论文部分内容阅读
以滨海盐土为研究对象,通过添加不同浓度的盐溶液并模拟蒸发过程,获取不同含水、含盐量的土壤样品,并测定土壤光谱和土壤含水量,分别运用光谱指数法和偏最小二乘回归法(PLSR)对土壤含水量进行预测。结果表明:由2 027 nm和1 878 nm构建的土壤水分差异化光谱指数(NDMI2027,1878)是预测土壤水分的最优指数,且适用于任何等级的盐渍化土壤,其建模集和验证集的预测结果均优于PLSR方法,验证集R2达0.99,RMSE仅为21.84 g/kg,可比较准确地预测盐渍化土壤的含水量。
Taking coastal saline soil as the research object, soil samples with different water and salinity were obtained by adding different concentrations of salt solution and simulating the evaporation process. The soil spectra and soil water content were determined. The spectral index method and partial least-squares regression Law (PLSR) to predict soil moisture content. The results showed that the soil moisture differentiated spectral index (NDMI2027, 1878) constructed by 2 027 nm and 1 878 nm was the best index for predicting soil moisture and was suitable for any grade of salinized soils. Its modeling set and validation The prediction results of the set are superior to the PLSR method. The validation set R2 reaches 0.99 and the RMSE is only 21.84 g / kg, which can predict the water content of salinized soil more accurately.