Space-time modeling of soil moisture Academic Article uri icon

abstract

  • 2017 Elsevier Ltd A physically derived space-time mathematical representation of the soil moisture field is carried out via the soil moisture balance equation driven by stochastic rainfall forcing. The model incorporates spatial diffusion and in its original version, it is shown to be unable to reproduce the relative fast decay in the spatial correlation functions observed in empirical data. This decay resulting from variations in local topography as well as in local soil and vegetation conditions is well reproduced via a jitter process acting multiplicatively over the space-time soil moisture field. The jitter is a multiplicative noise acting on the soil moisture dynamics with the objective to deflate its correlation structure at small spatial scales which are not embedded in the probabilistic structure of the rainfall process that drives the dynamics. These scales of order of several meters to several hundred meters are of great importance in ecohydrologic dynamics. Properties of space-time correlation functions and spectral densities of the model with jitter are explored analytically, and the influence of the jitter parameters, reflecting variabilities of soil moisture at different spatial and temporal scales, is investigated. A case study fitting the derived model to a soil moisture dataset is presented in detail.

published proceedings

  • ADVANCES IN WATER RESOURCES

author list (cited authors)

  • Chen, Z., Mohanty, B. P., & Rodriguez-Iturbe, I.

citation count

  • 8

complete list of authors

  • Chen, Zijuan||Mohanty, Binayak P||Rodriguez-Iturbe, Ignacio

publication date

  • November 2017