| Volumes and Issues Contents of Issue 5 |
www.hydrol-earth-syst-sci-discuss.net/6/5753/2009/
doi:10.5194/hessd-6-5753-2009
© Author(s) 2009. This work is distributed
under the Creative Commons Attribution 3.0 License.
Selection of an appropriately simple storm runoff model
CSIRO Land and Water, Canberra, ACT, Australia
Abstract. Alternative conceptual storm runoff models, including several published ones, were evaluated against storm flow time series for 260 catchments in Australia (23–1902 km2). The original daily streamflow data was separated into baseflow and storm flow components and from these, event rainfall and storm flow totals were estimated. For each tested model structure, the number of free parameters was reduced in stages. The appropriate balance between simplicity and explanatory power was decided based on Aikake's Final Prediction Error Criterion and evidence of parameter equivalence. The majority of catchments showed storm recession half-times in the order of a day, with more rapid drainage in dry catchments. Overland and channel travel time did not appear to be an important driver of storm flow recession. A storm runoff model with two free parameters (one related to storm event size, the other to antecedent baseflow) and a fixed initial loss of 12 mm provided the optimal model structure. The optimal model had some features similar to the Soil Conservation Service Curve Number technique, but performed an average 12 to 19% better. The non-linear relationship between event rainfall and event runoff may be associated with saturated area expansion during storms and/or the relationship between storm event size and peak rainfall intensity. Antecedent baseflow was a strong predictor of runoff response. A simple conceptual relationship between groundwater storage and saturated catchment area proved adequate and produced realistic estimates of saturated area of <0.1% for the driest and >5% for the wettest catchments.
Discussion Paper (PDF, 794 KB) Interactive Discussion (Closed, 4 Comments) Final Revised Paper (HESS)
Citation: van Dijk, A. I. J. M.: Selection of an appropriately simple storm runoff model, Hydrol. Earth Syst. Sci. Discuss., 6, 5753-5782, doi:10.5194/hessd-6-5753-2009, 2009. Bibtex EndNote Reference Manager XML
