASTS.

......00

19

zone between the two bodies of water the turbulent refraction of light

is very pronounced and is a good indication of the presence of fresh

water in the absence of t~mperature contrasts . (Such turbulence is

readily observed by-a diver using a face mask.)

Progressive dispersion away from the spring source occurs by the

mixing and spreading of the fresh water. Not all springs yield high

quality fresh water because mixing occurs within the ground-water body,

but most are at least brackish. The temperature conductivity contrasts

of springs on the island of Hawaii are listed in Table 3.

TABLE 3. CONTRAST IN INFRARED IMAGE DENSITY AND TEMPERATUREAND SALINITY OF SELECTED COASTAL SPRINGS ON THEISLAND OF HAWAI I.

LOCATION IMAGERY DENSITY CONTRASTl TEMPERATURE CONDUCTIVITYSPRING/SEAo F2 SPRING/SEA (mhos)2

REEDS BAY, HILO 0.70 69/76 14,000/40,000

EAST OF HILO 0.66

PUNALUU 1.48 64/75 4000/30,000

PAPA BAY 0.30 NO SPR ING FOUND /40,000(NEAR MILOll I , /76SO. KONA)

lDATA OBTAINED FROM HYDROLOGIC ATLAS HHA-218 , USGS2DATA OBTAINED IN DECEMBER 1966

The largest springs are found near Hilo. However, a large total

discharge probably occurs as dissipated flow along the coasts with uni­

form permeability . Many examples of small flows through beach sediments

are reported by swimmers and divers and have been examined as part of

this study, but their measurement has not yet been attempted due to the

difficulty of isolating the ground-water flow from tidal drawdown flow

(Figure 5).

I NSTRUMENTATI ON

Although other thermometers and salinity measuring devices have

been employed, the primary instrument used in this study was a Barnes

Engineering Co. Model IT-3E infrared thermometer. For field surface

studies and laboratory calibration, various mercury thermometers, me­

chanical thermometers, and thermistor thermometers have been used. For

salinity measurements an Industrial Instruments (Beckman) model RB3-3341

FIGURE 5. A SMALL UNDERWATER SPRING CHURNING SAND IN THE SHALLOW NEARSHORE WATERS ATKAWAILOA BEACH, OAHU (IN CIRCLE TO LEFT). NUMEROUS SUCH SPRINGS, WHICH CUTSMALL CANALS THROUGH THE SAND, ARE EVIDENT IN THIS AREA (THE NORTHERN BEACHES).PHOTO BY LARRY LEPLEY.

No

21

Conductivity Bridge was used.

The infrared thermometer is essentially a thermistor flake measuring

radiation flux filtered by indium bromide and itron lens to transmit the

7-14 micron passband. The sensing thermistor is chopped by a rotating

blade to produce an AC signal which is verified by a second reference in

an oven within the instrument. Temperature is determined as a difference

between the detecting thermistor and a reference thermistor. (See block

diagram in Figure 6.)

The theory and construction of the infrared thermometer are well

described by Clark and Frank (1963) and Lenschowand Dutton (1964). Sim­

ply stated, the radiant flux in the infrared "window" between 7 and 14

microns wavelength is measured (see Figures 7 and S).

Laboratory Testing of Instrument Response

Barnes IT-3E. The infrared thermometer used in this project was labora­

tory tested for repeatability of temperature readings, response time,

and temperature accuracy. The instrument is rated to measure temperature

within 1.S oC absolute accuracy and O.SoC resolution. Response time is

specified as 50 milliseconds for 63% rise to true temperature in "fast"

response position and 500 milliseconds in slow response.

The response of the instrument was tested by applying switched

reference temperature targets. The targets with approximately 10°C

difference in temperature were switched on at about 1/2-minute intervals

to produce a square wave temperature reference curve. The results are

shown in Figure 9.

The temperature readings in the fast mode of response fluctuated

2°C continuously with 4° to 5°C noise peaks occurring about 3 per minute

and SoC noise peaks about 1 per minute. Response time was immediate

within the recording instrument capability.

Readings in the slow mode of response were constant within about

1-1/2 0C after approximately 15 seconds log time of instrument to drift.

One second after switching, the instrument recorded about 3° to 4°C error

in temperature reading.

Factory service of the instrument was required shortly after it

arrived. Tests were not made prior to the service. However, the apparent

response on the meter readout was much better before the factory repair.

