摘要
离子型稀土的开采活动会导致矿区地表极其剧烈的生态扰动,并且会造成当地的生态环境问题,而矿区地表热环境分异变化能较好地反映矿区的生态扰动特点,是一种辨识地表生态扰动的重要参数。离子型稀土矿区存在矿点分散且单个矿点面积较小的特征,因此获取实用性强且空间分辨率更高的地表温度数据对稀土矿区生态环境的监测具有重要价值。构建了一种综合图像融合算法、线性光谱混合模型的地表温度降尺度模型。以赣州市辖区内的定南县岭北离子型稀土矿区作为研究区域,以Landsat 8卫星影像作为主要数据源,首先选取同一年份两个季相的数据,综合图像融合算法和线性光谱混合模型,将地表温度空间分辨率降至15 m;然后对降尺度后的地表温度结果进行定性定量的分析并检验其精度。结果表明,分解前、后矿区地表温度的空间分布和走向整体一致,降尺度后的地表温度能够更细致地反映矿区地表特征和空间差异性,研究区内两个季相的降尺度结果整体均方根误差分别为1.459 K、1.196 K,绝对误差分别为1.128 K、0.952 K,精度较高,表明该方法对于提升离子型稀土矿的地表温度空间分辨率有较好的适用性。
The mining activities of ion-type rare earth have caused extremely ecological disturbances on the surface of the mining area and caused local ecological and environmental problems.The variation of surface thermal environment in the mining area can better reflect the ecological disturbance characteristics of the mining area,and is an important parameter to identify surface ecological disturbances.The ion-type rare earth area has the characteristics of scattered ore and small single-site area,thus obtaining the surface temperature data with strong practicability and higher spatial resolution is valuable to the monitoring of the ecological environment of the rare earth mining area.We constructed a temperature downscale model with image fusion and spectral unmixing.The Lingbei ion-type rare earth district in Dingnan County of Ganzhou City is selected as the study area.The Landsat 8 satellite image is used as main data source.Firstly,we select data of two seasons in the same year,and combine the integrated image fusion algorithm and linear spectral mixture model.The surface temperature resolution of the surface is downscaled to 15 m.Then,the land surface temperature results after downscaling are qualitatively and quantitatively analyzed and tested for accuracy.The results show that the spatial distribution of the surface temperature and the overall trend of the mining area before and after the decomposition are consistent.The surface temperature after the downscaling can reflect the surface features and spatial differences of the mining area in more details.The overall root mean square error(RMSE) of the two seasonal phases in the study area are respectively for 1.459 K and 1.196 K,the mean absolute error(MAE) are 1.128 K and 0.952 K respectively with high accuracy.Our proposed method has high applicability for improving the spatial resolution of the surface temperature of the ionic rare earth.
作者
李恒凯
吴冠华
王秀丽
LI Hengkai;WU Guanhua;WANG Xiuli(School of Architectural and Surveying&Mapping Engineering,Jiangxi University of Science and Technology,Ganzhou 341000,China;College of Economic Management,Jiangxi University of Science and Technology,Ganzhou 341000,China)
出处
《武汉大学学报（信息科学版）》
EI
CSCD
北大核心
2021年第1期133-142,共10页
Geomatics and Information Science of Wuhan University
基金
教育部人文社科研究项目规划基金(18YJAZH040)
江西省自然科学基金(20181BAB206018)
江西省社会科学规划课题(17YJ20)。
关键词
地表温度
降尺度
离子型稀土矿
线性光谱混合模型
开采扰动
land surface temperature
downscaling
ion-type rare earth
linear spectral mixture model
mining disturbance
