POTENTIAL FOR DISCRIMINATING CROP RESIDUES FROM SOIL BY REFLECTANCE AND FLUORESCENCE

Crop residues left in the field after harvest can be important in cont rolling soil erosion. Current methods for quantifying percent crop res idue cover are tedious and somewhat subjective. There is a need for ne w methods to quantify residue cover that are rapid, accurate, and obje ctive. We evaluated reflectance and fluorescence techniques for discri minating crop residues from a wide range of soils. Reflectance and flu orescence spectra of 37 agricultural soils (wet and dry) and of recent ly harvested and weathered corn (Zea mays L.), soybean [Glycine mar (L .) Mere], sorghum [Sorghum bicolor (L.) Moench], and wheat (Triticum a estivum L.) residues were measured in the lab. Reflectance factors in the visible or near-infrared wavelengths did not uniquely distinguish all soils from all crop residues. Crop residues may be brighter or dar ker than a given soil, depending on soil moisture and residue age. Whe n illuminated with ultraviolet radiation, however, the crop residues f luoresced more than most of the soils. Fluorescence of crop residues w as a broad-band phenomenon centered between 420 to 520 nm and induced by a relatively broad range of excitation wavelengths centered between 350 to 400 nm. More than 90% of the crop residues <2 yr old could be discriminated from 33 of 37 dry soils and 36 of 37 wet soils using flu orescence. The threshold for discrimination can be optimized for class ification accuracy for each soil. Fluorescence techniques are less amb iguous than reflectance methods and are better suited for discriminati ng crop residues on soils. Furthermore, if properly implemented, fluor escence techniques can be used to quantify crop residue cover in the f ield.