【作者】 雷少刚；

【导师】 卞正富； John L.Daniels；

【作者基本信息】 中国矿业大学 ， 地图制图学与地理信息工程， 2009， 博士

【摘要】 为研究地下煤炭开采对荒漠区脆弱环境的影响规律,论文选择了植被、水体(土壤水、包气带水、地下水)作为荒漠矿区的关键环境要素,以具有代表性的神东矿区为研究区域,集RS、GIS、GPR、现场调查等多种技术手段,多尺度监测研究关键环境要素的时空演变规律,并在监测技术与采动影响规律方面取得了以下成果:在宏观尺度上,利用MODIS-NDVI实现了区域植被的月际时序监测分析,以及与气温、降水等气象因素的关联分析。结果表明研究区植被具有明显的物候年周期性,植被与气象因素具有明显的相关性,其中植被对降水变化的响应最为敏感。总体来讲,矿区植被呈现出区域性变化,近年来开展的生态建设影响,使得植被有明显好转,空间变异性增强。然而小尺度的植被现场调查与Landsat-NDVI监测表明,地下开采导致了部分矿井采区植被相对非采区小幅下降。为解释采矿对植被的影响规律,论文对矿区浅层土壤含水率进行了遥感定量反演。研究表明,相对于土壤反射率、土壤湿度、地形湿度指数,土壤含水率与表观热惯量之间具有更加显著的相关性。通过改进的土壤温度预测模型,建立了适用于TM或ETM+的土壤温差转换模型。利用该模型能更准确地提取任意深度的土壤温差信息,这比单纯利用MODIS影像能提取更高分辨率的表观热惯量。改进后的表观热惯量与10cm埋深土壤含水率的R2由0.264提高到了0.789。从而首次实现了基于表观热惯量法与TM/ETM+、MODIS影像相结合的土壤含水率高分辨率反演。反演结果与现场调查一致发现,受地表沉陷影响,采区土壤含水率均略偏小于非采区。对工作面上方地下水采前、采中、采后的长期观测表明,采矿对地下水具有显著的负面影响,且采后地下水的恢复过程缓慢。但是,该工作面上方植被并没有明显衰退。其原因在于,通过探地雷达(GPR)对包气带土壤含水的垂向分布研究给出了风积沙区地下水的临界作用埋深为8m,而该工作面的初始水位埋深远远大于临界埋深。此外,本文首次结合GPR与开采沉陷预计理论对沉陷变形影响下的包气带土壤含水的横向分布规律进行了分区研究,分析了土壤压缩、拉伸变形对包气带土壤含水分布的影响规律,指出今后沉陷变形区土壤特性的实验研究应分区进行。最后,根据对矿区植被、水体的地空一体化监测结果,给出了植被与土壤水的负倒数关系模型,地下水与土壤含水、植被的指数关系模型,以及地下水深埋区植被受采动影响的判别模型,提出了适用于荒漠矿区资源环境协调开采的建议与措施。更多还原

【Abstract】 The objective of this dissertation is to find the impact law of underground mining on the fragile environment at desert area. Therefore, the worldwide known Shendong mining area was taken as a case study area. The vegetation and water resource including the surface soil moisture, unsaturated zone water, and groundwater, were selected as the key environment elements based on abundant literature reviews. The spatiotemporal evolution law of which was monitored by integrating with RS, GIS, GPR, field investigation etc. The conclusions were made as follows.On large scale, monthly Series of MODIS-NDVI-250m/1km were summarized from 2000 to 2005, which showed the vegetation varied circannian, corresponding with the circannian variation of air temperature, rainfall, air pressure etc. Therefore, there was apparent relativity between NDVI and climate factors; however, vegetation is more susceptible to rainfall. Overall, the vegetation of study area varied regionally, and improved since 2002, and no abnormal was detected. On micro-scale, it was observed by both TM and field investigation that the vegetation of mined area was a slight less than the unmined area.The surface soil moisture of the study area was retrieved by advanced RS to further interpret the vegetation variation due to the underground mining. Compared with soil reflectance, soil wetness, topography wetness index, soil albedo, the apparent thermal inertia has a distinct relationship with soil moisture. For improving the resolution of predicted soil moisture, the soil temperature difference model was developed to model the soil moisture with higher resolution by method of apparent thermal inertia (ATI) and TM image. The ATI has stronger relationship between the observed soil moisture at depth of 10cm, than that of 20cm, and 40cm. Therefore, the square of correlation coefficients (R2) was improved to 0.789 by improved soil temperature difference model at depth of 10cm. Moreover, the nonlinear relationship between soil moisture and vegetation or topography was found. Also, it is observed by both the TM image and Field investigation that surface soil moisture at mined area is slight lower than at unmined area, especially the variation of loess region is more apparent than sand region. Overall, the difference was not remarkable, so the vegetation did not get worse very much.The groundwater (aquifer thickness, relative and absolute groundwater table) at 32201 mining face was observed from the beginning to the third year after the end of mining. It indicated that the groundwater was disturbed seriously by underground mining. But the vegetation on the 32201 mining face did not get worse as great loss of groundwater, which was explained by the GPR detection of unsaturated zone. The critical groundwater table depth (8m) was found by the analysis of vertical distribution of soil moisture at unsaturated zone. Because the original groundwater table depth before mining was already 30m, which meant that the vegetation did not depend on the groundwater and would not change dramatically, no matter how the groundwater table drop. Additionally, it was the first time to integrate GPR with Mining Subsidence Predicting Theory, then the lateral variation of soil moisture at different deform region was detected. It demonstrated that the compression and stretch deformation due to mining subsidence would change the soil moisture. So, more attention is needed to study the soil characteristics at deformation area in future.Finally, an inverse model of soil moisture and NDVI, and two exponential models of groundwater and NDVI, or soil moisture were developed based on the monitoring results. Furthermore, some recommendations were provided for green-mining at desert mining area.更多还原