Sampling uncertainty of climate change impacts on water resources: Case study using data from Iran and England

It is now generally accepted that anthropogenic greenhouse gas emissions are largely responsible for rapid global warming. Moreover, according to the United Nations Intergovernmental Panel on Climate Change (IPCC, 2001), it is likely that human-induced climate change will persist for many centuries to come. It is therefore inevitable that water resources, in arid and semi-arid regions in particular, will be subject to increased stress. Consequently, numerous studies have been carried out to assess the impact of climate change on water resource systems such as reservoirs. Although these studies employed different schemes for baseline climate perturbation, and different models for simulating the resulting catchment rainfallrunoff response, the impacts assessment is based on the single realisation of the perturbed and baseline climate. As a result, it has been difficult to estimate the sampling variability of the assessed impacts or to attach confidence limits to them. In the current study, Monte Carlo simulation experiments were performed to assess the sampling uncertainties in impacts on water resources of projected climate changes. The investigation employed data from basins in Iran, and catchments in Yorkshire, England, both of which incorporate water supply reservoirs. The approach enabled the population of “impacts” to be obtained, which allowed risks of having different levels of impacts to be quantified.