亚热带土壤铬元素的高光谱响应和反演模型

高光谱遥感技术已成为当前遥感领域的前沿技术，因其高分辨率的特点，可利用地物反射光谱特征定量反演地物的物理化学性质。目前土壤环境质量愈来愈受到关注，土壤重金属含量与土壤环境质量安全密切相关，以往土壤高光谱遥感技术研究多注重于土壤有机成分如土壤碳氮的光谱反演模型，对土壤重金属含量的高光谱反演研究普遍较少。土壤重金属污染已经成为影响土壤质量安全的关键因素，对土壤重金属尤其是污染元素普查是当务之急。传统土壤重金属的测试方法要求条件较高，测试周期较长，试图建立土壤高光谱与土壤铬元素(ICP-MS测定)含量之间的定量预测模型，以实现土壤铬元素的快速准确预测。采集福州市土壤样品135个，对土壤样品在350～2 500 nm的光谱反射率进行倒数、对数、微分等六种变换，筛选出对土壤总铬含量敏感的光谱波段，最后获得福州土壤铬元素高光谱反演优化模型。研究结果表明：亚热带红壤总铬的敏感光谱波段为：可见光520～530 nm和近红外1 440～1 450，2 010～2 020，2 230～2 240 nm;亚热带地区土壤总铬—高光谱反演的优化模型为: y=120.768e-7.037x(相关系数R为0.568，均方根误差为0.619 μg·g-1，检验相关系数R为0.484，均方根误差为1.426 μg·g-1)，该模型可以用于福州地区土壤全铬的光谱快速监测。

Spectral Inversion Models for Prediction of Total Chromium Content in Subtropical Soil

With the high requirements and long test cycle of traditional testing method of soil heavy metal, this paper tries to establish the quantitative prediction model between soil hyperspectral and soil chromium content(tested by ICP-MS) to realize the prediction of soil chromium element quickly and accurately. The paper studied the hyperspectral response characteristics of red soil, with 135 soil samples in Fuzhou city. After monitoring the hypersectral reflection of soil samples with ASD (analytical spectral device) and total chromium contents with ICP-MS, the paper gained the spectral reflection data between 350 and 2 500 nm and soil total chromium contents. Then the paper treated the hyperspectral reflection data with 6 mathematic changes such as reciprocal logarithmic change, differentials and continuum removal in advance. The next step was to calculate the correlation coefficient of soil chromium and the above spectral information, and select the sensitive spectral bands according to the highest correlation coefficient. Finally, six kinds of models were selected to build the soil total chromium content model, and the final optimal mathematic model between soil chromium and hyperspectral information was significantly determined. Results showed that 520～530, 1 440～1 450, 2 010～2 020, and 2 230～2 240 nm were the main sensitive bands to soil total chromium, y=120.768e-7.037x was the optimal soil total chromium predicting model(in the model, the correlation coefficient R and the RMSE of total chromium were 0.568 and 0.619 μg·g-1, and the inspection correlation coefficient R and the RMSE were 0.484 μg·g-1 and 1.426 μg·g-1 respectively). The model can be used to rapidly monitor soil total chromium with hyperspectral reflection in Fuzhou area.