风云卫星微波遥感土壤水分产品适用性验证分析
杨 纲, 郭 鹏, 李西灿, 万 红, 孟春红

(山东农业大学 信息科学与工程学院, 山东 泰安 271018)

微波遥感; 土壤水分; 风云; AMSR-2; 适用性

Assessment on Applicability of Soil Moisture Products of Fengyun Satellite Microwave Remote Sensing
YANG Gang, GUO Peng, LI Xican, WAN Hong, MENG Chunhong

(College of Information Science and Engineering, Shandong Agricultural University, Taian, Shandong 271018, China)

microwave remote sensing; soil moisture; FengYun; AMSR-2; applicability

备注

土壤水分是水文过程、生物生态过程和生物地球化学过程的关键因子,大面积土壤水分的长期观测对于研究旱涝监测,水资源管理和作物产量预测等至关重要。基于美国俄克拉荷马州西南部地区ARS(Agricultural Research Service)水文监测网络中LWREW(Little Washita Experimental Watershed)和FCREW(Ft.Cobb Reservoir Experimental Watershed)两个地区35个地面监测站的实测数据,对比分析了2016年5月1日至2018年5月31日的FY-3B,FY-3C和AMSR-2土壤水分产品在该区域的适用性。结果 表明:FY-3B和FY-3C土壤水分产品在LWREW和FCREW地区大部分时间都存在高估,FY-3B土壤水分产品在LWREW和FCREW地区的平均偏差(Bias)均值分别为0.057 0 m3/m3,0.048 4 m3/m3,FY-3C土壤水分产品在LWREW和FCREW地区的平均偏差(Bias)均值分别为0.060 2 m3/m3,0.053 2 m3/m3,且在降水较多的季节,高估现象更为明显。JAXA(Japan Aerospace Exploration Agency)的AMSR-2土壤水分产品在LWREW,FCREW地区存在低估,在两个地区平均偏差(Bias)均值分别为-0.100 5 m3/m3和-0.100 3 m3/m3。在LWREW地区FY-3B升轨土壤水分产品表现最好,与地面实测地表水分的相关系数(R=0.664 8)较高,并且有着较低的均方根(RMSE=0.075 5 m3/m3)和平均偏差(Bias=0.061 6 m3/m3)。在FCREW地区AMSR-2升轨土壤水分产品表现最好,有着最高的相关系数(R=0.726 1)和最低的无偏均方根(ubRMSE=0.035 7 m3/m3)。总体上,FY-3B,FY-3C,JAXA土壤水分产品都能够反映出LWREW,FCREW地区土壤水分随季节和降水变化的趋势。

Soil moisture is a key variable in hydrological processes, bioecological processes, and biogeochemical processes. Long-term observations of soil moisture over large areas are critical to research on flooding and drought monitoring, water resource management, and crop yield forecasts. This paper is based on the measured data of 35 ground monitoring stations in the Agricultural Research Service(ARS)hydrological monitoring network of Southwest Oklahoma, USA, LWREW(Little Washita Experimental Watershed)and FCREW(Ft. Cobb Reservoir Experiment Watershed). The applicability of moisture products of FY-3B, FY-3C and AMSR-2 soil from May 1, 2016 to May 31, 2018 in this region was analyzed and compared. The results show that the soil moisture products of FY-3B and FY-3C are overestimated compared to the measured soil moisture in LWREW and FCREW areas, the average biases of the FY-3B soil moisture products in the two regions are 0.070 m3/m3 and 0.048 4 m3/m3, and the average biases of the FY-3C soil moisture products in the two regions are 0.060 2 m3/m3 and 0.053 2 m3/m3. In the seasons with more precipitation, the retrieval results of FY-3B and FY-3C products increased significantly, and JAXA soil moisture products underestimated the soil moisture in LWREW and FCREW areas, with an average biases of -0.100 5 m3/m3 and -0.100 3 m3/m3 in both regions. In the LWREW area, the FY-3B ascending orbital soil moisture product performed best, with the higher correlation coefficient(R=0.664 8), the lower root mean square(RMSE=0.075 5 m3/m3)and the average bias(0.061 6 m3/m3). In the FCREW area, JAXA had the best performance, with the highest correlation coefficient(R=0.726 1)and lowest unbiased root mean square(ubRMSE=0.035 7 m3/m3). In general, FY-3B, FY-3C, and JAXA soil moisture products are able to reflect the seasonal changes of soil moisture in the LWREW and FCREW areas.