James E. McMurtrey

Ratio analysis of reflectance spectra (RARS) - An algorithm for the remote estimation of the concentrations of chlorophyll a, chlorophyll b, and carotenoids in soybean leaves

An algorithm utilizing reflectance spectra bands in the photosynthetically active radiation (PAR) region of the solar spectrum was developed for the remote estimation of the concentrations of chlorophyll a, chlorophyll b, and carotenoids in soybeans. The defining of specific bands in the reflectance spectrum that corresponded to absorption bands of the individual pigments was basic to the development of the algorithm. The detection of these bands was rendered difficult by the lack of detail in reflectance spectra. It was therefore necessary to manipulate the reflectance spectra so that absorption bands due to specific pigments could be detected and their spectral maxima defined. It was found that by dividing soybean reflectance spectra by an arbitrarily selected reference soybean reflectance spectrum, ratio spectra were obtained in which the absorption bands could be distinctly seen and their wavelength defined. These ratio spectra allowed the defining of those bands corresponding to the absorption bands of chlorophyll a chlorophyll b, and carotenoids. The strong linear relationships of certain combinations of the bands in the ratio spectra to the concentrations of the photosynthetic pigments made it possible to develop a ratio analysis of reflectance spectra algorithm (RARS) by which the concentrations of these pigments could be calculated from the reflectance spectra. The measurements necessary for the development of RARS were made using soybeans which were grown at different nitrogen levels in order to obtain a range of reflectance spectra. A test of the RARS algorithm using other soybean plants showed very good agreement between measured pigment values and those calculated using RARS.

Remote Sensing of Total Dry-Matter Accumulation in Winter Wheat

Red and photographic-infrared spectral data collected on 21 dates over the growing season with a hand-held radiometer were quantitatively correlated with total dry-matter accumulation in winter wheat. The spectral data were found to be highly related to vigor and condition of the plant canopy. Two periods of drought stress and subsequent recovery from it were readily apparent in the spectral data. Simple ratios of the spectral radiance data compensated for variations in solar intensities and, when integrated over the growing season, explained 79% of the variation in total above-ground accumulation of dry matter. A satellite system is proposed to provide large-area assessment of total dry accumulation or net primary production from terrestrial vegetation.

Laser-induced fluorescence of green plants. 1: A technique for the remote detection of plant stress and species differentiation

The laser-induced fluorescence (LIF) of green plants was evaluated as a means of remotely detecting plant stress and determining plant type. Corn and soybeans were used as representatives of monocots and dicots, respectively, in these studies. The fluorescence spectra of several plant pigments was excited with a nitrogen laser emitting at 337 nm. Intact leaves from corn and soybeans also fluoresced using the nitrogen laser. The two plant species exhibited fluorescence spectra which had three maxima in common at 440, 690, and 740 nm. However, the relative intensities of these maxima were distinctly different for the two species. Soybeans had an additional slight maxima at 525 nm. Potassium deficiency in corn caused an increase in fluorescence at 690 and 740 nm. Simulated water stress in soybeans resulted in increased fluorescence at 440, 525, 690, and 740 nm. The inhibition of photosynthesis in soybeans by 3-(3-4-dichlorophenyl)-1-1-dimethyl urea (DCMU) gave increased fluorescence primarily at 690 and 740 nm. Chlorosis as occurring in senescent soybean leaves caused a decrease in fluorescence at 690 and 740 nm. These studies indicate that LIF measurements of plants offer the potential for remotely detecting certain types of stress condition and also for differentiating plant species.

