Influence of Temporal Variability of Rainfall on Interception Loss

An interception model is derived that links the temporal variability of rainfall with the storm-based description of the interception process. Analytical formulae are obtained for precipitation occurrence with idealized statistical characteristics.

The analysis of the results indicates that point interception loss is controlled primarily by three time scales: the mean inter-storm arrival time $\tau_a$, the mean storm duration $\tau_r$, and the time to evaporate a saturated canopy $\tau_0$ which depends on canopy water holding capacity $W_c$ and the wet canopy potential evaporation rate $E_{I0}$, and less importantly, on storm intensity. Additional assumption about rainfall stationarity leads to a relation between long-term interception loss and gross rainfall that requires a very small amount of input data.

The interception loss predicted by the analytical model agree well with that of a Rutter model driven by a synthetic rainfall time series with the same statistics. Using the parameter values estimated from the observed rainfall data in the Amazon and southwestern France, the analytical results predict long-term interception losses very close to those observed.