|Loechel, Sara - UNIVERSITY OF MARYLAND|
|Watson, Lee - 3DI INC., EASTON, MD|
Submitted to: Intnl Conference On Geospatial Information In Agriculture And Forestry
Publication Type: Proceedings
Publication Acceptance Date: September 20, 1998
Publication Date: N/A
Interpretive Summary: The use of remote sensing to identify and assess causes of crop spatial variability requires an understanding of the biology of the agricultural system and the physics of the remote sensing process. Crop canopy architecture, foliage density, leaf optical properties, soil reflectance, and atmospheric conditions influence canopy reflectance. Tonal variations in airborne imagery depict differences in canopy reflectances due to changes in these parameters. Tonal variability in imagery of a six hectare corn field in Beltsville, Maryland was shown to be of the same magnitude as the variability of leaf optical properties and foliage density as quantified with the coefficient of variation. This relationship suggests that remote sensing image variability expressed as the coefficient of variation may be a useful tool for detecting stresses that induce changes in LAI and leaf optics. However, questions concerning confounding factors such as percent vegetation plant cover remain. The findings are of interest to those developing algorithms for vegetation condition and vigor assessments using remotely sensed data.
Technical Abstract: The use of remote sensing to identify and assess the causes of crop spatial variability requires an understanding of both the biology of the agricultural system and the physics of the remote sensing process. Crop canopy architecture, foliage density, leaf optical properties, soil reflectance, and atmospheric conditions influence canopy reflectance. Tonal variations in airborne imagery depict differences in canopy reflectances. It is essential to demonstrate that tonal variations are correlated with changes in certain biophysical factors that influence reflectance. An investigation of the spatial variability of foliage density expressed as leaf area index (LAI), leaf optical properties, mean leaf tip angle (MTA), and soil reflectance was conducted on a six hectare field located on the Beltsville Agricultural Research Center, Beltsville, Maryland. These variables were measured on 64 locations within ten days of the acquisition of airborne scanner data. Imagery of 544, 640, and 720 nm were compared with grey-tone data postings of the LAI and MTA for a mid-season date. The coefficient of variations (CV) from highest to lowest were MTA, LAI, airborne image digital numbers (DNs) at 640 nm, leaf reflectance at 640 nm, leaf reflectance at 544 nm, canopy image DNs at 544 nm, leaf reflectance at 720 nm, and canopy image DNs at 720 nm. All of the CVs ranged between 0.08 and 0.26 for all of the measurements except MTA which was 0.81. Questions concerning the separability of leaf optical properties from foliage density and percent vegetation cover are raised. The analysis suggests that remote sensing image CV may be useful for monitoring agricultural fields for variations due to stresses.