论文部分内容阅读
本文对克里雅河流域进行野外调查、采集土壤样品及其光谱反射特性的测量,通过比较不同光谱预处理的方法建立偏最小二乘回归(PLSR)模型,并利用决定系数(R2)、均方根误差(RMSEP)、残留预测偏差(RPD)对模型的稳定性和预测能力进行检验。结果表明:反射率一阶微分是预测土壤样本盐分含量的最佳光谱指标。PLSR模型在建立土壤光谱与盐分含量关系时较为适用,R2、RMSE和RPD分别为0.77、0.25和1.88。利用反射光谱估算土壤中盐分含量,通过各种光谱预处理方法可以提高估算精度,可以为该区土壤盐渍化评价和生态环境调查提供依据。
In this paper, we conducted field surveys in the Keriyaya basin, collected soil samples and measured the spectral reflectance characteristics. Partial least squares regression (PLSR) models were established by comparing different spectral pretreatment methods. The determination coefficients (R2) Root mean square error (RMSEP) and residual prediction error (RPD) were used to test the stability and predictive ability of the model. The results show that the first-order differential reflectance is the best spectral index for predicting the salinity of soil samples. The PLSR model is more suitable for establishing the relationship between soil spectra and salt content, R2, RMSE and RPD are 0.77, 0.25 and 1.88, respectively. Estimation of soil salt content by reflectance spectroscopy can improve the estimation accuracy through various spectral pretreatment methods, which can provide basis for soil salinization assessment and ecological environment investigation in this area.