Fluorescence Responses from Nitrogen Plant Stress in 4 Fraunhofer Band Regions
J.E. McMurtrey
1, E.M. Middleton
2, L.A. Corp
2, P.K. Entcheva Campbell
2, L.M. Butcher
2, E.W. Chappelle
2and W. B. Cook
31Hydrology & Remote Sensing Laboratory, Agricultural Research Service, USDA, Beltsville, MD 20705
2Biospheric Sciences Branch, Laboratory for Terrestrial Physics, NASA/GSFC, Greenbelt, MD 20771
3Electro-Optics & Controls Branch, NASA Langley Research Center, Hampton, VA 23681
Abstract- The potential of solar Fraunhofer line features centered at 532, 607, 677 and 745nm for tracking changes in plant canopy chlorophyll content and photosynthetic capacity was studied. Excitation wavelengths similar to full sun light were considered. Canopy changes were tested experimentally by monitoring treatments of plant stress due to nitrogen application rate in corn. Corn leaves were obtained from field plots that were given different nitrogen application rates at 20, 50, 100, and 150% of optimal N in 2001. The data infers that leaves in plant canopies that have the greatest photosynthetic performance potential can be identified. Information collected in Fraunhofer regions compared favorably with data taken by laser induced fluorescence excitation and detection methods in peak emission areas. The technique may be useful in projecting what can be expected if a space-born interferometer type sensor can be developed for capturing plant canopy fluorescence.
I. INTRODUCTION
This study details the fluorescence responses of 4 Fraunhofer band regions that result from plant stress due to nitrogen (N) application rate. The 4 spectral regions span ultra-narrow Fraunhofer line features centered in the green spectrum at 532 nm, the yellow at 607nm, the red at 677 nm, and the far-red at 745 nm. In active fluorescence sensing studies, the emissions observed in the green band has been useful when combined in ratios with those obtained from the red and far-red bands for distinguishing differences among vegetation subjected to a range of stress levels [1]. Emissions produced in the 607 nm yellow band are least effected by plant physiology activity and are minimal. Thus they are suggested as a reference band to normalize noise effects and to express the true differences in vegetation fluorescence emissions due to stress responses. Fraunhofer band regions are under study because they have potential for use with a future interferometer-type passive satellite system. These Fraunhofer band regions (centered on 532, 607, 677, & 745 nm) were chosen because they represent wavelengths where little or no solar energy reaches the Earth’s surface, and where atmospheric absorption is minimal. To be effectively sensed by a space-based sensor, the fluorescence signals produced by vegetation must penetrate back out through the Earth’s atmosphere. Fluorescence emissions occur at wavelengths greater than the incident excitation wavelength and emissions observed in Fraunhofer lines above the Earth’s atmosphere can be attributed to fluorescence instead of apparent surface reflectance. Passive instruments capable of
detecting solar stimulated chlorophyll fluorescence were first built to detect solar absorption features at Fraunhofer lines.
They measure luminescence based on the Fraunhofer line depth method [2,3,4]. A prototype Fraunhofer line discriminator has been used to collect imaging data from an airborne platform with encouraging results [5]. The Fraunhofer line depth method captures the total luminescence in each band, which when emanating from the earths surface is primarily fluorescence from plant life. This technology has been suggested as a passive method for detecting vegetation stress from orbit [6].
Live green vegetation when exposed to ambient sunlight radiation, dissipates a portion of the absorbed light energy as fluorescence. Primary fluorescence emission maxima for chlorophyll occur in bands at 685 and 740 nm. Under normal growing conditions the majority of light absorbed by the chlorophylls and carotenoids is utilized in photosynthesis, with less than 3% of the absorbed light energy being dissipated as heat or as fluorescence emissions near 685 nm (R) and 740 nm (FR). The magnitude of these fluorescence emissions on exposure of a plant to light are governed by chlorophyll concentration and photosynthetic activity. The N cycle is the driving force for carbon sequestration and the use of other nutrients. Experimentally N levels can be induced in the field. A major portion of R fluorescence emission has been attributed to chlorophylls associated with Photosystem II. A significant portion of the FR fluorescence emission has been attributed to antenna chlorophylls of Photosystem I [7].
The R/FR fluorescence ratio (685/740 nm) can relate to changes in the distribution of excitation energy between the two photosystems and has been studied as an indicator of chlorophyll content and stress condition in plants. Several researchers have demonstrated that linear and curvilinear relationships exist between certain ratios of fluorescence and pigment concentrations, and rates of photosynthesis [8,9].
The electron transfer status between Photosystem I and II can be an indicator of the health of vegetation. The plant adjusts physiologically and levels of chlorophyll change in response to the environment. During the growth cycle the R/FR fluorescence ratio decreases as a leaf chlorophyll concentration increases for most higher vegetation. The R/FR ratio increases with loss of chlorophyll under conditions of stress, such as nutrient or water shortage, excess light, or heat levels. The relationship between leaf chlorophyll concentration, absorption and the R/FR ratio of
US Government work not protected by US Copyright 1538
0-7803-7536-X/02/$17.00 (C) 2002 IEEE
chlorophyll fluorescence has been established as a way of detecting plant photosynthetic activity [10,11,12,13]. Our study investigated fluorescence emissions centered on 4 Fraunhofer line regions. Results are presented for corn (Zea mays L.) leaves from field plots grow under different N application rates (150, 100, 50, 20% of the 140kg/ha optimal N). Information collected in Fraunhofer regions compared favorably previous data [14] taken of emissions from laser induced fluorescence.