NN

I IIIIIIIIII I

THERMOSTAT

HEATER

I I I PREAMPLIFIER

,---------IIIIIIIII

'.1

------.., 1-------------,I I II I I

~I I iii m:et~ I !IIII

TO RECORDER I

I , ,:-POST-AMPUFIER ' METER III I AND RECTIFIER

II I/l / '-DETECTOR II I

· 1 OPTICSJ I I BIAS I'I II

BATTERYBOX I

I II II II II SENSING HEAD II ELECTRONICS CONSOLE IL · __~L ~

INCOMINGRADIATION

FIGURE 6. BLOCK DIAGRAM OF THE INFRARED THERMOM~TER (AFTER CLARKE & FRANK, 1963).

- _._- --- - -_._._- - - - - _. .__._ - ---

1.0

0.9

0 .8r-zw 0.7u-~ 0 .6w0 0.5u

z 0.40(J)

0.3(J)-::E(J) 0 .2z<tc:::

0.1r-

00

23

2 3 4 5 6 7 8 9 10 II 12 13 14WAVELENGTH (MICRONS)

FIGURE 7. INFRARED PASSBANDS IN THE TRANSMISSION OF RADIANT ENERGYTHROUGH ATMOSPHERE (AFTER HARRIS AND WOODBRIDGE, 1966). THEINFRARED THERMOMETER MEASURES THE PASSBAND FROM 7 TO 14 MI­CRONS WAVELENGTH.

24

2018166 8 10 12 14

WAVELENGTH (MICRONS)42o

0.0

0.01

0.10

0.09lJJ0::lJJIa. 0.08en~lJJI

0:: 0.07lJJ _600oKa.z

0.0600::()

~

0:: 0 .05lJJa.

N

"E0 .04o

enl-I-

~ 0.03

t(J

0.02

FIGURE 8. SPECTRAL DISTRIBUTION OF ENERGY FOR PERFECT EMITTERS. THIS SHOWS THE RE­LATIVE CHANGE IN RADIATED ENERGY TO TEMPERATURE MEASURED BY THE INFRAREDTHERMOMETER (AFTER HARRIS AND WOODBRIDGE, 1966).

1--1 MINUTE 'I

25

RECORDED "FAST RESPONSE" TEMPERATURE

L:'ESPONSE LAG

RECORDED "SLOW RESPONSE" TEMPERATURE

"----' 33°C

TARGET REFERENCE TEMPERATURE (SHOWN DIAGRAMATICALLY)

FIGURE 9. LABORATORY TESTS OF REPEATABILITY, RESPONSE TIME,ANDACCURACY.

26

The magnitude of the noise response in Figure 9 was considered by

the factory to be "on the high side" but met their specifications. l

Expectable instrument response times are shown in Figure 10 as

provided by Dr. R. Lyons of the Stanford University remote sensing labo­

ratory.

Field Application of Infrared Thermometer

The infrared thermometer is most efficiently utilized for detecting

thermal contrasts in shoreline waters by using a flying platform. Speed

. of surveying and ease of operation are primary advantages over sea-surface

measurements.

The resolution of sea-surface temperature measurements are directly

related to the speed of travel and to the response rate of the instrument

as discussed above and in Figures 10 and 11. Due to the improvement in

temperature accuracy with slowing of response time, a helicopter was used

for the instrument platform vehicle. 2 Typical operating velocities and

altitudes for the helicopter used are shown diagrammatically in Figure 11.

The planned field conditions were determined by comparison of data

shown in Figures 10 and 11. The 10-foot wide target area changing 8 times

per second would allow "fast" mode instrument response to about 65% of

the true temperature reading.

Maximum expectable resolution from the instrument and flying platform

limitations is the recording of 65% true temperature contrast in a body

of water 10 feet wide, and of greater than about 5 degrees temperature

contrast with the surrounding water. Smaller than 10-foot wide bodies

would be recorded with less temperature contrast proportional to their

size. Temperature contrasts of less than 5 degrees are masked by instru­

mental "noise." Occasional noise peaks make positive detection of a 10­

foot wide body of water (1/8 second exposure to thermometer) difficult,

if not almost impossible. However, contrasting temperatures of larger

springs (50 to 100 feet wide) should be detectable.

12

Personal communication, H. Maier, Barnes Engr. Co., December 1966.See Acknowledgements.

