Lidar observations during dusty infrared Test-1

Lidar observations during dusty infrared Test-1

Jagir S. Randhawa and Jan E. Van der Laan

Lidar measurements using ruby (0.7 -ttm) and CO2 (10.6-Mim) lidar systems during the dustry IR Test-i are

described. The test was conducted at the White Sands Missile Range (WSMR) in October 1978. Transmis-

sion comparisons are made between the two wavelengths through dust and smoke clouds generated by artil-

lery barrages, TNT explosions, and oil-rubber fire in a test zone midway (1 km) along the lidar path. A tar-

get at the end of the lidar path provided a reference backscatter return for the transmission measurements.

Results indicate that the broad particle size distribution present in the dust generated at WSMR produced

little if any wavelength-dependent transmission effects.

1. Introduction

Propagation of electromagnetic energy through theatmosphere can be adversely affected by atmosphericconstituents that scatter and absorb energy. Lidar(laser radar) offers the most promising approach forrapidly obtaining information about the optical and IRproperties of smoke and dust particulates dispersed overa test range. The characteristic narrow beamwidth andshort duration of the pulses transmitted by lidar makeit possible, by firing the lidar in different directions, tomap the general morphology of particulate plumes andhazes with considerable accuracy.1 Lidar may also beemployed to determine the extinction and backscattercoefficients of the aerosol clouds, provided certainauxiliary information is also available. 2 A field ex-periment was conducted at White Sands Missile Range(WSMR) to examine multiwavelength lidar techniquesfor measurement of optical properties of battlefield dustand smoke. This paper presents the results from thatexperiment where two lidar systems (0.7 and 10.6 Atm)were deployed to measure backscatter signals from dustclouds and from a passive reflector placed on the far sideof the clouds.

Jagir S. Randhawa is with U.S. Army Atmospheric Sciences Lab-

oratory, White Sands Missile Range, New Mexico 88002; and Jan v.

d. Laan is with SRI International, Menlo Park, California 94025.

Received 2 February 1980.

11. Experiment

Dusty IR test-1 (DIRT-1) was conducted by the At-mospheric Sciences Laboratory, 2-14 October 1978, inthe extreme southeast corner of WSMR, New Mexico.The terrain is gently sloping with large areas of mesquitecovered sandhills up to 2.5 m high and relatively flatareas covered with greasewood and other vegetationtypical of semiarid desert. The test area, however, wasgraded to remove all vegetation. The CO2 and ruby(Mark IX) lidar vans were positioned at the south endof the horizontal grid during the experiment.3 Speci-fications for the Atmospheric Sciences Laboratory(ASL), CO2 lidar, and Mark IX are given in Table I.The lidar systems were operated from generator powerrather than line power. A gray colored (10 X 5-m)target of very low reflectivity was placed at the far endof the grid so that the lidar-to-target path passed across

15 July 1980 / Vol. 19, No. 14 / APPLIED OPTICS 2291

Table 1. CO2 and Ruby Lidar Specifications

CO} idar Ruby Lidar

Transmi tter

Model

Type

Wavelength

Beam Diameter

Beam Divergence

Operation

Energy

PRF (Maximum)

Pulsewidth

Receiver

Telescope

Detector

PredetectionFilter

L~umo,ics Research,Inc.

TEA-10L-2

CO2

10600 nm

3.1 cm

1.2 mrad

Pulsed

250 mJ

1 pps

75 ns

30 c Newtonian

Honeywell HgCdTecooled photodiodc,100 Mhz bandwidthD*=1.3x10 1

cmHz1 /2 watt 1

None

0100,I

300

Ruby

694 nm

2 cs

0. 5 imrmad

Q-swi Lched

-1 J

1 lpS

30 ns

15 cm Newtonian

RCA 7265 PMT

05 A

_ .'155-mm

HOWITZER BATTERY

Fig. 1. Experimental configuration for two-wavelength lidar observations, DIRT-1.

the center of the impact zone (Fig. 1). The target pro-vided laser eye safety for distances beyond its location.It also provided a constant reflective surface so thatobserved lidar-reflected energy could be used to monitoratmospheric transmission over the lidar-to-target path.The two lidar systems were interfaced for the durationof the test, and the firing rate was set for every 2 sec witha 0.5-sec delay between the 10.6-Mum lidar firing and the0.7-yum firing.

The general lidar operating procedure used duringthe DIRT-1 program was to start lidar operation priorto an event to record a sufficient amount of pre-eventdata to establish reference conditions for backscatter

and transmission analysis. Operation would continuethrough the event and until backscatter conditions re-gained the pre-event values. Mark IX calibration wasperformed daily using optical attenuations, where asASL lidar used layers of low density polyethylene sheetsfor receiver linearity and dynamic range calibration.

111. Results

During DIRT-1, dust clouds were generated by ex-plosions from TNT charges, statically detonated 155-mhowitzer projectiles, and live firings of 155-mm howit-zers. Transmission data from the two lidar systems,when eight 155-mm projectiles were fired on the impact

2292 APPLIED OPTICS / Vol. 19, No. 14 / 15 July 1980

I . . a . .'1nnllfnrt11ror-lsllU-dl -U t:

11 I ~ ~~~~~I I I I _

00 10 1000 '2000 0 10601 00 20C

Fig. 3. Backscatter data at 10.6-,um for same event as in Fig. 2 (which is printed on page 2294).

