结合激光雷达和光学遥感研究人工纯林下水力土壤侵蚀

The Universal Soil Losing Equation (USLE) was constructed by multiplicating five factors such as Pricipitation,soil property,terrain,control practice and vegetation coverage. It has been used to evaluate regional soil erosion widely. And the Spectral Vegetation Index (SVI) obtained from optical remote sensing was always used to expressed the vegetation coverage factors.But SVI can't describe the difference of soil conservation caused by the difference of forest's constructure. Since 1980s, many plantation forest were buildup in Jiangxi Province to improve the environment of desert hills. These plantation forest are mainly composed of Pinus massoniana,Pinus elliottii or Cunninghamia Lanceolata. And severe soil erosion are still happening under forest canopy in despite of high coverage because of the simple constructure of the plantation forest. USLE can't evaluate these erosion under these canopy of plantation forest with high coverage if only SVI was used to depict the vegetation coverage factor.In recent years, the waveform from LiDAR (Laser Intensity Direction And Ranging) were used to collect the information of the forest's structure. And the Geoscience Laser Altimeter System (GLAS) carried by Ice,Cloud, and Land elevation Satellite(ICESAT) was lauched in 2003 and has collected the waveform of land surface for about 9 years. In this study,vegetation structure factor will be retrieved from ICESAT-GLAS waveform. and the vegetation coverage factor in the USLE will be relationed to vegetation structure factors retrieved from ICESAT-GLAS waveform by the soil erosion monitoring at these stations managed by Jiangxi Provincial Institute for Soil and Water Conservation. With the improved vegetation coverage factor and the other factors calculated by these methods that have been used and accepted widely, the soil erosion for Jiangxi Provin will be estimated and the soil conservation difference for these forests with different structure will be evaluated.

上世纪80年代以来，江西省在治理荒山过程中培育了大量以马尾松(Pinus massoniana)、湿地松(Pinus elliottii)和杉木等为主的人工林，这些人工林的林相结构单一，尽管覆盖度较高，但林下水土流失依然严重。通用土壤侵蚀方程(USLE)被广泛用于区域土壤侵蚀量评价中，由于遥感技术获取区域陆表参数的优势，USLE中的植被覆盖因子往往由光谱植被指数（SVI）表征，但SVI表征的植被覆盖因子难以合理评价人工林下土壤侵蚀。近年来，激光雷达(LiDAR)获取的波形信息被广泛用于森林结构提取。本研究拟通过野外径流小区和集水区产沙试验和林相结构参数的调查，建立植被覆盖因子与林相结构参数的关系，结合光学遥感和ICESAT卫星搭载的GLAS传感器所记录的波形信息获取区域的林相结构参数；在此基础上，计算不同林相结构森林的植被覆盖因子，以USLE评价鄱阳湖流域不同林相结构森林的水力侵蚀差异。

植树造林是开展生态恢复主要方式。鄱阳湖流域植树造林以以马尾松(Pinusmassoniana)、湿地松(Pinus elliottii)、杉木和果园等为主。尽管结合USLE方程和光学植被指数广泛用于区域尺度的土壤侵蚀评价，但难以合理评价人工林的水土保持效益。本课题试图结合光学遥感和大光斑雷达波形信息获取的林相结构参数，评价鄱阳湖流域植被恢复对土壤侵蚀的影响。课题组主要开展鄱阳湖流域人工林、果园分布制图和森林结构参数的激光雷达获取、土地利用变化制图、径流小区产沙试验、小流域卡口站径流泥沙观测以及鄱阳湖“五河”控制水文站径流泥沙资料的收集，从“五河”子流域、典型人工林小流域和径流小区等尺度，研究人类活动（植被恢复、水库建设和采砂等）等对流域径流、蒸散、产沙和输沙的影响，并结合光学遥感分析鄱阳湖水沙平衡。研究表明：.（1）结合大孔径激光雷达和光学遥感影像，能够较好地反演植被冠层高度和地上部分生物量；.（2）鄱阳湖流域人工林面积达到9.9×104km2，其中针叶林5.9×104km2；赣南柑橘果园面积达260万亩，果园和人工林下的水土流失都比较严重；.（3）径流小区试验的结果表明人工马尾松林下的植被灌丛对小区径流和产沙都有显著的影响；.（4）近三十年来鄱阳湖流域蒸散呈阶段性增长趋势，植被恢复和降水变化是造成蒸散空间差异的主要原因，植被恢复驱动蒸散呈增加趋势，气候变化导致蒸散年际波动；.（5）从鄱阳湖流域及其“五河”流域的径流泥沙观测资料和SWAT模拟表明，流域范围的水库建设是入湖泥沙减少的主要原因，而植被恢复对河道泥沙的减少效应不是很明显，水库蓄水改变了流域入湖径流的季节分配特征；.（6）鄱阳湖泥沙的平衡研究表明：上游水库运行显著降低了入湖泥沙量，而2000年以来鄱阳湖的采砂导致出湖泥沙显著增加，使鄱阳湖从净沉积环境逆转为净侵蚀环境，同时，鄱阳湖大规模无序采砂导致入江河道加深加宽，过水断面增大，泄流速度加快是近年来鄱阳湖水文环境发生显著变化的主要原因。.通过本课题的研究，建立了3个典型人工林小流域卡口站、1980-2015年长时间序列土地利用数据库、赣南柑橘果园空间分布、森林树高和生物量空间分布数据集、鄱阳湖流域水库数据库、收集了主要河流水文控制站径流泥沙资料等基础资料，为后续工作开展奠定基础，同时研究结论也为鄱阳湖及其流域的综合治理提供参考。