Parsimonious modelling the rainfall-runoff behaviour of large catchments in Thailand
PublisherUniversity of Lancaster
MetadataShow full item record
There is currently great concern in Thailand about the effects of land-use and climate change on the rainfall-runoff behaviour of the great rivers in Thailand. At the same time, hydrologists are becoming increasingly worried about uncertainties in the forecasts of physics-based models - due to the complexity of their structures and inappropriateness of the catchment characteristics incorporated. These concerns become particularly acute when such models are applied to large river catchments (100 - 10,000 km2) - the scales needed to link hydrological research with land-use management and planning.Within this thesis, the value of using parsimonious models (i. e., structurally very simple with few parameters but a high simulation efficiency), Normally restricted to engineering applications, is addressed. The particular modelling techniques used form part of new suite of tools known as Data-Based-Mechanistic (DBM) modelling. Three very large catchments, one in northern Thailand, one in the centre and one in the south were the focus for the study and each contained one or more smaller catchments. The results of the DBM modelling showed that much of the rainfall-runoff dynamics of these large catchments could be modelled with very simple (3 or 4-parameter) models and that these parameters were related to the catchment size, hydrogeology and climatic regime. The role of the climate was shown with the DBM tool known as the Dynamic Harmonic Regression (DHR) model. These new methods were complemented by traditional hydrological engineering analyses, which supported the findings of the DBM. Indeed, these traditional techniques clearly showed the effect of rapid and extensive urbanisation on catchment hydrology. In an attempt to show how such parsimonious methods might be extended to incorporate additional information, where it can be justified, catchment topographic analyses were compared with measured catchment moisture measurements. The poor performance of the topographic model evaluated did, however, mean that further validation work needs to be undertaken before this particular topographic model should be applied (without similar testing) throughout Thailand. Given the greater use of physics-based models of catchment systems in Thailand, this work demonstrates that parsimonious statistical approaches remain of great value to the interpretation of hydrological processes operating within large Thai catchments, and indeed to the identification of changes in these processes with land-use and climate change.