LUND UNIVERSITY

PO Box 117221 00 Lund+46 46-222 00 00

Laser-induced Fluorescence Monitoring of Vegetation in Tuscany

Edner, H; Johansson, Jonas; Svanberg, Sune; Wallinder, E; Bazzani, M; Breschi, B; Cecchi,G; Pantani, L; Radicati, B; Raimondi, V; Tirelli, D; Valmori, G; Mazzinghi, PPublished in:EARSeL Advances in Remote Sensing

1992

Link to publication

Citation for published version (APA):Edner, H., Johansson, J., Svanberg, S., Wallinder, E., Bazzani, M., Breschi, B., Cecchi, G., Pantani, L., Radicati,B., Raimondi, V., Tirelli, D., Valmori, G., & Mazzinghi, P. (1992). Laser-induced Fluorescence Monitoring ofVegetation in Tuscany. EARSeL Advances in Remote Sensing, 1, 119-130. http://www.earsel.org/Advances/1-2-1992/1-2_19_Edner.pdf

Total number of authors:13

General rightsUnless other specific re-use rights are stated the following general rights apply:Copyright and moral rights for the publications made accessible in the public portal are retained by the authorsand/or other copyright owners and it is a condition of accessing publications that users recognise and abide by thelegal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private studyor research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal

Read more about Creative commons licenses: https://creativecommons.org/licenses/Take down policyIf you believe that this document breaches copyright please contact us providing details, and we will removeaccess to the work immediately and investigate your claim.

Download date: 08. Jul. 2021

https://portal.research.lu.se/portal/en/publications/laserinduced-fluorescence-monitoring-of-vegetation-in-tuscany(983b75c1-6443-49c9-a43b-6fb382249a8c).htmlhttp://www.earsel.org/Advances/1-2-1992/1-2_19_Edner.pdfhttp://www.earsel.org/Advances/1-2-1992/1-2_19_Edner.pdf

EARSeL ADVANCES IN REMOTE SENSING, Vol. I, No.2 - II , 1992 119

Laser-Induced Fluorescence Monitoring of Vegetation in Tuscany

H. Edner, J. Johansson, S. Svanberg, E. Wallinder

Department of Physics, Lund Institue of Technology P.O. Box 118, S-221 00 Lund, Sweden

M. Bazzani, B. Breschi, G. Cecchi, L. Pantani, B. Radicati, V. Raimondi, D. Tirelli, G. Valmori

CNR lstituto di Ricerca sulle Onde Elettromagnetiche Via Panciatichi 64, I-50 127 Firenze, Italy

P. Mazzinghi

CNR Istituto di Elettronica Quantistica Via Panciatichi 56/30, I-50127 Firenze, Italy

ABSTRACT

Field experiments using laser-induced fluorescence were performed on different vegetation species in Tuscany, Italy. Remote sensing fluorescence lidar experiments were performed at Pian di Novello studying mainly Fagus Silvatica, but also some other species occurring in the beech forest. One of the fluorescence lidar operated on the third harmonic of a Nd:YAG laser (355 nm) while the other one was based on a excimer-pumped dye laser oper-ating et 480 nm. Both systems utilized optical multichan-nel detection. The lidar measurements were supplemented by 1luorescence point monitoring using a fluorosensor based on a nitrogen laser and an optical multichannel analyzer, and using a portable Be-Ne-based two channel instrument. High-quality remote spectra could be obtained from both lidar systems in distances ranging from 40-1 OOm. In addition, point monitoring on Picea Abies was carried out at the CNR Camporgiano research station on experimental plants subject to controlled stress condi-tions. A discussion of the optimum choice of excitation wavelength is performed.

and mapping of threatened areas. Satellites or airborne multi-spectral imagery provides excellent possibilities of area mapping with regard to major species. certain possi -bilities of forest damage assessment have also been de-monstrated, but the techniques seem to have limited applicability. Laser-induced fluorescence (LIF) provides a further spectral dimension for vegetation assessment (Lichtenthaler and Rinderle 1988) and research efforts have been directed to using LIF for improved vegetation characterisation (See e.g. , Cecchi and Pantani 1991, Hoge 1988, Moya et al . 1991, Pantani and Cecchi 1990, Svanberg 1990, Zimmermann and Gunther 1991). The concept of pont - and remote sensing monitoring of vegetation LIF is illustred in Fig. l. Examples of remotely recorded fluorescence spectra of different species are shown in Fig. 2.