zone in rapid fire, are shown in Fig. 2 (all rounds im-pacted within 20 see), thus indicating that both the rubyand CO2 signals were completely blocked for 40 secfollowed by a rapid recovery to clear air conditions. Itshould be noted that the receiver threshold for the twosystems is different, as can be seen during the blockageperiod. A polaroid photograph sequence of 10.6-Atmbackscatter data was also taken for this event and isshown in Fig. 3. Note that in these photographs largeaerosol return is observed in both the pretest andposttest clear air conditions; this return is due to in-creased humidity and haze on the morning of 13 Octo-ber. The small return seen at 100 m is off-axis reflec-tion from a reflector placed there to provide a range

mark reference, and the spike at the right edge corre-sponds to the gray target at 2 km. Figure 4 presents thetransmission data from another event when three155-mm projectiles, placed 0.6 m below the groundsurface and 15 m apart, were detonated. Transmissioncurves for both wavelengths are nearly identical. The10.6-,um backscatter sequence for this event is shownin Fig. 5. This particular sequence demonstrates theunique capability of the lidar to observe and distinguishbetween two (or more) distinct obscurant clouds. AtT - 30 sec, the dust cloud left over from an earlier testis seen to be still persisting at a 300-700-m range. AtT + 1 sec, the explosive cloud from the present event isobserved at a 1-km range. By T + 3 min, the dust


102

10

10-1

-1 0 1 2 3 4

TIME -- mn

Fig. 2. Transmission as observedby two lidars from dust cloudsgenerated by eight 155-mm how-

itzer projectiles, rapid fire.

Figure 3 is on page 2293.

I I . I

-3 -2 -1

0.7 pm

------- 10.6pm

0

TIME - mn

2 3

I Fig. 4. Transmission as observedI by two lidars from dust clouds -

generated by three 155-mm pro-4 jectiles (0.6 m deep, static

detonation).


C8

0

I

CU

I0Co

I-

12

10

10-1

I T I g g

- [ I t @ I s

I

I

Fig. 5. Backscatter data at10.6-pm for same event as

in Fig. 4.

I I I I I I I I0 100 1000 jnoo 0 100 1000 2000

RANGE - M RANGE

102

z

a

Iz0

U,zMC

10

10-1

Fig. 6. Transmission as observedby two lidars from dust clouds

generated by three 155-mmprojectiles

(surface, static detonation).

-3 -2 -1 0 1 2 3

TIME - min


14 OCTOBER 1978 T = 083610 MDT

tJ W0

0 -

M

0 100 1000)(''IANN I M

Fig. 7. Backscatter data at 10.6-!Lm for same event as in Fig. 6.

clouds from the event have blown out of the opticalpath, while the earlier cloud still persists at a 500-mrange. Similar transmission curves on both wave-lengths were obtained (Fig. 6) when three 155-mmprojectiles were detonated in a similar configuration butplaced on the surface instead of 0.6 m deep. Back-scatter photographs are presented in Fig. 7.

Analysis of the two-wavelength lidar data presentedhere indicates that little difference in transmission isobserved during these tests. The data are surprisinglyconsistent regarding recovery times, and the varioustwo-wavelength transmission feature changes (clouddensity variations) track very closely. This suggeststhat the slight differences in sample time (0.5 sec) andoptical path (2 m in test zone) do not present a sig-nificant problem in measurement of dust clouds. The

lack of wavelength dependence in the measurementsindicates that the particle size distribution in the gen-erated dust clouds at WSMR is quite broad and that arather substantial proportion of large particles (10AM) exists. This is consistent with preliminary sizedistribution measurements made during some eventsof the DIRT-1 program.3

Calculation of SNR as a function of range was madefor the ASL lidar to establish expected system perfor-mance. The backscatter signal as a function of rangewas compared with the receiver system noise equivalentpower (NEP) derived from the detector and receivercharacteristics. Comparing the system NEP with theanticipated received power, an SNR-1 was obtained ata 1-km range. Another event in the DIRT-1 programwas a diesel fuel, oil, and rubber fire burning from a


WIL0

_J 0

'000

Fig. 8. Backscatter data at 10.6-Mm and two-wavelength transmission obtained from diesel oil, rubber, and gasoline fire.

trench in the center of the test zone. Large volumes ofblack smoke were produced for a period of -37 min, butdue to the variable wind conditions, the plumes ofsmoke moved intermittently in and out of the lidaroptical path. Total blockage of the 0.7-gm transmissionwas observed quite often as the plumes crossed the op-tical path as seen in the transmission plot (Fig. 8). Thetransmission plot also shows that 10.6-gm transmissionwas never totally blocked. The backscatter record,which represents the maximum attenuation observedand is also shown in Fig. 8, was calibrated after the eventin clear air conditions and found to represent 12.2-dBattenuation (-6% transmission).

IV. Conclusions

Results of the DIRT-1 program indicate that thebroad particle size distribution present in the dustgenerated at WSMR produces little if any wave-length-dependent transmission effects. The few ob-served exceptions, where the larger 10.6-gm transmis-sion is indicated, generally can be explained by the

presence of wavelength-dependent smoke along theoptical path, which is also generated by the detonations.The oil-rubber fire generated dense black smoke thatwas totally opaque to the 0.7-gm lidar, while the 10.6-gim transmission measurements indicate a worse casetransmission of -6%.

The authors are indebted to E. Uthe of SRI Inter-national for providing helpful suggestions and assis-tance throughout the test program. Norman Nielsen,Rod March, and K. A. Ballard operated the lidar sys-tems during the test.

References1. E. E. Uthe and R. J. Allen, Opt. Quantum Electron. 7, 121

(1975).2. E. R. Murray, M. F. Williams, and J. E. Van der Laan, Appl. Opt.

17, 296 (1978).3. J. D. Lindberg, "Measured Effects of Battlefield Dust and Smoke

on Visible, Infrared, and Millimeter Wavelength Propagation,"Preliminary Report on DIRT-1, ASL-TR-0021, U.S. Army ASL,

WSMR, N. Mex. 88002 (Jan. 1979).


Documents

Lidar observations during dusty infrared Test-1