100

80

t 60wenzoa..enw0:: 40

20

oo 5

DOWN TO 2/3 AT 8.5 SAMPLES/SECONDi.e,~ 120 mSEC. SYSTEM RESPONSE

10 15 20 25 30

TERRAIN SAMPLES/SEC = 2 X CHOPPING FREQUENCY

35 40

N--.J

FIGURE 10. RESPONSE RATE OF INFRARED THERMOMETER. (AFTER DATA FROM R. LYON, STANFORD UNIVERSITY)

28

60MPH=88FT/SEC.- ~

"'----r-_,..........-

100'

- -10 FT. WIDE TARGET AREA...........~_ ...~~

FIGURE 11. DIAGRAM OF OPERATING VELOCITIES AND ALTITUDES FOR A HELICOPTER. THERESOLUTION OF SEA SURFACE TEMPERATURE RELATED TO PLATFORM HEIGHT,AND VELOCITY, AND TO THE I.R. THERMOMETER CONE OF VIEW AND RESPONSETIME. SLOWER OR LOWER FLIGHTS CAN ONLY BE ACCOMPLISHED WITH LOSSOF FLYING SAFETY.

29

Temperature Data Recording

To allow versatile handling of the data, the readout of the infrared

thermometer was recorded on magnetic tape. An automobile cartridge-type,

4-track tape playing mechanism was converted to record the 5- to 50-milli­

volt temperature analog data as an FM signal and simultaneously record

voice field-note data on another track. Tape deck conversion involved

the installation of an in-line 4-track recording head on the tape recorder

mechanism as well as the FM recording electronics constructed by the

University of Hawaii's Institute of Geophysics electronic shops. Data

playback in the laboratory was done on a Crown 4-track tape deck fed into

a chart recorder. Voice data was transferred to the chart by hand.

FIELD TESTING

Electric Power and Helicopter Platform

Field testing of the infrared thermometer and recording unit was

done by mounting the sensor and recorder units and a battery-inverter

power source onto a stretcher attached to a Hawaiian Army National Guard

Helicopter (Figure 12).

Two flights were made over Oahu. The first traversed the central

portion of the island recording agricultural and land temperature en

route to the north shore where recordings were made on a zig zag course

along the shore from Kahuku to Kaneohe Bay. Only the most outstanding

temperature contrasts were recorded due to instrumentation problems

with the magnetic tape recording apparatus.

The second flight traversed the shoreline from Diamond Head to

Makapuu Point, and from Kaneohe Bay to Kahana Bay. Traverses were

done over Kahana Bay to record the temperature contrasts of coastal

springs and the outlet of Kahana Stream. The flight path is shown

in Figure 13.

RESULTS AND ANALYSIS OF PRELIMINARY FIELD FLIGHTS

Records obtained on the preliminary test flights confirmed expec­

tations of the limitations of the instrumentation and served to eliminate

instrumental malfunctions. Usable records were obtained on each flight,

FIGURE 12. SHOCK-MOUNTED INSTRUMENTATION FOR FIELD TESTING. THE UNITS STRAPPED TO THESTRETCHER IN THE FOREGROUND ARE FROM LEFT TO RIGHTJ SENSORY HEADJ ELECTRONICSAND READOUT CONSOLE J FM MAGNETIC TAPE DATA RECORDERJ INVERTERJ AND BATTERY(PHOTO BY LARRY LEPLEY).

(,No

<.M.....

FIGURE 13. FLIGHT PATH OF PRELIMINARY TEST FLIGHTS.

32

but better records were made on the last flight due to improvements in'

the recording apparatus. In general, the infrared thermometer was found

to perform as specified in its design, but either the response rate or

elimination of instrumental noise should be improved.

Selected sections of the traverses indicate the good correlation

of records with ground features, but the difficulty imposed by instrumen­

tal noise masks any but large areal and temperature anomalies of the

ocean surface at this time.

Infrared temperature profiling from residential housing to a for­

ested area is shown in Figure 14. Major features such as houses, streets,

and lakes correlate well with the temperature profile. Variable emissi­

vity of the objects may account for apparent anomalies such as the gen­

erally high level of radiation from the forest area. Difficulties with

recording cross talk from the voice track to temperature data recording

(not shown on records) were corrected prior to the second flight. Tra­

verses shown in Figures IS and 16 record both land and sea temperature

profiles. Figure IS records infrared temperature readings across sea,

beach, park,and urban areas with good recognizable correlation. Records

over the sea water show excessive fluctuation. Sea water with uniform

emissivity should provide relatively uniform radiation levels, and,

therefore, more steady temperature records.

The Hanauma Bay profile, Figure 16 reveals excellent general

temperature relations. Uniform land temperature with cool vegetation

change decreased about 4° to SOF over sea water. The noise, however, is

distracting and masks any local temperature anomalies.

Several solutions of the causes of the excessive noise are being

sought. Some noise is quite clearly instrumental, as shown by laboratory

testing, some may be related to the field conditions, and others may be

due to terrain radiation characteristics.

The instrumental noise appears to be caused by the detector or

preamplifier of the infrared thermometer and may be improved by compo­

nent or circuit improvements or by refrigerating the components during

use .