In the present paper we report on the result of field test employing point-monitoring and lidar LIF on vegetation in Tuscany (Italy). The field experiments were per-formed at the end of September 1990. An important aspect of the work is lidar remote measurements on beech (Fagus Silvatica) in Pian eli Novello using two mobile systems. A photograph of the two measurement

INTRODUCTION systems at the test area is shown in Fig. 3. In addition to the measurements on beech, point monitoring of Picea

During recent years it has become obvious that forests are Abies at the CNR test site at Campogiano was carried endangered because of the influence of atmospheric pol- out. lution. There is a great need for methods of early detection

120 EARScL ADVANCES lN REMOTE SENS ING, Vol.l , No.2- II , 1992

LIDAR FLUOROSENSORS

POINT FLUOROSENSOR

PORTABLEFLUOROSENSOR

Fig. 1 -Artist's conception o(poinr and ron ore Ll F lll-' ~ -w ()

z w () (/) w a: 0 ::J ...J u._

>-' z w () z w () (/) w a: 0 ::J ...J u._

400

400

450 500

450 500

ROWAN

550 600 650 WAVELENGTH (nm)

550 600 650 WAVELENGTH (nm)

700

700

750

750

800

800

>-' z w () z w () (/) w a: 0 ::J ...J u._

400 450 500 550 600 650 WAVELENGTH (nm)

!ROE - FUDAR 2

700 750 800

16000~-~---~----~---~---~--

14000 ~- MYRTLE (Myrtuo communit>) _1_ CUVE-TREE (Oioa ouropaeo)

12000 l , fV( (Hudar:J hnflx) _1_ CHICORY (Cichorium lnty buo)

2_ LAUREL (Lauruo r.ob ilid

600 650 700 750

wavelength (nm) !ROE - CNR

Fig. 2 - Examples of".fluorescence spectra of" vegetation recorded 1vith a) rhe S\vedish 111ohile system (excitat ion wavelength 355 nm) and b) the Ita l ian mobile system (excitation wavelength 4RO nm).

Edner el a/.: Laser-Induced Fluorescence Mo.-"n-'-'it"'o-'-'ri-'-'n-"'o ________ _ 121 - ---

Fit;. 3 - Photograph of" two 111 obi/e f luorescence lidor svstem s a /m easurements site at Pian di Nove llo.

l. FLUORESCENCE EQUIPMENT

In the present measurements four diffe rent fluorosensor of capturing vegetation fluorescence spec tra were employed. For point measurements a multi -channel and a two-chan-nel instrument were used. Remote measurements were carried out wi th two independent mo bile fluorescence lidar sys tems.

1.1 Point monitors

1. ! .1 Multichannel f luorosensor

A multi-channel fluoro sensor was used fo r point monitor-

ing of the whole fluorescence spec trum or leaves and needles . The equipment has been described by ANDERS -SON -ENGLES et a/. (199 1) and a schematic di agram is

shown in Fig. 4. As an excitati on source a Laser Science V LS337 pul sed nitrogen laser (A = 337 nm) was used at a repetition rate of I 0 Hz. Either this laser was used directl y or as a pump source for a small dye laser (Laser Science Model VSL-DCM-3), which in the present measurements

was used to generate light at 405 nm or 470 nm . The laser light was foc used into a 600 m quartz fiber, that in the present experiment had a length of 5 m , but which could be extended to tens of meters. The fibre was pul in direc t contact with the pl ants to be studied and induced flu ores-cence was collected with the same fi bre and conducted back to the meas urement equipment. The fluorescence

light passes through a 45° dichroic mirror, that is highl y re ll ecting fo r the laser light and is used for injecting it into the fibre. A Schott high-pass coloured glass-filLer was used to full y reject elastically back-scattered laser light while pass ing the Stokes-shifted flu orescence light. The

fluorescence li ght was focused at the entrance slit of a Jarrel-Ash 0 .275 m polychromator with a 400 m entrance slit.

122 EARSeL ADVANCES IN_ REMOTE SENSING, Vol. I , No.2 -ll, 1992

ELECTRICAL SIGNALS TO

AND FROM OMA MAINFRA~~E

DETECTOR

POL YCHROMATOR

LIGHT TO ANO

FRmA TARGET

OPTICAL FIBRE

CUT-OFF=;~ FILTER

LENS DICHROIC

MI RROR

NITROGEN LASER

FLIP-IN MIRROR

DYE LASER

Fig. 4- Schemwic diagralll u,ftlte Lund Jillilll monitoring mullichanne/ jluorosensor.

In the focal plane of the spectrometer an image- intensified diode-array detector (EG&G Model 1421) was placed covering the wavelength range 300-800 nm. The captured fluorescence spectrum was transferred to an EG&G OMA mainframe. The spectra were stored on floppy di sk for later evaluation usiong a specially developed software

package.

In the field measurements the t1uorosensor was trans-ported in a van and electricity was provided from a small, portable motor generator.