II. METHODS A. Plant Material
Leaf level samples were collected at a field site located at the USDA Beltsville Agricultural Research Center, Beltsville, MD. The site is an intensive test site for a multi-disciplinary project, Optimizing Production Inputs for Economic and Environmental Enhancement (OPE3) and is a NASA, MODIS validation site. One focus of OPE3 is to study N management and movement from field to the near ecological zone. Four hydrologically separated 4 ha watersheds are on the site, and flow into a wooded riparian wetland, and a first-order stream.
Corn N treatment plots large enough to capture the spatial variability of crop and soil parameters have been established within the OPE3 field site.
B. Fluorescence Measurements
A Hitachi F-4500 fluorescence spectrophotometer was used to collect fluorescence emissions at spectral excitation wavelengths from freshly excised corn leaves. The fluorescence spectrophotometer utilizes two 0.22 m single monochrometers, one to direct wavelengths of excitation and the other to discern intensities of emission. The excitation monochrometer stepped through specific wavelengths of light directed from a 150 watt xenon lamp onto the leaf, while the emission monochrometer collected ≈ counts/sec. (CPS) intensity at a specific Fraunhofer line (532, 607, 677 or 745 nm). A rhodamine B quantum counter was used to correct for instrument response and spectral accuracy. Excitation spectra of a specific emission wavelength at 1 mm slit width were acquired yielding a 1.7 nm resolution center on the line. Leaf samples were held in place by an anodized aluminum holder.
III. RESULTS AND DISCUSSION
The cumulative solar spectrum energy from 350 nm up to a specific Fraunhofer line will be the predominate excitation contributors to that fluorescence emission in vegetation. The data taken simulates the primary regions of the solar spectrum where differences in plant stress can be distinguished when captured by passive instruments capable of collecting fluorescence intensity at specific Fraunhofer lines.
The contributions to total 677 nm emissions enabled separation of N treatments with blue excitations, 420-490 nm only (Fig. 1). For contributions to total 745 nm emissions N treatments were separated in both the blue region, 420-490 nm, and in the red region, approaching 650 nm.
677 Emission at % of Optimal Nitrogen Fertilization
0 200 400 600 800 1000
355nm (X100) 430nm (Cha) 470nm (Chb) 550nm (< Ch) Significant Excitation Wavelength Regions
Intensity
150% 100% 50% 20%
Fig. 1. 677 nm emissions for N fertilization levels.
Emissions at 677 nm (Fig. 2) and 745 nm (Fig. 3) excited throughout the photosynthetically active spectrum show that the greatest portion of the total difference in fluorescence at these Fraunhofer line centers will come from excitation energy in the blue region of the spectrum where the chlorophylls are optimally excited. A large contribution in emission intensity centered at both 677 and 745 nm comes from wavelengths where chlorophyll a is optimally absorbed and excited in the 430 nm region. The region causing highest emissions extends through 470 nm, where chlorophyll b is optimally absorbed and excited.
0 300 600 900 1200
350 450 550 650
Excitation Wavelengths
CPS Intensity
150% 100% 50% 20% (N)
Fig. 2. Mean 677 nm emissions (n=7).
0 300 600 900 1200
350 400 450 500 550 600 650 Excitation Wavelengths
CPS Intensity
150% 100% 50% 20% (N)
Fig. 3. Mean 745 nm emissions (n=7).
US Government work not protected by US Copyright 1539
Ratios of two bands (Fig. 4) have been more successful than single bands for remote sensing purposes from space borne platforms. Corp [10] noted that a 685/740 nm ratio appeared to follow the shape of the reported photosynthetic action spectra. A similar 677/740 nm mean ratio at Fraunhofer lines with the current studies data provided separation but incomplete ranking of optimal N level according expected photosynthetic performance level (data not shown). The ratio of 677/532 nm gave the most logical ranking for the N fertilizer treatment regimes.
0 5 10 15 20
350 400 450 500
Excitation Wavelengths
Ratio Value
150% 100% 50% 20% (N)
Fig. 4, Mean emission 677/532 nm (n=7).
Field plots with different N levels produced plant canopies with different physiological attributes (Table 1). Optimal vegetative performance occurred for the recommended N (100%) fertilizer level in this Maryland area of production.
