北京典型耕作土壤养分的近红外光谱分析

为研究土壤养分含量分布信息,以从北京郊区一块试验田采集的72个土壤样品为试验材料,应用傅里叶变换近红外光谱技术分析了土样的全氮、全钾、有机质养分含量和pH值。采用偏最小二乘法(PLS)对光谱数据与土壤养分实测值进行回归分析,建立预测模型,以模型决定系数(R2)、校正标准差(RMSECV)、预测标准差(RMSEP)和相对分析误差(RPD)作为模型精度的评价指标。结果表明,利用该模型与光谱数据对土壤全氮、全钾、有机质养分含量和pH值进行预测,结果与实测数据具有较好的一致性,最高决定系数R2达到0.9544。偏最小二乘回归方法建立的养分预测模型能准确地对北京地区褐土土质全氮、有机质、全钾和pH值4种养分进行预测。

Analysis of soil nutrient content based on near infrared reflectance spectroscopy in Beijing region

To study the distribution of soil nutrients, Fourier transform infrared spectroscopy techniques were used to predict total nitrogen, organic matter, total potassium and pH values of soil. With 72 soil samples collected from the experimental field in the suburbs of Beijing, the models were constructed using partial least-squares (PLS) regression based on the spectral data and measured soil nutrient. The model accuracy was evaluated using determination coefficient (R2), adjusted standard deviation (RMSECV), standard deviation of prediction (RMSEP), and residual prediction deviation (RPD). The results showed that, good consistency can be found between the prediction models and the spectral data of total nitrogen, total potassium, organic matter and pH value, the results and the measured data has, the highest coefficient of determination is R2=0.9544. Nutrient prediction model established by PLS could predict total nitrogen, organic matter, and total potassium and pH values accurately.