1.1.2 Portable two-channel j luorosensor

Point measurements were also pelformecl with the proto-type of the portable, battery operated, two channel fluorometer, developed at IEQ and named LEAF (Laser

Excited Automatic Fluorometer). LEAF uses a Hc-Ne laser (0.5 mW A = 633 nm) as a fluorescence inducing source and a single optical fiber both to carry the exc ita-tion light to the leaf and to collect fluorescence. The returnin g beam is separated from the laser beam by a set of dichroic mirrors and interference filetrs (Fig. 5). The detectors are two miniature-size , head-om photomulli -pliers (Hamamatsu) , with S20 photocathode for a good sensitivity in the red spec tral region. The instrument con-tain s all the detection electronics, including the analog ampli fiers, sample-and-hold circuits and 12 bit. s analogue-to-digital converters. A microprocessor board acquires data and stores them in the internal RAM, for later dump via a RS 232 seri al in terface. The internal battery is sufficient for 4 hours of continuous use, but an auto power off software turn s off the instrument when not in use. Durin g field operation thi s battery capac ity was found sufficient for a fully clay of operation.

Edner et al.: Laser-Induced Fluorescence Monitoring

~--------------------------,

1 DM

I I I I I I I I I I I

LASER

MICRO PROCESSOR

BATTERY

i li~~~~~~~~~~~ I INSTRUMENT BODY L--- ----- --------- - --------"

Fig. 5 - Schenwtic diagram of CNR-IEQ portable Iwo-channel fluorosensor. FC~Fiber Coupler; DM~Dichroic Mirror; TM~Totally reflecting Mirror; IF=In!e~ference Filter; AF=Ab-sorption Filter.

LEAF can be measure chlorophy 11 fluorescence at 685 and 730 nm, with a band-pass pf TA = 5 nm. In spite of the different excitation wavelength it can perform the measurements of the F685fF370 ratio as the high resolu-tion system. Its low weight (5.6 Kg, battery included) and its small waterproof aluminium case /37x29xll cm3) make it very useful as a "ground truth" instrument for the remote sensing systems and its simplicity is an advantage.

1.2 Mobile fluorescence lidar systems

1.2.1 Nd: YAG-based system

The Swedish mobile fluorescence lidar system employed in the present measurements is based on the remote sensing system described by ENDER at al. ( 1987). This system has been mainly beenused for atmospheric pollu-tion monitoring as reviewed by SV ANBERG (J 991 ). For remote fluorescence monitoring the point multi-channel fluorosensor described above was connected to the output of the receiving optical telescope as also described by EDNER eta/. (1991).

The system adopted to vegetation fluorescence measure-ments is schematically shown in Fig. 6. The lay-out of the optical and electroning system is given in Fig. 2 of the just mentioned reference (this volume). A short description of the system will be given here.

Fig. 6- Lund mobile fluorescence lidar systetn.

123

The system is housed 111 a covered truck and eklectric power is supplied by a 20 KV A diesel motor generator, towed by the truck. As a laser transmitter a powerful pulsed Nd:Y AG laser (Continuum Model YG 682) with a primary IR pulse energy of 1.2 J is used at a repetition rate of 20Hz. For inducing fluorescence it was operated at its third harmonic (A= 355 nm), where output pulse energies up to 250 mJ were available. The system is equipped with a Continuum Model TDL 60 dye laser, which is capable of producing pulses of about 20 mJ energy at 480 nm, where chlorophyll-a absorbs strongly. for practical rea-sons only the Nd:Y AG third harmonic was used in the present experiments. The UV laser are transmitted into the atmosphere via a large first surface aluminum mirror on the top of the lidar van. With this cpmputer-controlled mirror the beam is directed onto the object under study. Fluorescence light induced in the target is reflected by the same minor down into a vertical Newtonian telescope of the diameter 40 em. In the image plane of the telescope a 600 m diameter quartz fibre is placed to collect fluores-cence light and guide it to the 400 m wide entrance slit of the optical multichannel analyzer. The fibre is placed in the center of a small hole in a metal mirror, that directs the rest of the light collected by the telescope to a TV camera, used for aiming the system. On the TV screen the hole appears as black circle indicating the aiming point of the laser beam. The recording of the who 1 e fluorescence spectrum from about 370 nm to 800 nm follows as de-scribed above in connection with the multi-channel point monitor. ln order for the system to sensitively detect vegetation tluorescence in the presence of ful I sun light the multichannel plate of the image intensifier is gated to a time window of about 500 ns centered at correct time delay con·esponding to the target distance.

1.2.2 Excimer-based system

The Italian nobile fluorescence lidar system used in the present experiments is basically the system described by CECCHI eta/. (1991) . It employes a XeCl pumped dye laser, operating at 480 nm as transmitter. The excimer laser can operate at 10Hz with an output energy of 80 mJ. The dye laser conversion efficiency is about 10 percent. Mounted together with and aligned with a Newtonian telescope of diameter 25 em, the integrated tluorosensor arrangement can be directed toward the target. Collected fluorescence light is focused into a fibre bundle and brought to the entrance slit of an intensified optical mul-tichannel analyzer system. Spectra are stored on a hard disk, already preprocessed, for further analysis. During the field test the equipment was installed in a temporary vehicle awaiting the completion of the specially arranged

-'-'t 2=--4:__ _ _ _ ____ ___ ___ _ _____ E-:cA~R~S=eL ADY ANCES IN REMOTE SENSING, Vo l. l, No.2- II, 1992

permanent vehicle. A schematic diagram of this fluores -cence lidar system is shown in Figure 7.

