INFLUENCE OF SOLAR ILLUMINATION ANGLE AND CULTURAL PRACTICES ON THE REFLECTANCE PROPERTIES OF SOYBEAN CANOPIES

In recent years considerable progress has been made toward the use of remote sensing technology to inventory crop acreages, assess crop stresses, and predict yields. Since soybeans (Glycine max L. Merr.) are a major crop in the United States, with over 28,800,000 hectares planted in 1979, it is important to fully understand those factors that are sources of variation in spectral measurements collected over soybeans. Spectral and agronomic measurements were acquired for soybean plots at the Purdue Agronomy farm several times during the 1978 and 1979 growing seasons. A Landsat band radiometer which acquired data in four spectral bands (0.5-0.6, 0.6-0.7, 0.7-0.8, and 0.8-1.1 (mu)m) was used both years. In 1978, 81 plots were observed periodically throughout the season to study the effect of row width, population, and cultivar on spectral response (bidirectional reflectance factor) of soybeans. A similar study was conducted in 1979 to examine the effects of row width, cultivar, planting date, and soil type on the spectral response on soybeans. As the 1978 soybean cultural practices data were analyzed, it was apparent that the time of the day or sun angle was related to the bidirectional reflectance factor (BRF). Therefore, an experiment to study specific sun angle and row azimuth interactive effects was conducted in 1979 by observing nine row directions, one complete canopy cover, and one bare soil plot in three diurnal studies. Analyses included the use of a near infrared/red reflectance ratio and greenness transformation. Results have shown a strong correlation of the bidirectional reflectance factor (BRF) of all the bands and transformations with percent soil cover. The changes in BRF associated with row width, planting date, or cultivar differences were often associated with differences in soil cover, leaf area index, and biomass. Differences in the BRF value due to cultivar were often noted late in the season when cultivars senesced at different rates. Differences in soil background due to soil moisture or soil type had an effect on the reflectance factor measurements and thus the correlations to changes in the vegetative canopy. Factors that affect the soil background such as soil color, soil moisture, or shadow due to sun angle-row azimuth interactions significantly affect the spectral response of an incomplete canopy cover, especially in the visible region. Diurnal variations in the BRF due to the shadow component in the canopy were greater than 100 percent in the visible wavelength region, while only very minor changes were noted in the near infrared. A function that utilizes both the solar azimuth and zenith angle explained most of the variation in reflectance observed for a given plot during the day