Hydrological model sensitivity to climate change can be defined as the response of a particular hydrological model to a known quantum of climate change. This paper estimates the hydrological sensitivity, measured as the percentage change in mean annual runoff, of two lumped parameter rainfall-runoff models, SIMHYD and AWBM and an empirical model, Zhang01, to changes in rainfall and potential evaporation. These changes are estimated for 22 Australian catchments covering a range of climates, from cool temperate to tropical and moist to arid. The results show that the models display different sensitivities to both rainfall and potential evaporation changes. The SIMHYD, AWBM and Zhang01models show mean sensitivities of 2.4%, 2.5% and 2.1% change in mean annual flow for every 1% change in mean annual rainfall, respectively. All rainfall sensitivities have a lower limit of 1.8% and show upper limits of 4.1%, 3.4% and 2.5%, respectively. The results for potential evaporation change are-0.5%,-0.8% and-1.0% for every 1% increase in mean annual potential evaporation, respectively, with changes rainfall being approximately 3 to 5 times more sensitive than changes in potential evaporation for each 1% change in climate. Despite these differences, the results show similar correlations for several catchment characteristics. The most significant relationship is between percent change in annual rainfall and potential evaporation to the catchment runoff coefficient. The sensitivity of both A and B factors decreases with an increasing runoff coefficient, as does the uncertainty in this relationship. The results suggest that a firstorder relationship can be used to give a rough estimate of changes in runoff using estimates of change in rainfall and potential evaporation representing small to modest changes in climate. Further work will develop these methods further, by investigating other regions and changes on the subannual scale.

No abstract

No abstract

The origin and evolution of the USDA SCS curve number method for estimating runoff from small ungauged rural catchments is traced, and the characteristics of the method are examined. When the method is expressed as an infiltration equation, the infiltration rate becomes dependent on both total storm rainfall and rainfall intensity. When expressed as a spatially varied saturation overland flow model, the method implies that some part of any catchment has infinite surface storage capacity. The lack of physical reality in the formulation of the method is an inherent limitation to any further development. A major weakness is the sensitivity of estimated runoff to errors in the selection of the curve number. Changes of about 15-20% in the curve number doubles or halves the total estimated runoff. The results of some Australian studies where curve numbers have been calibrated against actual runoff data are collated.

Multiple linear regressions are used to relate average annual runoff to average annual rainfall and areal potential evapotranspiration (PET) using data from 213 catchments grouped according to location in six of the major Drainage Divisions of Australia. A method is presented for estimating daily runoff from daily rainfall data using the AWBM model, which self-calibrates its surface storage parameters to the estimate of average annual runoff from the regressions, and using default values for its baseflow parameters. Two-thirds of the estimates of average annual runoff were within +/-25% of the actual value. The approach can also estimate satisfactorily the monthly and annual runoff series in many catchments, with the simulations being only slightly poorer than those obtained by directly calibrating the AWBM against recorded runoff.