|Chang, Chen - IOWA STATE UNIVERSITY|
Submitted to: Agronomy Abstracts
Publication Type: Abstract Only
Publication Acceptance Date: November 4, 1999
Publication Date: N/A
Technical Abstract: Near-infrared reflectance spectroscopy (NIRS) is a fast and convenient technique for analysis of soil properties. The objective of this study is to understand the influence of moisture content on near-infrared spectra of soils and on the accuracy of NIRS predictions. Both moist and air-dried soils from an agricultural field were analyzed by NIRS and the spectra were related to measured soil properties using principle component regression. Correlation coefficients (r) between measured and predicted values for moisture content, total C and N, CEC, pH, and carbonate content are higher than 0.95 and 0.92 for the air-dried and moist samples, respectively. The r values for %sand and %clay are higher than 0.89 for the air-dried soils and 0.86 for the moist soils. Although NIRS analysis is more accurate for air-dried soils, the acceptable accuracy obtained for moist soils indicates the potential application of NIRS in the field.