Turbulent air from helicopter blades and forward flight air motion

may cause the apparent increase in field noise over laboratory test noise

levels. Extension of the sensor protective tube and use of conventional

~~-f!!~.!""·1~""~:!~!.a-'.~!~~Q9~~Q'-)~99"<::::79 Q ~~_9!)99399~Q~9 <i 9 9 <i 999~

NUUANU VALLEY

FIGURE 14. INFRARED TEMPERATURE PROFILE OF RESIDENTIAL AND FOREST AREA IN NUUANU VALLEY. A CORRELATION WITHTHE TERRAIN SHOWN DIAG~~TICALLY. MAJOR FEATURES, SUCH AS HOUSE ROOFS , SHOW DISTINCTLY BUT DE­TAIL IS COVERED BY INSTRUMENTAL NOISE.

c.Nc.N

Vol.;::.

LEWERSa COOKE

HONOLULU

FIGURE 15. INFRARED TEMPERATURE PROFILE OF SEA, BEACH, PARK, AND URBAN AREAS. A DISTINGUISHABLE CORRELATIONIS EVIDENT BUT THE VARIABILITY OVER WATER, WHICH SHOULD BE MORE CONSTANT WITH ITS GREATER UNIFORM­ITY OF EMISSIVITY, IS EXCESSIVE.

35

75°COOL

KOKO HEADPENINSULA

FIGURE 16. CORRELATION OF THE TERRAIN TO INFRARED TEMPER­ATURE PROFILE. CONSTANT EARTH TEMPERATURESBROKEN BY COOL VEGETATION, AND BY SEA WATERBECOMES PROGRESSIVELY COOLER AWAY FROM SHORE.POSSIBLE THERMAL ANOMALIES IN SEA WATER AREMASKED BY NOISE.

a i r craf t may help to eliminate these sources of noisy response.

Af t er the conclusion of fi eld tests i t was learned that similar

problems of instrument instability and noisy response were solved by

Mr. M. J. Doyle, Jr . , of the Pacific Gas and El ect r i c Company . He mea­

sured warm water discharge from coastal power gener a t i ng plants. A

factory modification he reported effective was the installation of a

sealed detector to e l i mi nat e wind turbulence in the instrument. He

also installed a long wind isol ation tube on the sensor and used a light

Cessna aircraft gliding over the test ar ea to e l i minat e propeller tur­

bul ence. He has made successful recordings of 1 degree temperature va­

riations with these methods. Similar modifications to the Center's

instrument may a l so i mpr ove records.

A natural characteristic which may introduce noisy radiation records

over the sea is the specul ar reflection of sunlight by waves. Comparative

testing in sun and shade or at night should provide an evaluation of this

possibil i t y.

36

SUMMARY AND CONCLUSIONS

data on previously

flows which sig-

springs will provide

on changes in spring

fresh-water lens.

Hawaiian coastal springs are anticipated to be particularly amenable

to detection and mapping by synoptic sensors due to the large surface sa­

linity and temperature anomalies found by field investigation. Infrared

scanners have successfully delineated some such springs (Fischer, et aZ.,

1964) and visible and near visible optical methods are being pursued.

Laboratory testing and preliminary flight testing of the infrared

thermometer with a National Guard helicopter has shown that heat sensing

is applicable to rapid reconnaissance detection of coastal springs and

that with innovated accessory equipment, it is adaptable to flight porta­

bility.

Airborne mapping of offshore

unmeasured ground-water losses and

nal overuse of the Ghyben-Herzberg

ACKNOWLEDGEMENT

The authors gratefully acknowledge the valuable assistance rendered

by Hawaiian National Guard in providing the helicopter which made possible

the aerial traverses over the island of Oahu. The assistance of Generals

V. Siefermann and B. F. Webster in the arrangement of the helicopter

transport and Major Phillips, who piloted the aircraft, is deeply ap-

preciated.

37

BIBLIOGRAPHY

Ackermann, W. C., 1965, A look at data and instrumentation, remote sen­sing, AGU Transactions~ 46(4):702-704.

Antonini, G. S., 1965, Infrared image applications in studies of themarine environment, I n Proc. of the Third Symposium on RemoteSensing~ Univ. of Mich. Inst. Sci. and Tech., Oct. 1964, pp. 761­788.

Burt, W. V., 1956, A light scattering diagram, Mar i ne Res. J.~ 15:76-80.

Cameron, H. L., Currents and photograrnrnetry, In Oceanography from Space.(Proceedings of conference on the feasibility of conducting oceano­graphic exp l orat ions from aircraft~ manned~ orbital~ and lunar lab­oratories.) Woods Hole Oceanogr. Inst., Ref. 65-10, p. 341.