I 1 OPTICAL MODULE I

FLIDAR- 2 I I OMA I I : i\ECTOR I SPECTROMETER I

: [TEL~ ~G TARGET I -----------

LASER BEAM

I I I IJb:J UV LASER I LASER L _ ---1--~! -- --- -"" "= I - 1

I I I HV POWE:SUPPLY h ------ ------I I I I

I I I I I

~~~ I I I

I 1-I I I

I 24 VDC I I He Ne Xe HCI I . I I I I

I /OL \u~ I I I OMA

I I I I I ELECTRONIC MODULE I I GAS HANDLING MODULE I L I I_ I

Fig. 7- Schemalic diagram (!l CNR-IROE mobile j!uoresce11< c lidar svsrem.

2. REMOTE AND POINT MONITORING MEASUREMENTS IN PIAN Dl NOVELLO

In Pian eli Novello, close to Abetone Pass in northern Tuscany there are wide-spread forests of beech (Fagus Silvatica). Our measurements were performed at the beginning of the senescence period when fully green as well as ye llowish leaves were ava il able . The field ex-periments were started by point monitoring of beech leaves using the two point-monitoring devices described above .

As a starting point we consider the respon se of green leaves to excitation light of different wavelengths. Spectra were recorded using the multi-channel instrument and examples are given in Fig. 8. Using the Nrlascr-pumpcd dye laser, spectra were taken for 405 and 470 nm, where chlorophyll has a strong absorption and the radiation has a good penetration of the leaf. Correspondingly strong chlorophyll spectra featuring the characteristic peaks at about 690 and 735 nnl. The ratio 1(690 nm)/1(735 nm) is of special interest, since it is an indicator of the proper functioning of the photosynthetic apparatus (LICHTEN-THALER and RINDERLE 1988). Generally speaking a low ratio is obtained for a healthy plant while hi gh ratios may indicate stress conditions. However , the ratio is also intluenced by the amount of chlorophyll present and thus changes between the upper and lower side of a leaf It is

also strongly dependent on the ambient light level and other factors and thus no straight-fo rward relation be-tween the ratio and the health status exists, but additional inrormation is needed.

,..: z w () z w () (f) w 0: 0 ::l ...J u.

,..: z w () z w () (f) w 0: 0 ::l ...J u.

,..: z w () z w () (f) w 0: 0 ::l ...J u.

400

400

400

450 500

450 500

450 500

FAGUS SILVATIGA

550 600 WAVELENGTH (nm)

550 600 650 WAVELENGTH (nm)

550 600 650 WAVELENGTH (nm)

700 750 800

700 800

700 750 800

Fig. 8 - Point moniwring fluorescence spectra of beech leaves (Fig us selvarica) measured or diflerent exciwrion wave/engrhs using rile Lund mulrichannel syslem.

Edner er a/.: Laser-Induced F_::l:::_u~o r...::·c:::s :::cc::.:n.:.::c:.::c_:M.:..:.::o:.:.nr:.:_·ro:::r_:·ir:.:.lg,__ ______________ _ 125

It can be noti ced from Fig. 8 that the broad-band fluores-cence in th e green spectral region ga ins in importance for the shorter exci tation wavelength. For 337 nm excitation no chl orophyll 11uorcscence is induced, which is a com-

mon observation for most species (See e.g. CELANDER et a f. 1978, C HAPPELLE 1990, S V ANBERG 1990). The explanation is partly a spec tra l mismatch with the chlorophyll absorption but also reduced penetration into the leaf because of superficial '·wax··. For beech the wax is only weakly tluorescent as seen in Fig. 8. (Actually , I 0

times as many lase r shots \>.' ere used for recording the 337 nm curve than for the other two). Normally the wax fluorescence is very strong, as e.g. seen in Fig. 14.

The change in f luorescence ratio between the upper and lower sides of beech leaves is illu strated in Fig. 9 showing

recordings with the two-channel point monitoring system.

1.4.------------ -1.3

~ : :~ ~ 0 .9 ::: 0 .8 -~ 0.7 g 0.6 ~ 0.5 :g 0.4 -~ 0.3

Fagus sllvatlca

o.z .... ~=-:::=;.--... 0.1

o +-.-~~~~~

UPPER SIDE

o r t.as 1 ;:-JJo + r 7Jo

• F 685

PAR = 0

Fig. 9 - Poilllmon ituring oj"upperwullowel'.l·ides r!lheeclr lew ·es t.tsing rlre CNR-!EQ two- channel jluurosensor. Individual chan -nel intensities and rhe ralio of the two channel readings are p/oued.

In the remote sensing experiments measurements were performed on leaves of young beech trees growing at the edge of an open area in the forest. The llu orescence lidar systems were placed as shown in the photograph in Fig. 3 shooting across the open area to hit targets in 40-60 m

distance.

Examples of remote spectra recorded with the Nd: Y AG-