Laser-induced fluorescence of green plants. 2: LIF caused by nutrient deficiencies in corn

The effects of nutrient deficiencies on the laser-induced fluorescence spectra of intact corn plants were studied to determine the utility of the LIF technique as a field and remote sensing tool for detection of nutrient deficiencies. A pulsed nitrogen laser emitting at 337 nm was used as the excitation source. The fluorescence maxima in corn were at 440, 690, and 740 nm. A decrease in fluorescence at 690 and 740 nm was observed for those plants deprived of phosphorus, nitrogen, and iron. The absence of nitrogen and iron also caused a small decrease in fluorescence at 440 nm. Plants deprived of calcium, sulfur, and magnesium showed no significant change in fluorescence at any of the bands. The lack of potassium increased the fluorescence at 690 and 740 nm more than threefold along with a small decrease at 440 nm.

Laser-induced fluorescence of green plants. 3: LIF spectral signatures of five major plant types

A technique amenable to remote sensing use which utilizes laser-induced fluorescence (LIF) properties of plants has been successfully used in the laboratory to identify five major plant types. These included herbaceous dicots, herbaceous monocots, conifers, hardwoods, and algae. Each of these plant types exhibited a characteristic LIF spectra when excited by a pulsed N2 laser emitting at 337 nm. Although monocots and dicots possess common fluorescence maxima at 440, 685, and 740 nm, they could be differentiated from one another by using the ratio of the square of the fluorescence intensity at 440 nm to the nonsquared intensity at 685 nm, i.e., (440)2/685. In all cases, monocots yielded a significantly higher ratio. Conifers have fluorescence maxima at 440, 525, and 740 nm but none at 685 nm. Hardwoods exhibited fluorescence at 440, 525, 685, and 740 nm. Algae had very low fluorescence at 440 nm, no fluorescence at 525 nm, and fluorescence maxima at 685 and 740 nm. For algae, the ratio of the fluorescence intensity at 685 nm to that at 740 nm was much greater than that for monocots, dicots, and hardwoods. The potential use of the LIF technique for individual species identification is suggested.

Identification of the pigment responsible for the blue fluorescence band in the laser induced fluorescence (LIF) spectra of green plants, and the potential use of this band in remotely estimating rates of photosynthesis

rln 1he laser-induced fluorescence (LIF) of vegetation is being investigated in this laboratory for use as a technique for the remote detection of the effects of environmental stress upon vegetation, as well as for plant identification. The fluorescence band with a maximum at 440 nm, in conjunction with the chlorophyll bands with maxima at 685 and 740 nm, has been found to be a critical band in the development of algorithms for detecting stress, and identifying plant types. The identification of the plant constituent responsible for this band is vital to understanding the mechanism underlying its fluorescence changes in response to environmental and physiological changes. The identification was achieved as follows: The laser induced fluorescence (LIF) spectra of pure plant pigments were determined. Fluorescence bands with maxima at 420 nm, 440 nm, 490 nm, and 525 nm were observed for vitamin K 1, reduced nicotinamide adenine dinucleotide (NADPH), beta-carotene, and riboflavin, respectively. The LIF spectra of water extracts and acetone extracts of clover leaves were also measured. It was found that the blue fluorescence band was associated with the water extract. NADPH which is a water-soluble compound, and the water extract of clover had no fluorescence after oxida

Fluorescence sensing systems: In vivo detection of biophysical variations in field corn due to nitrogen supply

