论文部分内容阅读
本项工作在黑河流域中上游区域内,利用地下4cm深度的地面实测土壤水分数据验证了2012年7月至2014年12月期间AMSR2的两种算法产品——日本宇航局标准算法土壤水分产品(JAXA产品)和阿姆斯特丹自由大学联合美国宇航局开发的陆表参数反演模型算法土壤水分产品(LPRM产品)。验证结果显示:与地面实测数据相比,所有验证像元上两种土壤水分产品的均方根误差RMSE(Root Mean Square Error)普遍超过了0.1m~3/m~3。JAXA产品动态变化范围较小,升轨产品的总体精度略高于降轨,相比地面实测数据均存在明显的低估,在冻季与实测数据比较接近。LPRM产品动态范围较大,降轨产品在冻季不可用,在未冻季升轨产品精度高于降轨,相比地面实测数据有高估的倾向。同时,还进一步讨论并分析了两种算法对土壤温度和植被的不同处理方式对土壤水产品精度的可能影响,指出了算法可能的改进方向。
This work validated the two algorithms of AMSR2 from July 2012 to December 2014 using the ground-based measured soil moisture data at a depth of 4 cm underground in the mid-upper reaches of the Heihe River basin - Japan’s Nasa standard algorithm Soil Moisture Products JAXA product) and Amsterdam Free University joint NASA development of surface parameters inversion model algorithm for soil moisture products (LPRM products). The validation results show that the Root Mean Square Error (RMSE) of the two soil moisture products over the validated pixels generally exceeds 0.1 m ~ 3 / m ~ 3 compared with the measured data. The dynamic range of JAXA products is small, and the overall accuracy of the ascending rail products is slightly higher than that of the falling orbit. Compared with the ground-based measured data, there is a significant underestimation, which is close to the measured data in the frozen season. The dynamic range of LPRM products is large, and the product falling track is not available in the frozen season. The accuracy of the product rise in the unfrozen quarter is higher than that of the falling track, which is overestimated compared with the measured data on the ground. At the same time, the possible influence of the two algorithms on the precision of soil aquatic products under different treatments of soil temperature and vegetation is also discussed and analyzed. The possible improvement directions of the algorithm are pointed out.