The derivation and validation of a new model for the interception of rainfall by forests

A dynamic model and an analytical model derived therefrom for predicting the grossinterception of forest canopies are described. The dynamic model is validated using arunning balance in time between rainfall, throughfall, evaporation and changes in thecanopy storage. The canopy storage is increased by rainfall interception and depletedby evaporation and drainage. The evaporation rate varies with the amount of water in thecanopy and is estimated by numerical methods because sufficient meteorological dataare unavailable. Drainage rate expressions, similar to Rutter's exponential relationshipbetween drip rate and water storage, are shown to be inadequate during the period ofrainfall. A new drip expression which explicitly includes the rain rate, in addition to thestored water, is proposed and tested. This new expression fits the observed drip ratebetter and gives significantly better model prediction with fewer empirical parametersthan the exponential form of drip rate. However, because one of the drip parameters inthe dynamic model varied from one storm to the next and is a complicted function ofrainfall characteristics, the dynamic model was simplified to an analytical model forpredicting the gross interception loss. This analytical expression is easier to use becauseit is not sensitive to the exact value of the drip parameters; it predicted the total observedgross interception loss for the 20 storms tested to within 4%. It is suggested that the newdrip expression used in the dynamic model describes better the dislodgement of previouslyintercepted rain droplets by falling rain droplets. Finally, a new model for estimatingevaporation rates from forest canopies is proposed and discussed.