Soil water content (θ) is an important factor for the crop growth and crop production .The objectives of this study were to (i) test various regression models for estimating θ based on spectral feature parameters,and (ii) compare the performance of the proposed models by using artificial neural networks (ANN) and spectral feature parameters .The θ data of sand and loam and concurrent spectral parameters were acquired at the laboratory experiment in 2014 .The results showed that: (1) the maximum reflectance with 900~970 nm and the sum reflectance within 900~970 nm estimate θ had the significant,when sand bulk density was 140 g·cm-3; the maximum reflectance with blue edge and the s um reflectance within 900~970 nm had the best correlation (R2>070) when s and bulk density was 150 g·cm-3; while soil bulk density was 160 g·cm-3,the sum reflectance within 780~970 nm and normalized absorption depth in 560~760 nm reached a significant (R2>090); when soil bulk density was 170 g·cm-3,the maximum reflectance with 900~970 nm and the sum reflectance within 900~970 nm had the best correlation estimate θ (R2>088) .2) When the soil type was loam,the maximum reflectance with 900~970 nm and the sum reflectance within 900~970 nm had a best correlation estimate θ .The spectral feature parameters the sum reflectance within blue edge (R2=026 and RMSE=009 m3·m-3) and 780~970 nm absorption depth (R2=032 and RMSE=0 .10 m3·m-3) were best correlated with θ in the sand .The θ model based on maximum reflectance with 900~970 nm (R2=092 and RMSE=005 m3·m-3) and the sum reflectance within 900~970 nm had a high correlation (R2=092 and RMSE=004 m3·m-3) in the loam .The BP-ANN model presented a better estimation accuracy of θ (R2=087 and RMSE=005 m3·m-3) in two soils .Thus,the ANN model has great potential for estimating θ .Thus,the BP-ANN model has great potential for θ estimation .%土壤含水量(θ)是影响作物生长和作物产量的主要因素之一.旨在评估基于光谱特征参数的各种回归模型估算土壤含水量的精度,并比较人工神经网络(BP-ANN)和光谱特征参数模型的性能.2014年在室内获取砂土和壤土的土壤含水量和光谱反射率数据.结果表明:(1)当砂土容重为140 g·cm-3时,900~970 nm最大反射率和900~970 nm反射率总和估算θ达到极显著水平(R2超过090);容重为150 g·cm-3时,用蓝边最大反射率和900~970 nm反射率总和估算θ相关性最好(超过070);容重为160 g·cm-3时,780~970 nm反射率总和与560~760 nm归一化吸收深度的R2均超过090,达到极显著水平;容重为170 g·cm-3时,900~970 nm最大反射率和900~970 nm反射率总和的R2为088,呈极显著水平.(2)当土壤类型为壤土时,用900~970 nm最大反射率和900~970 nm反射率总和估算θ相关性最好.(3)蓝边反射率总和(R2=026和RMSE=009 m3·m-3)和780~970 nm吸收深度(R2=032和RMSE=010 m3·m-3)估算砂土的含水量相关性最好.在估算壤土的含水量时,900~970 nm最大反射率(R2=092和R MSE=005 m3·m-3)与900~970 nm反射率总和估算模型的精度最高(R2=0 92和RMSE=004 m3·m-3).(4)用人工神经网络模型能够更好地估算两种土壤的含水量(R2=087和RMSE=005 m3·m-3).因此,人工神经网络模型对θ估算具有巨大的潜力.
展开▼