based system O"cxc = 355 nm) are shown in Fi g. 10 using a measuring distance of 60 m and a pulse energy of 25 mJ. Each spec trum was averaged for 500 laser pul ses. The divergence of the exciting lase r beam was chosen wi th a

beam expanding telescope to a spot of selec table size, in our case with about 10 em. di ameter. Thus, 2~3 leaves were covered by the laser spot. Spectra are shown for green and yellow shadow leaves, featuring the characteris-

tic ratio increase for the senescent leaves. In both spec tra a s ignal possibly due to carotenoids is shown in the spec-tral reg ion at 520 nm. Included in Fi g. I 0 arc also spectra for the dead, brown leaves covering the ground of the

beech fo rest and for the grey bark of the living beech trunks. It can be seen that the beech bark f luorescence is very st rong and whiti sh and also features a chlorophyll signal. For airborne meas urements thi s might be a problem and needs some consideration s.

,__.: z w 0 z w 0 (f) w a: 0 ::> _j

LL

~TE LIF BEECH SPECTRA

Vl ~~ 355nmEXC

25 mJ/PULSE BARK 60 m DISTANCE

YELLOW LEAF

400 450 500 550 600 650 700 750 800 WAVELENGTH (nm)

Fig. I 0 - Erwnples 1~/ rellwleflunrescence recordings 11{ beec!t frees ar Pian di No vello using the Lund mobile fluorescence lidar svstem.

With the same system also passive remote monitoring of the leaf refl ec tance was performed with the laser switched off. The spec trall y resolved reflected intensity from the leaves was first recorded and stored and then the sky light spectral distribution under the same circumstances was

recorded by directing the telescope viewing to a first sur-face aluminized 30x40 cm2 min-or angled to receive the verti cal impinging light. By di viding the leaf spectrum with sky spectrum an approx imate recording of the leaf reflec-tance is obtained. The results for green and yel low leaves are shown in F ig. II , showing the expected increased inten-

sity towards the red spectral region. A prominent feature is the strong leaf refl ectance in the ncar IR reg ion, where the chlorophyll docs not absorb any longer. In assessing forest decline the combination of using active (laser) and passive remote sensi ng may prove to be important.

126 ___________________ E_A_R_S_e_L_A_D_ Y_A_N_C_E_S_-_!N_R_E_M!JTE SENSING, Vol. I, No.2- Il, 1992

SKY

400 450

400 450

REMOTE REFLECTANCE OF BEECH LEAF

LEAF

500 550 600 650

REFLECTANCE- !(LEAF) !(SKY)

500 550 600 650 WAVELENGTH (nm)

t H,O

700

700

t o,

H20

750 800

750 800

Fig. II - Remole recording of green leaf rejleued imensily and quasi-simultaneous sky intensity distrihution for a beech free at Pian di Novello. h1 the power pari of" !he figure !he resultinf? reflectance curve is shown for the green leaves andalsofor yellow leaves. The recording.1· were taken with the Lund .fluorescence lidar system operating in ils passive (non -laserj mode.

Examples of remotely recorded spectra obtained with ex-cimer-based fluorescence lidar system are given in Fig. 12 for 40 m measuring distance.

Simultaneous measurements on the same leaves using both fluorescence lidar systems were performed. Ex-amples of such recordings are given in Fig. 13 for green and yellow shadow leaves. A quite detailed and very satisfying agreement between the chlorophyll signals re-corded wi th the two lidar systems is obtained.

Night-time remote recordings on green shadow leaves were taken for the case of dark adaption following li ght exposure of the leaves from a flash-light. Signals obtained with the two fluorescence lidar systems investigating the same leaves are shown in Fig. 14.

Pian di Novello 10/10190 woo,---~--------~---.----,--,

3500 -

3000 - - __ __ __ _.: ________ - -- -green-leaf

2500 .. yell~h-_leaf

"' § 2000 8

1500

650 700

wavelength (run)

Fig. 12 - Examples of green and yellow beech leqf' spectra recorded remotely with the CNR- JROEjluoresccnce Lidar system.

355 nm EXCITATION

YELLOW LEAF

LJ 0

400 450 500 550 600 650 700 750 800 WAVELENGTH (nm}

GREEN LEAF

~ z w 0 z w 0 (}) w a: 0 :;:J ...J u..

400 450 500 550 600 650 700 750 800 WAVELENGTH (nm)

Edner et al.: Laser-Induced Fluorescence Moni torin o

480 nm EXCITATION

500 550 600 650 700 750

wavelength (nm)

GREEN LEAF ., .5

