Inversely Calculating the Roughness of Bare Soil Surface in Cold-arid Irrigation Regions Using the SAR Method
WANG Xue
LIU Quanming
MA Teng
Abstract：Quick calculation of spatial distribution of soil surface roughness is important both practically and sci-entifically. In this paper we investigated the feasibility of using the radar image of RADARSAT-2 to inversely cal-culate the surface roughness of Jiefangzha Irrigation area in Hetao Irrigation District of Inner Mongolia. The sur-face roughness in the radar image was calculated by the sectional plate method. We used both back propagation (BP) artificial neural network and the Levenberg-Marquardt back propagation (LMBP) artificial neural network to calculate and verify the inverse model for quantifying the surface roughness. The results showed that the LMBP model was superior to the BP model, witha R-squared coefficient of 0.8883 and 0.6892 respectively. Cal-culating soil surface roughness inversely using artificial intelligent model and the radar backscatter coefficient is quick, providing important basic parameters for using microwave remote sensing to monitorsoil moisture and soil salinization.
Keywords：soil surface roughnessLMBP neural networkSARmodeling
Publication Date：2017-01-01
Online Publishing Date：2026-09-12(First online date of this platform, not the publication date of the document)
Pages：7( 74-80 )