In recent decades there has been a significant increase in reported flood events and flood losses. Although these losses may be attributed to improvements in reporting and increased exposure in flood prone areas, a perception now exists that flood risk is increasing as a direct result of anthropogenic global warming. In response to this perceived risk, research focused on producing projections of future flood risk has been receiving considerable attention. Indeed, climate impact studies are now being used to guide and test government policy. However, there are significant uncertainties associated with the application of climate model output in flood impact studies. Moreover, there is a disparity between current impact studies and the information required by decision makers, with studies typically focussing on changing river flows. This thesis aims to bridge this gap, cascading climate model output through to impacts at the building scale under an uncertainty framework. The overall aim is to explore the feasibility of using future flood projections as a decision making tool and ultimately to better inform decision makers. The first component of this research was focussed on investigating current climate models and exploring the suitability of their application in flood impact studies. It was found that poor model performance currently precludes their application in assessing flood risk in some regions. The results also provided recommendations for future flood impact studies; these were then used to inform the cascade modelling framework. The rest of the thesis details the development of the modelling framework, driving ensemble climate projections through hydrological, hydraulic and damage models. The modelling framework was also structured to enable uncertainty under current climate conditions to be explored. For the first time, this work presents uncertain climate projections in terms of damage at the building scale. The results reveal that there is an increase in flood magnitude under future climate conditions, however there is significant variability between projections. In fact, the results reveal that there is significant uncertainty under current climate conditions with the potential for exceptional flooding regardless of any future change. The results also emphasise the need to model damage in impact studies as the assumption of linearity between changing hazard and changing risk is invalid. The research presented here has proposed suitable methods for informing decision makers and demonstrated that there is significant scope for improvement in climate impact studies.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:683693 |
| Date | January 2014 |
| Creators | Smith, Andrew M. |
| Publisher | University of Bristol |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0018 seconds