~

H g [i;

500 550 600 650 700 750

wa..:length (nm)

Fig. 13 - Comparison between simultaneous remote recordings oftlze same green and yellow leaves obtained with the two mobile f luorescence lidar systems.

The difference obtained with the two lidar systems might be due to the two different excitation wavelengths, and/or an actinic effect due to the 20Hz 355 nm excita-tion.

3. POINT MONITORING MEASUREMENTS ON PICEAS ABIES AT CAMPORGIANO

In addition to the remote flu orescence lidar measurements on beech trees in Pian di Novello point monitoring on spruce plants (Picea Abies) was performed at the CNR test site at Camporgiano (Garfagnana). The studied plants had a height of about 80 em, and consisted of normal reference specimen and plaqts that had been subject to detergent (ABS) exposure affecting the wax surface and/or had been subject to water (dehydration) stress. In particular the water stress was tenninated six months before experiment, while the ABS stress was on going (once a week).

.....: z UJ u z UJ (.) (f) UJ a: 0 ::J __J

u..

.....: z UJ (.) z UJ (.) (f) UJ a: 0 ::J __J

u...

400

400

4000

3500

3000

2500

1500

1000

450

450

GREEN LEAF WITH LAMP

500 550 600 650 WAVELENGTH (nm)

GREEN LEAF WITHOUT LAMP

500 550 600 650 WAVELENGTH (nm)

700

700

Pian di Novello 10/10190 - Leaf 16

17:35 lamp on 17:38 lluDp off

650

wavelength (om)

700

127

750 800

750 800

750

Fig. 14 - Remote night-time flu orescence recordings of green beech leaves. Spectra we re recorded in darkness and during light exposure with a flash light. a) Recordings·wirh the Lu1ul system (/exc-355 nm) . b) Reco rdings with rh e CNR-IROE system ( lexc=480 nm).

128 _ _ _ __________ E_~A_R--"--Se:_L_A_D_V_A_N_C~ES IN REMOTE SENSING, Vol. I , No.2 - II, 1 99~

Spectra for different excitation wavelengths are shown in Fig. 15 as recorded with the multi- spectral fluorosensor. We have already commented on the strong wax fluores-cence and the absence of chlorophyll iluorescence when

using 337 nm excitation .

f.-' ~ w 0 z w 0 (/) w a: 0 ::> ..J u..

f.-' ~ w 0 z w 0 (fl w a: 0 ::> _J

u.

f.-' ~ w 0 z w 0 (/) w a: 0 ::> _J

u..

500

500

350

550

550

400

PICEAABIES

600 650 700 750 800 WAVELENGTH (nm)

600 650 700 750 800 WAVELENGTH (nm)

450 500 550 600 650 700 WAVELENGTH (nm)

Fig. 15- Point monitoring qt:fluorescence spectra ofPicea Abies obtained with the Lund 1nultichannel .fluornsensor ope raring at three different wavelengths.

In Fig. 16 data on the ratio of the two chlorophyll iluores-cence peaks intensities 1(690 nm)/1(735 nm) are give n for a healthy plant and an even visually damaged pl ant su bj ect to both detergent and wate r stress. Resu lts obtained with

the two point monitors are presented in separate diagrams. In these meas urements the multi-spectral in strument was operated with 470 nm excitati on. The tluorescence data can here clearly distingui sh the two plants [rom each other as cou ld the naked eye.

2.0~------;::::::====;---------~

1470 nm exc I

1.5 X

(/) X w (/)

Q (/) 0 X

X

a: 0 X

in 1.0 STRESSED cry

!::::-""-

CONTROL 8 m ~

0.5

0.0+-----,,-----,---- -.-- --.-----l 2 3 4

PLANT GROUP

2r----- - - ------- ----- -, 0

"' r--u.. ~ ;n ~ 1.5 li.

0 z

~ 0.5 0 :::J _J

li.

DAMAGED PLANT

685 nm 730 nm ·--EJ--- ·····b····

HEAL TY PLANT

Fig. 16- Comparison betweenfluorescence datafo r a normal and even visually strongly stressed spruce plan/. a) Daw obwined with the Lund multichannel point-monitoring j luorosensor (lexc=4700 nm). b) Data obtained with the CNR-!EQ two-chan -nel poinl-lnonitoringfluorosensor (lexc=633 nm).

Measurements were performed on four classes of plants as shown in Fig. 17 for an irradiation corresponding to partl y cloudy ski es (PAR values were also measured). The excitati on wavelength was chosen to 405 nm and

one spec trum recorded for each of 5 plants in each group. The different pl ants could not be visuall y distinguished from each other. The ch lo roph yll peak ratio 1(690 nm)/1(735 nm) as well as the blue to reel nuorescence

Ed ner eta/.: Laser-Induced Fluorescence Monitorin g

rati o 1(490 nm)/1 (735 nm) are given. As can be seen the

stati stical spread is quite large . Two more series of data of the same kind were recorded for othe r li ght intensiti es,

