The effect of intercepted rainfall on the water balance of a hardwood forest (original) (raw)

1979, Water Resources Research

Evaporation from a large hardwood forest is estimated from measurements of the required meteorological variables and from measured stomatal resistances. A correction factor is derived to overcome the incorrect assumption that evaporative demand remains the same during wet and dry periods. The factor is based on the ratio of the isothermal resistances during the wet and subsequent dry periods. The stomatal measurements were converted to canopy resistances by dividing by the leaf area index and were used to obtain the water balance for the entire season. The results are analyzed to show that evaporation from a wet canopy is often 2-3 times greater than transpiration from the same surface. Rates of interceptionai loss, calculated from the wet and dry evaporation rates, were verified by direct measurements of throughfall and stemflow. Net interceptional loss, equal to the excess evaporation from a wet canopy over a dry one, depended on rainfall duration and character and was on the average about 60-80% of total interception, In the overall summer water balance of 442 mm of precipitation and 52-mm depletion of soil storage, transpiration via the trees accounted for 261 mm; evaporation from the wetted leaves and branches for 111 mm, and runoff for 131 mm, giving an error of closure in the water balance of only 9 mm. If transpiration only had been used instead of interception when the canopy was wet, the error in the water balance would have been 100 mm. INTRODUCTION At present there is much controversy regarding the effect of intercepted rainfall on the water balance of forested watersheds. The core of the argument is whether rain retained by vegetation represents a total loss or no loss of moisture beyond the normal evapotranspiration of the canopy. The dis-' agreement stems largely from the way in which intercepted rainfall is viewed. If it is considered as a reduction in rainfall reaching the ground, then it could indeed be a total loss. This early view was shared by hydrologists who looked at interception only in terms of the inputs of the hydrologic cycle. But if the water cycle of the entire soil-vegetation complex is considered, then intercepted rainfall is not a total loss, because when the foliage is wetted, transpirational withdrawal of soil moisture is reduced. The magnitude of transpirational saving during the evaporation of intercepted rainfall is therefore critical in determining how much of a moisture loss interception constitutes. If the rate of water loss is the same whether the vegetation is wet or dry, then it matters little whether evaporative demand is satisfied by intercepted water or soil moisture. Burgy and Pomeroy [1958] found that in vigorously growing laboratory grass plots the evaporation of a given amount of intercepted water (sprinkled on the plots) corresponded to an almost equal reduction in the amount of transpiration from the plants, in that total moisture loss was approximately the same for plots with wetted and dry leaf surfaces. Field studies by McMillan and Burgy [1960] on short, clipped grass in California gave similar results. The early ideas, based on these experiments, maintained that since a given supply of energy will evaporate only a given amount of water, then the evaporation of the moisture retained by the foliage must be compensated for by an equal reduction in transpiration. If, on the other hand, evaporation of intercepted rainfall is Copyright ¸ 1979 by the American Geophysical Union.