Leaf and canopy fluorescence properties of field corn (Zea mays L.) grown under varying levels of nitrogen (N) fertilization were characterized to provide an improved N sensing capability which may assist growers in site-specific N management decisions. In vivo fluorescence emissions can occur in the wavelength region from 300 to 800 nm and are dependent on the wavelength of illumination. These light emissions have been grouped into five primary bands with maxima most frequently received from corn at 320 nm (UV), 450 nm (blue), 530 nm (green), 685 nm (red), and 740 nm (far-red). Two active fluorescence sensing systems have been custom developed; a leaf level Fluorescence Imaging System (FIS), and a canopy level Laser Induced Fluorescence Imaging System (LIFIS). FIS sequentially acquires high-resolution images of fluorescence emission bands under darkened laboratory conditions, while LIFIS simultaneously acquires four band images of plant canopies z 1m 2 under ambient sunlit conditions. Fluorescence emissions induced by these systems along with additional biophysical measures of crop condition; namely, chlorophyll content, N/C ratio, leaf area index (LAI), and grain yield, exhibited similar curvilinear responses to levels of supplied N. A number of significant linear correlations were found among band emissions and several band ratios versus measures of crop condition. Significant differences were obtained for several fluorescence band ratios with respect to the level of supplied N. Leaf adaxial versus abaxial surface emissions exhibited opposing trends with respect to the level of supplied N. Evidence supports that this confounding effect could be removed in part by the green/blue and green/red ratio images. The FIS and LIFIS active fluorescence sensor systems yielded results which support the underlying hypothesis that leaf and canopy fluorescence emissions are associated with other biophysical attributes of crop growth and this information could potentially assist in the site-specific management of variable-rate N fertilization programs.

Fluorescence sensing techniques for vegetation assessment

Active fluorescence (F) sensing systems have long been suggested as a means to identify species composition and determine physiological status of plants. Passive F systems for large-scale remote assessment of vegetation will undoubtedly rely on solar-induced F (SIF), and this information could potentially be obtained from the Fraunhofer line depth (FLD) principle. However, understanding the relationships between the information and knowledge gained from active and passive systems remains to be addressed. Here we present an approach in which actively induced F spectral data are used to simulate and project the magnitude of SIF that can be expected from near-ground observations within selected solar Fraunhofer line regions. Comparisons among vegetative species and nitrogen (N) supply treatments were made with three F approaches: the passive FLD principle applied to telluric oxygen (O2) bands from field-acquired canopy reflectance spectra, simulated SIF from actively induced laboratory emission spectra of leaves at a series of solar Fraunhofer lines ranging from 422 to 758 nm, and examination of two dual-F excitation algorithms developed from laboratory data. From these analyses we infer that SIF from whole-plant canopies can be simulated by use of laboratory data from active systems on individual leaves and that SIF has application for the large-scale assessment of vegetation.

Radiometric measurements over bare and vegetated fields at 1.4-GHz and 5-GHz frequencies

Results of radiometric measurements over bare and vegetated fields with dual-polarized microwave radiometers at 1.4-GHz and 5-GHz frequencies are presented. The measured brightness temperatures over bare fields are shown to compare favorably with those calculated from radiative transfer theory with two constant parameters characterizing surface roughness effect. The presence of vegetation cover is found to reduce the sensitivity to soil moisture variation. This sensitivity reduction is generally more pronounced the denser the vegetation cover and the higher the frequency of observation. The effect of vegetation cover is also examined with respect to the measured polarization factor at both frequencies. With the exception of dry corn fields, the measured polarization factor over vegetated fields is found appreciably reduced compared to that over bare fields. A much larger reduction in this factor is found at 5 GHz than at 1.4 GHz.

Demonstration of the accuracy of improved-resolution hyperspectral imagery

Numerous researchers have demonstrated the accuracy and utility of improved spatial resolution multispectral imagery by sharpening it with higher spatial resolution panchromatic imagery. A much more limited number of researchers have sharpened hyperspectral imagery with panchromatic imagery. In this research we have developed an algorithm that spatially sharpens specific ranges of hyperspectral bands with spectrally correlated multispectral bands of a higher spatial resolution to improve the spatial resolution of the hyperspectral imagery while maintaining or improving its spectral fidelity. Preliminary validation of the algorithm has been conducted using a 7m AVIRIS scene of the Maryland Eastern Shore containing corn, soybean, and wheat fields. This data was used to simulate 28m HSI and 7m MSI that were used in the sharpening process. Initial analysis has verified the spectral accuracy of the sharpened data. In the next phase of the study, airborne spectral data from two different sensors will be used in the sharpening process with the results used as input for USDA/ARS crop yield and stress models.