Table 1. ANOVA Physiological Attributes N Level Ch a
µg/cm2 Ch b
µg/cm2 % leaf N Yield kg/ha
150% 43.5 a 10.6 a 3.2 b 10175 a
100% 44.0 a 11.2 a 3.4 a 10784 a
50% 32.2 a 10.5 a 3.0 c 9727 a
20% 17.9 b 5.5 b 2.4 d 6211 b
IV. CONCLUSIONS
Corn grown under 4 different levels of N fertilization produced plant physiological conditions that effected: 1) Chlorophyll a, b, total chlorophyll, and total caroteniod concentration; 2) %N accumulation; 3) Carbon/N ratio; 4) rate of photosynthesis; 5) biomass accumulation; and 6) grain yield. The fluorescence responses indicated that most of the differences in intensity that can fill the Fraunhofer lines at 677 and 745 nm come from excitation of the chlorophylls in the blue (410-510 nm) region of the spectrum. In the blue, chlorophyll a is maximally excited at 430 nm while chlorophyll b is maximally excited at 470 nm. Significant differences among the upper levels of N fertilization (i.e. 150, 100, & 50% of optimal) showed mean separation for emissions at 677 nm when excited near 470 nm. This is the region where chlorophyll b absorbs the most and has been reported to be associated with the electron transfer activity in
photosystem I. A ratio of 677/532 nm regions maintained these N treatment separations. Emission intensities at 745 nm did not separate the highest 3 N levels with blue excitation.
Unlike the 677 nm line region, emissions at 745 nm did cause significant separation as they approached the red absorption area for the chlorophylls (630-650 nm). Emissions at 532 and 607 nm were low and remained relatively flat until excitation energy approached the longer wavelengths, making them good candidates for ratios with the chlorophyll Fraunhofer line bands to normalize extraneous noise effects from the emissions. Corn was used as a model in this experiment, but the levels of differences in plant physiology and fluorescence response behavior in corn is expected to be in the range of what is expected for other plant species.
REFERENCES
[1] E. Chappelle, F. Wood, J. McMurtrey, and W. Newcomb, "Laser induced fluorescence (LIF) of green plants.1,2,&3,” Applied Optics, 23:134-138, 139-142, 1984, 24:74-80, 1985.
[2] J. Plascyk, “The MKII Fraunhofer line discriminator (FLD-II) for airborne and orbital remote sensing of solar-stimulated luminescence:, Optical Engineering 14:339-346, 1975.
[3] G. Carter, J. Jones, R. Mitchell, C. Brewer, “Detection of solar-excited chlorophyll a fluorescence and leaf photosynthethic capacity using a Fraunhofer line radiometer”, Remote Sens. Environ. 55:89-92, 1996.
[4] P. Kebabian, A. Theisen, S. Kallelis, and A. Freedman, “A passive two- band sensor of sunlight-excited plant fluorescence”, Rev. Sci. Instrum.
70(11):4386-4393, 1999.
[5] W. Hemphill and M. Settle, “Workshop on applications of
luminescence techniques to earth resource studies”, LPI Tech. Report 81-03, Lunar and Planetary institute, Houston, TX, 1981.
[6] A. Theisen, “A passive method for detecting vegetation stress from orbit: Chlorophyll flurescence spectra from Fraunhofer lines, ”Proc.
Remote Sensing 2000: From Spectroscopy to Remotely Sensed Spectra.
pp. 100-108. 2000.
[7] S. Bose, “Chlorophyll fluorescence in green plants and energy transfer pathways in photosynthesis”, Photochem. Photobiol., 36:725-731, 1982.
[8] E. Chappelle, J. McMurtrey, and M. Kim,"Identification of the pigment responsible for the blue fluorescence band in the laser induced fluorescence (LIF) spectra and potential use in remote estimation of rates of photosynthesis”, Remote Sens. Environ., 36:213-218, 1991.
[9] L. Corp, J. McMurtrey, E. Chappelle, M. Kim, C. Daughtrey,
“Optimizing Fluorescence Excitation Wavelengths for the Detection of Stress in Vegetation,” IEEE, IGARSS Digest, 3:1812-1815,1996.
[10] H. Lichtenthaler, and U. Rinderle, “The role of chlorophyll fluorescence in the detection of stress conditions in plants”, Critical Reviews in Analytical Chemistry, Vol. 9, Supplement 1, CRC Press, Inc. S29-S85, 1988.
[11] R. Hak ,H. Lichtenthaler, and U. Rinderle, “Decrease of the fluorescence ratio F685/F730 during greening and development of leaves”, Rad. Environ. Biophysics 29:329-336, 1990.
[12] G. Krause, and E. Weis, “Chlorophyll fluorescence & photosynthesis:”, Annu. Rev. Plant Physiol. Plant Mol. Biol. 42:313-349, 1991.
[13] A. Gitelson, C. Buschmann, and H. Lichtenthaler, “Leaf chlorophyll fluorescence corrected for re-absorption by means of absorption and reflectance measurements”, J. Plant Physiol. 152:283-296, 1998.
[14] J. McMurtrey, E. Chappelle, M. Kim, J. Mesinger and L. Corp,
“Development of algorithms for detecting N fertilization levels in field corn with laser induced fluorescence,” Remote Sens. of Environ., 47:
36-44, 1994.
US Government work not protected by US Copyright 1540