and again a large sta ti stical scatter was ev ident. Using

337 nm excitat ion no chlorophyll is exc ited. However, th e bluish wax flu orescence does not provide any marked demarcation either. The conclu sion from these data is that , based on fluorescence ratio measurements

alone , o nl y strongl y stressed plants can be distingui shed from healthy ones. This is in accordance with earlie r

finding s by ANDERSSON- ENGELS et al. ( 1988) , who investigated young Swedi sh spruce and pine plants sub-ject to ozone stress.

1405 nm exc. I

0.5

X

2.0,-- ----------------,====rr

X

0.4 1.5

X X

{f) X X ~ {f) w {f) t')

0

X X X w ;::; ;::; a: ..:

X 0.3 :::J a: 0 X X

0 (/)

if) 1.0 if) C') C')

t::. t::. 0

0.2 "'-0

m ~

0.5

0 m 0 0

~ 0

[l

0 0.1 0 0 ~ 0 tl ~ 0

0

0.0+------.-------.------.------.-------+ 0.0 2 3 4

PLANT GROUP

129

tion on the wax layer coverin g the leaf. The 355 nm output from a frequency trip led Nd:YAG laser is readil y ava ilable at high power levels and is eyesafe in practical measurements , but cannot exc ite the chl orophyll f1 uores-

ce nce in sun lea ves of beec h trees. An opt imum wavelength in view of eye-safety consideratio n mi ght be

just below 400 nm.

The 633 nm excitation is also useful for compari sons with many ex isting data (LICHTETHALER and RlNDERLE 1988) on F685/F730 ratio changes due to different chlorophyll concentra tion and/or stress states. The red wavelength is also a suitable probe of chlorophyll fluore s-cence since it directly excites the first triplet states of

ch lorophyll molecule, without any tran sfer from other pigments. However thi s wavelength is not very suitable for lidar operation for the above-mentioned reasons of eye-safety and for the poor availability of high peak power red lasers .

Point measurements on spruce subjec t to dehydration stress and/or detergent ex posure did not provide any stati s-tically sign ificant demarcation be tween stressed and nor-mal plants. Only when changes were visible to the eye a flu orescence demarcation was also observed. For the re-

mote beech monitoring c lear fl uorescence Jiflcrcnces were a lso obse rved when the vis ible co lor o f the leaves changed from green to ye ll ow. However, the material collected for beech was not well controlled enough to

Fig. 17 - Fluorescence data on4 classes of Picea Abies, su!Jjecr make any conclusive observation on the poss ibilities to to difFe rent environmental conditio11s. Class 1: 11ormal referellce detect stress on beech using fluorescence. plants, Class 2: water stress, Class 3: deterge11t (A BS) stress, Class 4: warer and derergent stress. Excitation wavelength was 405 11111.

4. DISCUSSION

Valuabl e experi ences in remote fluorescence monitoring was obtaj ned in the fi e ld experim ents described in the

present paper. High-quality remote spec tra co uld be re-corded with both mobil e sys tems . Measurements ranges of 40-100 m we re use d. Ex c itation in the blue wavelength region yie ld s optimum chl orophyll flum·es-cence wh ile a shorte r wavelength also provides informa-

LIF measurements directly inte rfere with the photo syn-thetic process and thus mi ght have the potential to reveal stress conditions to vegetation before these are de-tectable by convention a l remote-sensin g tec hniques

based on exte rnal charac te ri stics of the pl ant such as spectral reflectance . The present study shows that early

de tection of stress is quite difficult. Probably it is neccs-. sary to use different pi eces of infonnation coll ected in fluoresce nce measurements (a t diffe re nt excitation waveleng ths, and possibly al so time-reso lved) and in

reflectance measurements to assess early damaged to fo rests. In view of the paramount importance of such an assessment the effort is motivated even if no easy way

directly presents itse lf.

::ci3::...:0::__ ________ _ _ _______ _:E::.:_A_:_:R_:_:S::::e-:::L:_-A_:_:D--_V_.__,_,A::_N:..::C-:::E~S IN REMOTE SENSING, Vol.! , No.2 - II , 1992

ACKNOWLEDGEMENTS

