Soil moisture active and passive microwave products: intercomparison and evaluation over a Sahelian site

Hydrol. Earth Syst. Sci., 14, 141–156, 2010 www.hydrol-earth-syst-sci.net/14/141/2010/ © Author(s) 2010. This work is distributed under the Creative Commons Attribution 3.0 License. Hydrology and Earth System Sciences Soil moisture active and passive microwave products: intercomparison and evaluation over a Sahelian site C. Gruhier1 , P. de Rosnay2 , S. Hasenauer3 , T. Holmes4 , R. de Jeu5 , Y. Kerr1 , E. Mougin1 , E. Njoku6 , F. Timouk1 , W. Wagner3 , and M. Zribi1 1 Centre d’Études Spatiales de la BIOsphère, UMR 5126 (CNRS, CNES, IRD, UPS), Toulouse, France 2 European Centre for Medium-Range Weather Forecasts, Reading, UK 3 Institute of Photogrammetry and Remote Sensing, Vienna University of Technology, Vienna, Austria 4 USDA-ARS Hydrology and Remote Sensing Laboratory, Beltsville, USA 5 Faculty of Earth and Life Sciences, Vrije Universiteit, Amsterdam, The Netherlands 6 Jet Propulsion Laboratory, Pasadena, USA Received: 3 August 2009 – Published in Hydrol. Earth Syst. Sci. Discuss.: 5 August 2009 Revised: 18 December 2009 – Accepted: 23 December 2009 – Published: 22 January 2010 Abstract. This paper presents a comparison and an evaluation of five soil moisture products based on satellite-based passive and active microwave measurements. Products are evaluated for 2005–2006 against ground measurements obtained from the soil moisture network deployed in Mali (Sahel) in the framework of the African Monsoon Multidisciplinary Analysis project. It is shown that the accuracy of the soil moisture products is sensitive to the retrieval approach as well as to the sensor type (active or passive) and to the signal frequency (from 5.6 GHz to 18.8 GHz). The spatial patterns of surface soil moisture are compared between the different products at meso-scale (14.5◦ N – 17.5◦ N and 2◦ W – 1◦ W). A general good consistency between the different satellite soil moisture products is shown in terms of meso-scale spatial distribution, in particular after convective rainfall occurrences. Comparison to ground measurement shows that although soil moisture products obtained from satellite generally over-estimate soil moisture values during the dry season, most of them capture soil moisture temporal variations in good agreement with ground station measurements. 1 Introduction Surface soil moisture is a key variable which controls the water and energy exchanges at the soil-vegetation-atmosphere interface. Koster et al. (2004) showed that the soil moisture feedback with precipitation is very strong in the three regions Correspondence to: C. Gruhier () of the US Great Plains, Asia and West Africa. In particular, in the Sahelian region of West Africa, Taylor et al. (2007) and Taylor (2008) showed that soil moisture and land surface processes influence meso-scale convective systems dynamics. Quantitative soil moisture assessment is crucial for land surface modelling and understanding as well as for numerical weather prediction purpose. However, due to its high temporal and spatial variability, it is difficult to provide accurate quantitative information on soil moisture at regional and global scales. Several coordinated land surface modelling activities have provided insight into quantitative soil moisture characterisation at regional and global scale (Dirmeyer et al., 2006; Boone et al., 2009). Satellite remote sensing approaches also open the possibility to provide spatially integrated information on soil moisture over large areas. Microwave remote sensing at low frequencies is the most efficient approach to characterise soil moisture from space, with low atmospheric contribution (Njoku and Entekhabi, 1996; Jones et al., 2004; Wagner et al., 2007; Kerr, 2007). Various active and passive microwave sensors have been measuring Earth emissions and reflection for several years. The Advanced Microwave Scanning Radiometer on Earth Observing System (AMSR-E) on the AQUA satellite is a passive microwave sensor. It has been providing brightness temperature at five frequencies from 6.9 to 89 GHz since 2002. AMSR-E C-band (6.9 GHz) and X-band (10.7 GHz) channels are suitable for soil moisture remote sensing (Njoku et al., 2003). On the Tropical Rainfall Measuring Mission (TRMM) satellite, the TRMM Microwave Imager (TMI) has been measuring microwave emission at five frequencies from 10.7 GHz to 85.5 GHz since 1997. The wind scatterometer on the European Remote Sensing (ERS) satellites have Published by Copernicus Publications on behalf of the European Geosciences Union. 142 C. Gruhier et al.: Evaluation of microwave soil moisture products been performing continuous active microwave measurements at C-band (5.3 GHz) for 1991–1996 (ERS-1) and since 1996 (ERS-2) (European Space Agency, 1997). Their continuity has been ensured since 2006 by the Advanced Scatterometer (ASCAT) on the Meteorological Operational satellite (METOP). METOP/ASCAT has been providing near realtime soil moisture products since 2008. The ERS/SCAT and METOP/ASCAT series provides the longest consistent and continuous global scale soil moisture data set since 1992. SMOS (Soil Moisture and Ocean Salinity) satellite of the European Space Agency (ESA), launched on 2 November 2009, is the first satellite devoted to soil moisture remote sensing. SMOS measurements use an L-band interferometer which has been shown to be optimal to capture soil moisture information from space (Kerr et al., 2001). From 2014 it should be followed by the Soil Moisture Active and Passive (SMAP) satellite of NASA which, by combining active and passive approaches, will provide soil moisture products at high resolution (http://smap.jpl.nasa.gov/). Soil moisture retrieval is based on the relationship between soil moisture and soil dielectric constant which influences brightness temperatures and scatterometer coefficient from passive and active microwaves sensors, respectively. The sensitivity to soil water content might also be affected by Radio Frequency Interference (RFI) and vegetation optical depth, which are both accounted for in the retrieval algorithms. Although these soil moisture products are provided at relatively coarse resolutions, disaggregation approaches have been investigated in the past few years (Merlin et al., 2008). They proved to be highly relevant to provide soil moisture information at kilometer scale. An important issue in remote sensing approaches concerns products validation. Several papers investigated soil moisture products evaluation (Dirmeyer et al., 2004; Pellarin et al., 2006; Wagner et al., 2007; Draper et al., 2009; Rüdiger et al., 2009). Draper et al. (2009) provided a comparison of four soil moisture products all based on AMSRE sensor over a temperate climate in Australia during 2006. Rüdiger et al. (2009), showed a comparison of three products (and one simulation) over the mainland of France from 2003 to 2005, in addition to a ground measurements comparison. Gruhier et al. (2008) provided an evaluation of the AMSR-E soil moist (...truncated)