FALLING AT THE FIRST HURDLE: SPATIAL RAINFALL VARIABILITY AND THE PROBLEM OF CLOSING CATCHMENT WATER BUDGETS IN TROPICAL MONTANE ENVIRONMENTS

摘要

The calculation of the catchment water budget is fundamental to understanding a catchment's hydrology. Water budgets are usually calculated by measuring those fluxes that are easiest to quantify (in theory precipitation, P and streamflow, Q) and using the difference between them as an estimate of the more difficult to measure evaporation term (ET). Other (usually unmeasured) fluxes such as catchment leakage, changes in soil water storage and other forms of precipitation input (such as snow and fog) are corrected for at best but often they are simply ignored. This approach is not particularly objective but works where precipitation inputs can be confidently predicted and where leakage (L) is small. In areas where precipitation and/or leakage estimates are potentially subject to significant error, closure of the budget is much more difficult. Many different combinations of P, ET and L can then close the budget but all but one of these combinations will be incorrect. Tropical montane volcanic uplands represent a case in point where inputs are extremely variable as a result of wind effects on precipitation and subterranean leakage can be both high and spatially variable. These combine to make standard water budget calculations extremely error prone. Without adequate knowledge of the spatial variability of precipitation inputs we 'fall at the first hurdle' in the race to quantify catchment water budgets.