Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Titre : Climate impacts on streamflow in Australian catchments

Auteur : Ye, Wei

Université de soutenance : Australian National University

Grade : Doctor of Philosophy (1996)

Résumé partiel
Through conceptual rainfall-runoff model analysis and development, one objective of this thesis is to develop and assess a methodology for predicting the impact of climate variability and possible climate change on streamflow in the Australian region. A simple lumped dynamics-encapsulating conceptual rainfall-runoff model, IHACRES (Identification of Hydrographs And Components from Rainfall, Evapotranspiration and Streamflow data), was successfully extended to low-yielding, ephemeral catchments which represent the basic characteristics of a large number of Australian catchments located in arid and semi-arid regions. Comprehensive comparison of its performance with three other more complex conceptual models demonstrated its performance to be very competitive. The model shows a strong capability to extract the information in rainfall-runoff records. Models to predict daily catchment streamflow can be developed on just a few years of data, allowing the most reliable period of record to be identified for model calibration. The applicability of the IHACRES model was then verified through application to the Australian benchmark catchment network (Chapter 6). The model was successfully applied to 19 Australian benchmark catchments to predict daily runoff. Comparison of the model’s performance with another complex conceptual model indicates that it is competitive in calibration (fitting) statistics with that more complex model. The simple model has been additionally tested for its ability to predict independent discharge time series. Satisfactory "validation" based on calibration models obtained from relatively short calibration periods (2-5 years) further manifests the ability of the simple conceptual model to extract enough "information" from the data to be useful for predicting long term variation in runoff from climate records.

Présentation

Version intégrale ( 166 Mb)