Premium
Modeling the Effect of Moisture on the Reflectance of Crop Residues
Author(s) -
Wang Changkun,
Pan Xianzhang,
Liu Ya,
Li Yanli,
Zhou Rui,
Xie Xianli
Publication year - 2012
Publication title -
agronomy journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.752
H-Index - 131
eISSN - 1435-0645
pISSN - 0002-1962
DOI - 10.2134/agronj2012.0133
Subject(s) - reflectivity , residue (chemistry) , crop residue , water content , mean squared error , environmental science , moisture , remote sensing , wavelength , crop , coefficient of determination , soil science , shortwave , mathematics , agronomy , radiative transfer , agriculture , chemistry , materials science , meteorology , statistics , geology , optics , ecology , geography , biochemistry , physics , geotechnical engineering , optoelectronics , biology
Models of crop residue reflectance under changing moisture conditions are required to better quantify management of crop residue in agricultural fields by remote sensing. In this study, reflectance spectra (400–2400 nm) of four different crop residues were measured in the laboratory under various moisture conditions. Crop residue reflectance decreased with increasing water content across all wavelengths, especially in the shortwave infrared (SWIR) region. For each crop residue, both linear and exponential models were capable of quantifying the relationship between reflectance and mass water content (MWC) with high accuracy [e.g., coefficient of determination ( R 2 ) > 0.9 and root mean square error (RMSE) < 0.025 at each wavelength in the SWIR]. The best‐fitting parameters of the two models could be directly derived from the reflectance of dry and saturated crop residues and saturated water content, which simplified the construction of the models. The results of regression analyses between the predicted MWC calculated by the inversion of the two models and measured MWC indicated that MWC could be estimated with high accuracy ( R 2 > 0.9, RMSE < 0.15, slope close to 1.0, and intercept close to 0.0) at nearly every wavelength in the SWIR. To construct a general model, the reflectance was normalized, and the results suggested that this method was valid. The results of this study help to quantify the effect of water on the spectral reflectance of crop residue and should contribute to better correct the effects of water on remote sensing estimation of crop residue cover in the future.