干旱半干旱草原区露天煤矿土壤光谱特征模型研究

The rehabilitated soil of surface mining site is originally from surrounding nature area with homogeneity. The rehabilitated topsoil under reconstructed landscape is of the spatial heterogeneity, which is the key indicator for detail digital soil mapping in the mining area. Visible and near infrared reflectance (VIS–NIR) spectroscopy is a physical non-destructive, rapid, reproducible method that provides inexpensive prediction of soil physical, chemical and biological properties according to their reflectance in the wavelength range from 400 to 2500 nm. In the large sruface coal mine site, the soil spectrum characteristic and model under different landform segment can be used to indicate rehabilitated soil change over years. This study is aimed to extract and analysis the rehabilitated soil feature from various rehabilitation ages response from different landscape using field VIS-NIR spectroscopy in the large surface coal mine area：1）developing soil organic matter content prediction models using field-collected VIS–NIR spectra; 2) analyzing the rehabilitation temporal and spatial process of topsoil from mining area based on the quantitatively relationship between soil spectral feature and topography. This study based on the field spectroscopy would effectively improve the accuracy of soil monitoring using advanced remotely sensed method in the large surface coal mine area of semi-arid grassland zone.

露天煤矿复垦区覆土多源于原生土壤，理化性质比较均一，经过多年复垦及不同地形条件下土壤会产生空间差异，是矿区复垦土壤精细监测调查的重点。将复垦区进行土壤-景观分类，并以此为单元，利用成像光谱技术快速提取土壤理化参数，从时空尺度上分析构建露天煤矿区土壤光谱特征及模型，可以揭示复垦土壤演变规律，直接指导矿区的生态修复。本项目选择干旱半干旱草原区伊敏露天煤矿作为研究区，利用光谱仪测定复垦土壤的全波段光谱，获得不同地貌单元不同复垦年限的土壤特征光谱：1）对土壤实测光谱特征定量分析，选择土壤有机质作为评价复垦土壤有效指标，建立矿区尺度土壤有机质含量光谱快速估测模型；2）通过不同复垦年限土壤光谱特征与地形因素之间定量关系，建立矿区尺度土壤-景观模型，深入分析土壤光谱在露天煤矿区地形重塑和生态重建中的时空演替规律。该研究利用光谱技术将有效提高模型在干旱半干旱草原露天煤矿区土壤遥感反演的精度和监测精度。

露天煤矿复垦区覆土多源于原生土壤，理化性质比较均一，经过多年复垦及不同地形条件下土壤会产生空间差异，是矿区复垦土壤精细监测调查的重点。相比较传统理化测试方法，光谱技术可以实现土壤有机质等特性的快速估测，进一步揭示复垦土壤演变规律，直接指导矿区的生态修复。本项目选择了平朔露天煤矿区、伊敏露天煤矿区为研究区，通过采集复垦土壤样品，测试有机质等理化参数，获取土壤可见光-近红外及热红外光谱曲线，分析排土场不同地貌以及不同复垦年限土壤有机质光谱特征，得出以下结论：1）针对露天煤矿区排土场不同地形条件下排土场复垦土壤，光谱665nm处可以作为土壤有机质的分类特征；提出基于R2294nm/R2286nm的比值指数与下坡和中坡土壤有机质特征相关性最高；2）针对露天煤矿区不同复垦年限土壤，与原地貌及复垦大于10年土壤相比较，复垦小于5年的土壤有机质含量最低，光谱反射率最高。随着有机质含量的增加，在350-500nm处的土壤吸收明显。土壤光谱可见光波段（360-760nm）经过LOG变换后，PLSR建模精度最高（R2 =0.90 , RMSE =9.55 ），全波段（350-2500nm）经过一阶导数变换后，PLSR-SVM建模精度最高(R2 =0.96 , RMSE =2.53)。3）对于草原露天煤矿区土壤，利用可见光-近红外比值指数RI（1930nm/663nm）和热红外差值指数DI（8.65μm-8.16μm）进行联合建模预测有机质精度较高。研究成果为矿区土地复垦监测提供了新的技术方法，推动了高光谱技术在露天煤矿区复垦监测中的应用。