Clarke, G. L., 1965, Transparency, bioluminescence, and plankton, InOceanography from Space . (Proceedings of conf erence on the feasi­bi l i ty of conducti ng oceanographic expl orat i ons from aircraft~

manned~ orbital~ and lunar laboratories.) Woods Hole Oceanogr.Inst., Ref. 65-10.

Clarke, G. L. and James, G. R., 1939, Selective absorption of light bysea water, JOSA~ 29:43-55.

Clarke, J. R. and Frank, J. L., 1963, Infrared measurement of sea sur­face temperature, Under sea Technology~ pp. 20-23.

Colwell, R. N., 1961, Some practical applications of multiband spectralreconnaissance, American Scientist~ 49(1):9-36.

Colwell, R. N., et al.~ 1965, Basic of matter and energy relationshipsinvolved in remote reconnaissance, Photogram. Eng.~ 29(5):761-799.

Colwell, R. N., Draeger, W. C., Lent, J. D., and Thorley, G. A., 1966,A multispectral photographic experiment based on statistical ana­lysis of spectrometric data, School of Forestry, Univ. of Calif.,a final report.

Cox, Charles and Munk, and Walter, 1954, Measurement of the roughnessof the sea surface from photographs of the sun's glitter, JOSA~

44:838-850.

Cox, Charles and Munk, and Walter, 1955, Some problems in optical oceano­graphy, Marine Res. J.~ 14:63-78.

Ewing, G. C., ed., 1965, In Oceanography from Space. (Proceedings ofconference on the f easibility of conducting oceanographic explora­tions from aircraft~ manned~ orbital~ and lunar laboratories.)Woods Hole Oceanogr. Inst., Aug. 1964.

Fischer, W. A., Moxham, R. M., Polcyn, F., and Landis, G. H., 1964,

38

Infrared surveys of Hawaiian volcanoes, Science3 146(3645):733-742.

Fischer, W. A., Davis, D. A., and Sousa, T. M., 1966, Fresh water springsof Hawaii from infrared images, U. S . Geolog. Surv. Hydrol. Atlas3HA-218.

Gates, D. M., 1962, Radiation instruments, In Energy Exchange in theBiosphere3 Harper &Rox, pp. 74-93.

Kalle, K., 1961, What do we know about "ge1bstoffe?" Tn Symposiwn ofRadiant Energy in the Sea3 Aug. 1960, Helsinki, Jer1ov, N. G., ed.

Kohout, F. A., 1966, Submarine springs: a neglected phenomenon of coastalhydrology, In Symposiwn on Hydrology and Water Resources Develop­ment3 Ankara, Turkey, pp. 391-413.

Land, E. H., 1959, Color vision and the natural image, Proc. Nat. Acad.Sci' 3 45(1):115-129,636-644.

Legrande, Y., 1959, About theories of color vision, Proc. Nat. Acad. Sci' 345:89-96.

Lenschow, D. H., and Dutton, J. A., 1964, Surface temperature variationsmeasured from an airplane over several surface types, J. AppliedMeteor' 3 3(1):65-69.

Lohman, S. W., and Robinove, C. J., 1964, Photographic description andappraisal of water resources, Photogrammetric XIX 1962-19643 No.3, 21 p.

Mink, J. F., 1964, Groundwater temperatures in a tropical island environ­ment, JGR3 69(24):5225-5230.

Mo1ineux, C. E., 1964, Aerial reconnaissance of surface features withthe mu1tiband spectral systems, In Proc. of the Third Symposiwnon Remote Sensing of Environment3 Univ. of Mich .

Moore, J. G., 1947, The determination of the depths and extinction co­efficients of shallow water by air photography using color filters,Roy. Soc' 3 phil. Trans' 3 240A:163-217.

Morse, R., 1966, Boundaries of man's vision, Life3 61(26):105-112.

Ragotzkie, R. A., 1966, Airborne infrared radiometry study measuringthe surface temperature of Lake Superior, National Center for At­mospheric Research, No.2, pp. 2-3.

Robinove, C. J., 1963, Photography and imagery -- a clarification ofterms, Photogram. Eng'3 pp. 850-881.

Robinove, C. J., 1965, Infrared photography and imagery in water resour­ces research, J. Am. Water Works Assn' 3 57(7):834-840.

Skibitzke, H. E., and Brown, R. H., 1965, Remote sensing of hydrologic

39

phenomena, In Proc . of the Symposium on Remote Sensing~ Huntsville,Ala.

Stomme1 , H., 1963, Rap id aerial survey of the Gulf Stream wi t h cameraand radiation thermometer, Science~ 117:639-640.