My current research fits into 5 main themes:
The sustained European drought in 2003 brought the issue of the impact of climate change on water resources to the attention of the media. Realistically, we are already facing problems with water supplies in many parts of the UK, particularly in the South East and Thames regions. Here, Thames Water have applied for permission to build a desalinisation plant in Kent (the most expensive way to produce a potable water supply) and for permission to build 7 new reservoirs. On the 14th June 2004, the Environment Agency called for additional measures to conserve water as "some parts of the country have less water per head than the desert states of Syria and Sudan." The Government's plans to build an additional 900,000 homes in the South East will worsen the crisis.
Water resource reliability, resilience and vulnerability under climate variability and change was the central theme in my PhD "The impacts of climatic change and variability on water resources in Yorkshire", which was completed in December 2000.
As part of the SWURVE (Sustainable Water, Uncertainty, Risk and Vulnerability in Europe) project, we conducted an investigation into the impacts of climate change, in particular the UKCIP02 Medium-High Scenario, upon water resources in northwest England. This work was in collaboration with the Environment Agency and United Utilities and considered the impacts of climate change on a very complex water resource system, the northwest Integrated Resource Zone. Results from this study were published in two papers in 2007 and can be found here.
The impact of climate variability and change on water resources is also central to many of the catchment and river basin case studies with the FP6 AquaTerra project which started in June 2004. Preliminary analyses of the impacts of climate change on groundwater systems in the Brevilles aquifer, France and the Geer catchment (part of the Meuse river basin) in Belgium and surface water systems in the Centa catchment in Italy and the Gallego catchment in Spain have now been carried out. This work is not yet published but watch this space and that of the AquaTerra website for details.
During the late 1980s and early 1990s there were a sequence of drought events in the UK, culminating in the serious drought of 1995-96 which mainly affected the north of England. The Yorkshire region was particularly badly affected and road tankering was necessary to sustain water supplies. My PhD research investigated the incidence of historic drought in the Yorkshire region, and how climatic variability and atmospheric circulation patterns are linked to the occurrence of drought in the UK.
More recently, together with a colleague Stephen Blenkinsop, I have investigated how the occurrence of drought may change in the future using outputs from regional climate models. Studies were part of the AquaTerra project and were published in 2 papers in 2007, one on the impacts of climate change on drought at the European scale and one on droughts in the UK at the water resource zone resolution. Details of this study can be found here.
During the SWURVE project, partially as a reaction to recent severe flooding in both the UK (autumn 2000) and Europe (summer 2002), we investigated how extreme rainfall events have changed over the observed record from 1960-2000. Similarly, we investigated how we expect these extreme events to change under future climate, and the impacts on future flood risk, using the results from the HadRM3H model from the UK Met Office Hadley Centre. As an update to this work, as part of my NERC fellowship, we investigated how extreme precipitation is expected to change under global warming at the European scale using the results from the PRUDENCE project - 6 different regional climate models (RCMs) were investigated and the first probabilistic projections of future change in extremes was made (using equal weighting of models) and published in 2007. Details of this study can be found here. A more recent study has taken the results of 13 RCMs from PRUDENCE and looked at changes projected to extreme rainfall across the UK but tried to weight the results by how well the models reproduce both the magnitude and the spatial pattern of extremes.
We will use the information on observed and projected trends in extreme precipitation and droughts to establish weighting schemes for RCMs to apply to produce probabilistic projections for the AquaTerra project.
In the simplest terms, an increase in flooding can be caused by either an increase in the magnitude and frequency of extreme rainfall events or a change within the catchment that causes the rain that falls to reach the drainage network faster. It has been suggested that land use change in the Ouse catchment has increased the speed with which rain reaches the drainage network, thus increasing both the magnitude and frequency of flood events at York (Lane, 2002). He argues that there is 'no trend in rainfall that can be linked to the observed increase in flooding'.
Research undertaken by myself and Chris Kilsby, has already shown that rainfall patterns have changed substantially in the Yorkshire region over the past 100 years. We have also shown that extreme rainfall events in the UK are increasing in both magnitude and frequency, and are likely to continue this trend under future climate change.
Further work will address the relative contributions of land use change and climate change/variability to the apparent increase in flooding at York.
It is thought that mountainous regions may provide important and sensitive early indicators of global climate change as temperatures are observed to be rising more rapidly at higher elevations. High quality precipitation and temperature data is available for the Upper Indus Basin (UIB) in the Karakoram region, Pakistan, extending back to around 1900. These records may be important as anthropogenic changes to the environment which impact on measurements of temperature are likely to be small and thus have the potential to provide a clearer indication of the impact of global climatic change than lowland or foothill stations further south in India and Pakistan or even on the southern slopes of the Himalayas.
The climate of the UIB, and particularly the Trans-Himalayan region of the Hindu Kush and Karakoram Mountains, is also of great significance both locally and regionally as more than 80% of the flow in the River Indus as it emerges onto the Punjab plains is derived from melting snow and ice. This has been shown to be highly correlated with summer mean temperature. Temperature change over the UIB may therefore significantly alter glacier mass balance and thus the hydrological regime of the Indus, affecting both local rural livelihoods in the mountains and the productivity of agriculture on the Indus plains.
Preliminary analysis of temperature and precipitation data has revealed contrasting trends. There is evidence of significant 'winter warming', coupled with increases in winter precipitation throughout the region. In summer however, there is evidence of cooling but increases in precipitation. There is also evidence of connections to North Atlantic weather systems that are also thought to be changing.
A PhD student, Nathan Forsythe, funded by NSF is due to take up this topic for his PhD in fall 2008.
As part of the SWURVE project we assessed the control climate of the HadRM3H model (used for the UKCIP02 climate scenarios). We assessed the ability of the climate model to reproduce both mean climate and extremes (extreme rainfall and drought events). As part of the BETWIXT project we are further assessing the ability of the model to reproduce observed daily rainfall statistics which can then be used in rainfall modelling.
Downscaling of information from climate models is a very important part of impact assessment and there are two main types: statistical and dynamical downscaling. Statistical downscaling generally develops relationships between atmospheric circulation parameters and rainfall statistics on the ground. These can then be used directly or as input to a rainfall model. In my PhD, I investigated the use of statistical downscaling of climate information combined with rainfall modelling. More recently, as part of the SWURVE project, we have investigated the use of dynamical downscaling, or the direct use of data output from regional climate models. It is hoped that the results of this research will be available soon.
Uncertainty in the impacts of climate change comes from many sources, not least the downscaling methodology, model and emissions scenario chosen. The effects of uncertainty have been investigated as part of the SWURVE project and results will be available soon in a special issue of Hydrology and Earth System Sciences. This work will be continued and expanded under the CRANIUM project.