THE WATER-USE OF THE BALQUHIDDER CATCHMENTS - A PROCESSES APPROACH

One annual and four seasonal models were operated with meteorological data for 1984 and 1985 to estimate the water use of the Balquhidder catchments. The most physically realistic models operated on three classes of vegetation (grass, heather and forest) distributed in three altitudinal zones and included transpiration estimates based on the Penman-Monteith equation and a daily exponential model of interception loss. Comparison of the model estimates indicates that the simpler models involving a larger number of assumptions but requiring fewer input variables gave results similar to those of the more physically sound models. The reasons for this are discussed. The most physically realistic model was operated with meteorological data for 1984-1989 to produce estimates of the water use of the Monachyle (grassland and heathland) and Kirkton (forest and grassland) catchments. The estimates differ systematically from catchment measurements of precipitation minus runoff but lie within experimental uncertainties. Reasons for the differences are examined and the sensitivity of the model to different assumptions and uncertainties is investigated.