Validation of the ESA CCI Soil Moisture Product in China.
Ru An,Ling Zhang,Zhe Wang,Jonathan Arthur Quaye-Ballard,Jiajun You,Xiaoji Shen,Wei Gao,Lijun Huang,Yinghui Zhao,Zunyou Ke
DOI: https://doi.org/10.1016/j.jag.2015.09.009
IF: 7.5
2016-01-01
International Journal of Applied Earth Observation and Geoinformation
Abstract:The quality of a newly merged soil moisture product (ECV_SM v0.1) from active and passive microwave sensors has attracted widespread international attention. The performance evaluation of this product will benefit studies on climate, meteorology, agriculture, hydrology, ecology and the environment. In this study, meteorological station data and the Noah soil moisture product were used to validate the ECV_SM product in China. First, some conventional statistical measures, such as correlation coefficients, bias, root mean Square difference (RMSD) and mean relative error (MRE), were computed to describe the level of agreement between the meteorological station data and ECV_SM values. The accuracy was moderately high (the correlation was significant at the 0.05 level), although the two datasets differed slightly for various types of land cover. Compared with cropland and urban and built-up areas, the performance of ECV_SM was best in grassland regions. Second, the triple collocation technique was used to assess the random error in the meteorological station data, Noah soil moisture product and ECV_SM product. The mean errors in these three datasets were 0.108, 0.079 and 0.075 m(3) m(-3), respectively, on July 8, 2010 and 0.099, 0.061 and 0.059 m(3) m(-3), respectively, on October 8, 2010. Only two days of data were used for the triple collocation test as a representative, but this cannot precisely indicate that the test results on any other day correspond with the test results on these two days. Additionally, a trend analysis of ECV_SM during 2003-2010 was carried out using the Mann-Kendall trend test. (C) 2015 Elsevier B.V. All rights reserved.