Development of a simple, catchment-scale, rainfall-evapotranspiration-runoff model.

Representation of the hydrological interaction between the land surface and the atmosphere requires considerable improvement, particularly for predicting evapotranpiration feedbacksas inputs to climate and earth system models. The predictive model developed here attempts to utilize a water balance approach that aims to extract information from the masses of catchment-scale time series data available on precipitation, energy related variables and stream discharge. Thus it begins with a few simple assumptions in order to seek systematically some synthesis of the climate and landscape controls on evapotranspiration feedbacks and discharge yields. The model adopts the hydrograph identification approach in the linear module of the rainfall-runoff model IHACRES but replaces the previously statistically based non-linear evapotranspiration loss module by a catchment moisture deficit accounting scheme. One advantage of this more conceptual approach is that evapotranspiration can be output on the same time step at which precipitation and energy variables are available (such as from GCMs), and this time step can be shorter (e.g half hourly) than the discharge time step (e.g. daily) used to calibrate the model parameters.

Figure 3: Evapotranspiration in the Scott catchment