| Popis: |
Soil moisture information has been used for irrigation scheduling, site-specific management of diseases and pests, and improving crop yield prediction. Spectral remote sensing offers the potential to provide more information for making better-informed management decisions in real time. In contrast, the tradition methods for irrigation management such as tensiometers and oven dry for estimating moisture content are generally time consuming, numerous observations are required to characterize them. The aim of this study was to investigate the suitability of hyperspectral reflectance sensor to estimate the moisture content of different soils. For that the spectral indices of soil were tested to assessment the moisture content by wetted the soil from dry to saturation conditions. The results showed that the three water spectral indices R960/R940, R970/R940 and R970/R900 showed close and highly significant associations with moisture content of sandy soil, and coefficients of determination reach up to R2 = 0.98. The three water spectral indices R878/R862, R956/R926 and R1056/R994 showed close and highly significant associations with moisture content of sandy loam soil, and coefficients of determination reach up to R2 = 0.84. As well as the three water spectral indices R956/R924, R956/R926 and R956/R9284 showed close and highly significant associations with moisture content of clay soil, and coefficients of determination reach up to R2 = 0.86. In conclusion, the use of spectral remote sensing may open an avenue in irrigation management for fast, high-throughput assessments of water status of soil samples. |