The authors gratefully acknowledge collaboration with the Plant Physiology group of Prof. E. Giordano, Universit della Tuscia, and the technical assistance by P. Olsson. We also would like to thank Dr. P. Raddi , CNR, Firenze and Prof. R. Giannini , Universit di Firenze, for kind support. This project was performed within the framework of the EUREKA Project LASFLEUR and was supported by the Swedish Natural Science Research Council, the Swedish Board for Space Activities and the Consiglio Nazionale delle Ricerche - CNR-RAISA.

REFERENCES

Andersson-Engels, S., Callander, K. , and Galle, B., 1988. Inves-tigation of the possibilities to use laser- Induced fluorescence to map conifer fo rest damage caused by ozone, IVL Report L88/l46 (Goteborg: Swedish Environmental Research Inst itute).

Andersson-Engels , S. , Elner, A, Johansson, J., Karlsson , S.E, SALFORD, L.G., SVANBERG, K., and SVANBERG, S. , 199 1. Clinical recordings of laser-induced fluorescence spectra for eval-uation of tumor demarcation feasibility in selected clinical spe-cialities, Lasers in Medical Sciences, in press.

Cecchi, G ., and Pantan i, L. , 1991 , Vegetat ion monitoring by mea ns of spectral resolved fluore scence lid ar, in Physical Measurements and Signatures in Remote Sensing, Courchevel, France, 14-1 8 Jan. , 1991 , Proc. of the 5th Int. Colloquium , (ESA SP-3!9, May 1991 ) 2, p. 687.

Cecch i, G ., Pantani , L., Breschi , B., Tirelli , D. , and Valmori, G. , 1991. Flidar: a multi -purpose fluoros ensor/spec trometer, in PAN-TANT,-L. , and REUTER, R. (eds.), this volume.

Celander, L. , Fredriksson, K. , Galle, B ., Svanberg, S. , 1978. Investigation of laser-induced fluorescence with application to remote sensing of environmental parameters, (Goteborg, lnst. of Physics Rep01t GIPR 149, 1978).

Chappelle, E.w., Wood , F.rn ., Newcomb, W.w., and Mc.Murtrey Ill , J.E. , 1985 . Laser-induced fluore scence of green plants. 3: LIF spectral signatures of f ive major plant types, Appl.Opt. 24, 74.

Chappe lle, E.w., M cmunrey, J.e., Kim, M.s. !990. Laser induced

blue fluorescence in vegetation , in IGARSS '90, Remote Sensing Science for the Nineties, 3, 191 9-22 .

Edner, H., Fredriksson, K., Su nesson, A:, Svanberg, S., Uncus. L. , and Wendt, W., 1987. Mobil e lidar system for environmental monitoring, Appl. Opt. 26, 4330.

Edner, H. , Johansson, J., Svanberg, S., Wallinder, E. , Cecchi , G. , and Pantani , L. , 199 1. Fluorescence lidar monitoring of the Arn o River, in PANT ANI, L. , and RE UTER, R. (eds.), thi s volume.

Hoge, F. E. , 1988, Ocean and ten·estri al remote sensing, in MEAS-URES, R.M. (ed.), Laser Remote Chemical Analys is (New York: Wiley-Interscience).

Hunt, J..I. , (ed. ), I 99 1, Physica l Measurement s and Signatures in Remote Sensing, ESA SP-3 19 (Noordwijk: ESA).

Lichtenthaler , H.k. , and Rinder le, U., 1988 , The rol e of chloroph yll. fluorescence in the detection of stress conditions in plants. CRC Ctitical Reviews in Analytical Chemi stry 19, Suppl. 1,29.

Mazzinghi , P. , 199 1. Leaf, a fi bre-optic fluorometer for field measurements of chlorophyll fluorescence, SPIE 1510 (Belling-ham: SPIE) p. 18. LEAF is a registered trademark ofCNR. Patent pending.

Moya, I. , Goulas, Y., Guyot, G., Bertolini, G. , and Schmuck, G. , 1991. Contra I e de !' tat de co uvert s vgtaux par tldtec tion des mission de fluorescence dan s le rg ions bleue et rouge du spectre visible, in Physical Measurements and Signatures in Remote Sensing, Courcheve l, France, 14- 18 .Jan. , 1991 , Proc. of the 5th ln t. Coll oquium, (ESA SP-319, May 199 1) 2 p. 68 1.

Pantani , L. , und Cecchi, G. , J 990. Fluorescence lidars in en-vironmenta l remote sensing, in Martellucci, S., and C Hester , A.N., (eds.) Optoelectronics for Environmental Science, (New York: Pl enum Press), 13 1.

Svanberg, S., 1990, Laser llu orescence spectroscopy in en-vironmental monitoring, in Marte llucc i, S., and Chester, A.N ., (eels.) Optoelectronics for Env ironmental Science, (New York: Pl enum Press), 15 .

Svanberg, S. , 199 1, Environmental monitoring using optical tech-niques, in Inguscio, M ., and Demtroder, W ., (eds.), Applied Laser Spectroscopy (New York: Plenum Press) .

Zimmermann , R ., and Gunth er, K.p., 1986. Laser-induced chlorophyll- a fluorescence of terres trial plants, IGARSS '86, ESA 5P-254 (Noord wijk: ESA) 3 p. 1609.

1.2_19_0011.2_19_0021.2_19_0031.2_19_0041.2_19_0051.2_19_0061.2_19_006b1.2_19_0071.2_19_0081.2_19_0091.2_19_0101.2_19_011