A terrain and data-based method for generating the spatial distribution of soil moisture

Soil moisture is a critical environmental variable, as it significantly affects infiltration, evapotranspiration and surface and subsurface runoff processes. Measurement of soil moisture, however, has only recently become possible over large scales. In modelling applications, the spatial distribution of soil moisture is typically assumed to mirror that of a terrain attribute such as the wetness index of [Beven KJ, Kirkby MJ. A physically based, variable contributing area model of basin hydrology. Hydrol Sci Bull 1979;24(1):43-69] (e.g. TOPMODEL [2]. Intense ground-based measurements of the spatial distribution of soil moisture in the root zone have been conducted in Australia and New Zealand, allowing examination of the predictive power of various terrain attributes with respect to soil moisture. Here a method is developed that enables the generation of realistic spatio-temporal patterns of soil moisture based on the changing importance of spatial attributes with average wetness condition (theta) over bar (t). Maps of terrain and other spatial attributes are combined and weighted according to relationships dependent on (theta) over bar (t) to generate simulated spatial patterns of soil moisture. The method is presently based on field data and requires spatial patterns of soil moisture for a range of (theta) over bar (t) conditions in order to establish relationships between (theta) over bar (t) and the attributes. Observed variability is introduced by adding an error term whose statistics are also a function of (theta) over bar (t). Ultimately generalised weightings could be derived, based on the literature or subjective assessment by the user. (C) 2004 Elsevier Ltd. All rights reserved.