Influences of Soil Bulk Density and Texture on Estimation of Surface Soil Moisture Using Spectral Feature Parameters and an Artificial Neural Network Algorithm

Effective monitoring of soil moisture (θ) by non-destructive means is important for crop irrigation management. Soil bulk density (ρ) a major factor that affects potential application θ estimation models using remotely-sensed data. However, few researchers have focused on and quantified the effect ρ spectral reflectance with different textures. Therefore, we influences texture θ, evaluated performance from combining feature parameters artificial neural network (ANN) algorithm to estimate θ. The conclusions are as follows: (1) sandy soil, most strongly correlated were Sg (sum in green edge) A_Depth780–970 (absorption depth at 780–970 nm). (2) had significant correlation R900–970 (maximum 900–970 nm) S900–970 loamy soil. (3) best clay loam respectively. (4) showed higher accuracy estimating achieved all four Combining ANN produced (R2 = 0.95 RMSE 0.03 m3 m−3)