Spatial variability of remotely sensed soil moisture in a temperate-humid grassland catchment

Authors

  • Wen Liu,

    1. Centre for Hydrology, Micrometeorology and Climate Change, Department of Civil and Environmental Engineering, University College Cork, Cork, Ireland
    Search for more papers by this author
  • Xianli Xu,

    Corresponding author
    1. Centre for Hydrology, Micrometeorology and Climate Change, Department of Civil and Environmental Engineering, University College Cork, Cork, Ireland
    2. Center for Sustainable Water Resources, Bureau of Economic Geology, Jackson School of Geosciences, The University of Texas at Austin, TX, USA
    • X. Xu, Center for Sustainable Water Resources, Bureau of Economic Geology, Jackson School of Geosciences, The University of Texas at Austin, J.J. Pickle Research Campus, Bldg. 130, 10100 Burnet Rd., Austin, TX 78758, USA.

      E-mail: xuxianliww@gmail.com

    Search for more papers by this author
  • Gerard Kiely

    1. Centre for Hydrology, Micrometeorology and Climate Change, Department of Civil and Environmental Engineering, University College Cork, Cork, Ireland
    Search for more papers by this author

ABSTRACT

Knowledge of the spatial distribution of soil moisture is very important for understanding eco-hydrological processes, but monitoring of soil moisture over extensive areas remains a challenge because of its high spatial variability and temporal dynamics. This study, taking an Irish temperate-humid catchment as an example, shows that the backscatter coefficient acquired from Environmental Satellite (ENVISAT) Advanced Synthetic Aperture Radar (ASAR) Wide Swath (WS) image (150 m resolution) is a good estimator of the surface (top 5 cm) soil moisture, leading us to propose an empirical model for soil moisture estimation. Statistical analysis of the remotely sensed soil moisture (produced from 35 ENVISAT/ASAR WS images spanning both wet and dry regimes in 2006) revealed that the spatial variation (standard deviation and coefficient of variance) mainly decreased with increasing mean soil moisture for this wet catchment. Geostatistical analysis showed that the spatial dependence of the soil moisture field was moderate with most of the nugget ratios ranging from 45% (25% percentile) to 55% (75% percentile), and the spatial correlation range was around 554–854 m. The exponential model was able to accurately fit the sample semivariograms and was a reliable estimator of the characteristics of the remotely sensed soil moisture field. This type of analysis can provide meaningful information for soil moisture monitoring at the catchment or regional scale. Copyright © 2011 John Wiley & Sons, Ltd.

Ancillary