Home
Remote sensing methods for soil moisture assessments
Remote sensing methods provide viable alternatives as numerous studies have established relationships between satellite observations, surface wetness, and soil moisture mapping (Georgakakos et al. 1996; Lakshmi et al. 1997; Basist et al. 1998; Verhoest et al. 1998; and others). The remote sensing approach to measure soil moisture uses solar, thermal infrared, and microwave radiation. In particular, active (radar) and passive (radiometry) microwave sensors have shown strong sensitivity to soil moisture content (Ulaby et al. 1982; Altese et al. 1996; Vinnikov et al. 1999; Space Studies Board and Oki et al. 2000; NRC 2001; Seto et al. 2003; and others). In contrast to direct in situ (or point) measurements, remote sensing techniques can provide improved spatial coverage that point-based observations cannot (Georgakakos et al. 1996). In addition, remote sensing techniques provide cost-effective ways to collect large datasets rapidly over large areas encompassing soil types with various textures, slopes, vegetation, and climatic conditions on a repetitive basis (Georgakakos et al. 1996).
Remote sensing methods can provide soil moisture maps for areas as small as 100 m2 to areas as large as 1,000 km2, encompassing soil types with various soil texture, various slopes, vegetation, and climatic conditions (Georgakakos et al. 1996). However, remote sensing techniques also have their disadvantages, including their inability to accurately represent spatiotemporal variability of soil moisture, the limited spatial resolution of sensors, and limiting factors of environmental conditions during remote sensing processes (Mohanty et al. 2000). Chauhan et al. (2003) used a combination of space-borne microwave (SSM/I) and optical/IR methods (AVHRR) to achieve high-resolution soil moisture maps. Although the soil moisture map generated at 25 km and over (low resolution) from SSM/I and at 1 km (high resolution) from AVHRR showed reasonably similar trends in magnitude and spatiotemporal pattern (Jackson et al. 1999; Chauhan et al. 2003), the soil moisture map at 1 km is not adequate for field scale studies since soil moisture varies spatially and temporally in much shorter steps (Fahsi et al. 1997; Bonta 1998). Soil moisture variability is noted in a scale of less than 100 m (Famiglietti et al. 1999). Another difficulty with analyzing remotely sensed soil moisture is its variability at different scales (Stewart et al. 1995)
as the spatial resolution of most spaceborne sensors ranges from ±30 m (Landsat ETM) to ±250–500 m (MODIS) to ±1,100 m (NOAA/AVHRR, MODIS) to ±5,000 m (METEOSAT) and even to ±50,000 m (ERS Scatterometer) or more (Verstraeten et al. 2008). Crow and Wood (2002) used a downscaling approach for coarse-scale mapping of this property and concluded that their methodology provided a simplified, and at times inaccurate, representation of subfootprint-scale soil moisture heterogeneity. The primary advantage of the downscaling procedure, based on spatial scaling, lies in its simplicity and ability to predict heterogeneity of soil moisture in fine scales when ancillary data are not available (Crow & Wood 2002). They also suggested that one promising strategy for fine-scale soil moisture mapping is the integration of high-resolution land use and soil data with the remotely sensed soil moisture (Crow & Wood 2002; Reichle & Koster 2004).
Remote sensing methods provide viable alternatives as numerous studies have established relationships between satellite observations, surface wetness, and soil moisture mapping (Georgakakos et al. 1996; Lakshmi et al. 1997; Basist et al. 1998; Verhoest et al. 1998; and others). The remote sensing approach to measure soil moisture uses solar, thermal infrared, and microwave radiation. In particular, active (radar) and passive (radiometry) microwave sensors have shown strong sensitivity to soil moisture content (Ulaby et al. 1982; Altese et al. 1996; Vinnikov et al. 1999; Space Studies Board and Oki et al. 2000; NRC 2001; Seto et al. 2003; and others). In contrast to direct in situ (or point) measurements, remote sensing techniques can provide improved spatial coverage that point-based observations cannot (Georgakakos et al. 1996). In addition, remote sensing techniques provide cost-effective ways to collect large datasets rapidly over large areas encompassing soil types with various textures, slopes, vegetation, and climatic conditions on a repetitive basis (Georgakakos et al. 1996).
Remote sensing methods can provide soil moisture maps for areas as small as 100 m2 to areas as large as 1,000 km2, encompassing soil types with various soil texture, various slopes, vegetation, and climatic conditions (Georgakakos et al. 1996). However, remote sensing techniques also have their disadvantages, including their inability to accurately represent spatiotemporal variability of soil moisture, the limited spatial resolution of sensors, and limiting factors of environmental conditions during remote sensing processes (Mohanty et al. 2000). Chauhan et al. (2003) used a combination of space-borne microwave (SSM/I) and optical/IR methods (AVHRR) to achieve high-resolution soil moisture maps. Although the soil moisture map generated at 25 km and over (low resolution) from SSM/I and at 1 km (high resolution) from AVHRR showed reasonably similar trends in magnitude and spatiotemporal pattern (Jackson et al. 1999; Chauhan et al. 2003), the soil moisture map at 1 km is not adequate for field scale studies since soil moisture varies spatially and temporally in much shorter steps (Fahsi et al. 1997; Bonta 1998). Soil moisture variability is noted in a scale of less than 100 m (Famiglietti et al. 1999). Another difficulty with analyzing remotely sensed soil moisture is its variability at different scales (Stewart et al. 1995)
as the spatial resolution of most spaceborne sensors ranges from ±30 m (Landsat ETM) to ±250–500 m (MODIS) to ±1,100 m (NOAA/AVHRR, MODIS) to ±5,000 m (METEOSAT) and even to ±50,000 m (ERS Scatterometer) or more (Verstraeten et al. 2008). Crow and Wood (2002) used a downscaling approach for coarse-scale mapping of this property and concluded that their methodology provided a simplified, and at times inaccurate, representation of subfootprint-scale soil moisture heterogeneity. The primary advantage of the downscaling procedure, based on spatial scaling, lies in its simplicity and ability to predict heterogeneity of soil moisture in fine scales when ancillary data are not available (Crow & Wood 2002). They also suggested that one promising strategy for fine-scale soil moisture mapping is the integration of high-resolution land use and soil data with the remotely sensed soil moisture (Crow & Wood 2002; Reichle & Koster 2004).