Physical basis of time series models for water table depths

The relationship between precipitation excess and water table depth can be described by empirical time series models such as transfer function noise models, autoregressive exogenous variable models (ARX), and threshold autoregressive self-exciting open loop models. In this paper these models are interpreted in terms of the water balance of a soil column. A physically based ARX model is used to predict the effect of an intervention on the water table dynamics at two locations. It is shown that the physically based ARX model predicts the effect of interventions reasonably well.