For decades, scientists have copiously researched and ferreted out the unusual nature of the global significant climate changes. The preponderance of the climate change evidence is demonstrated from the level of agreement across diverse studies which indicate a high degree of robustness. Considering the last 1000 years, the recent warming is strikingly higher than previous warm periods. Other noticeable effects of global warming have been sea level rises, increased ice melting, increased floods and droughts among others. There is ample evidence that this widespread warming has precipitated changes in hydrology and it is anticipated that the future changes will be enhanced. Anticipated climate change impacts on hydrology may lead to both opportunities and challenges. The frequency of floods and droughts could challenge existing water management measures, agricultural practices and the biodiversity in aquatic ecosystems. Increased flows are not necessarily negative; increased volumes of water could be beneficial for hydropower generation, groundwater recharge and agriculture. However, the confidence in the future changes is still a concern today because of the uncertainties in the climate change models. But climate models remain the only scientifically credible tools for investigating the dynamics of the global climate, and for investigating diverse scenarios. Moreover, the confidence in the climate models has been increasing over the last decades as modifications have been made to exploit the new advances in climate science and computer processing. Higher resolution Regional Climate Models (RCMs) have been nested within Global Climate Models (GCMs) to improve on the representation of the local climates which are less reliable in the coarse GCMs. With the improvements in the GCMs and RCMs, the increased climate change evidence and regional policy requirements, hydrological studies related to climate change have increased. The interfacing of climate modelling with hydrological modelling is one area that has received increased attention but challenging questions still exist. Thus, the principal objective of this research was to explore climate variability and future climate change in Belgium with the aim of understanding the possible consequences on hydrological regimes. The research benefited from the state of the art high resolution climate models and the availability of long term records. The first objective involved the diagnosis of the observed trends in extremes from long term observed records. Trend studies are important because they provide extra knowledge on the possible climate change effects, and the nature of the natural climate variability in Belgium. The presence of cycles, a natural variability phenomenon, may debunk the argument of unprecedented climate change while the presence of significant trends may support it. This study provides a historical view that elucidates whether climate change in Belgium is plausible or implausible. The historical diagnosis showed some signals of anthropogenic influence. However, the evidence of the anthropogenic signals appeared to depend on the season with winter showing the clearest evidence. The major outcome of the trend study was that the change signal in time series was explained by either natural variability or the superposition of natural variability and climate change. Hence, before concluding that the recent extremes are due to anthropogenic climate change, other studies are necessary to establish whether natural variability can not explain the observed extremes. Since climate models contain the best current understanding for how the physical processes within the climate system interact, the second objective was to evaluate the physical soundness of the climate models. This was motivated by the availability of many GCMs and RCMs. The fidelity of climate models was tested against the climate models and it was concluded that the models were still biased even at the higher resolutions. Hence, the outputs of the regional climate models require further refining or downscaling before direct use in hydrological studies. Climate change impact on hydrological regimes is conventionally investigated by comparing the outputs from the historical and perturbed inputs. Hence, it is necessary to apply a downscaling methodology that ensures that the climate change perturbations are applied to the hydrological variables in a more reasonable way; for instance to account for the changes in variability. Therefore, a method was formulated which ensured that the changes in extremes were realistically represented in the future climate series. This way, the future changes in extremes such as rainfall extremes (and flows) were better represented. The other research objective sought to analyse the impacts from the array of climate models and hydrological models to determine the direction and magnitude of climate change impact on hydrological regimes such as the peak flows and low flows. It was established that future trends suggest mixed consequences. Should the high flows be too high to be buffered, flood damages may increase above historic levels. However, the high flow projections also indicate a possibility of reduced high flows, which means that the future direction of change for high flows is not certain. In contrast, the impact of the future climate change impact on the low flows, as projected from the climate models, consistently exhibited reduced flows, which does not bode well for navigation and water quality during low flow periods. In addition, it was apparent that the hydrological model structure partly explained the magnitude of the changes with some models tending to have higher /lower extremes. This finding suggests that the effect of hydrological models, especially regarding extremes, should not be overlooked. Further, there is an increasing interest and demand for climate change scenarios for many local and regional impact studies in Belgium. One of the challenges such studies face is how to select a limited but useful set of climate scenarios. Hence, another goal for this research was to develop climate scenarios that would be surrogates for the examined large set of scenarios. Indeed, the spread of future climate model simulations is set to increase in the fifth IPCC assessment report making the challenge of impact analysis more intense. The demand for tailored scenarios can not be overstated. This research devised a methodology for transforming the simulations to a few scenarios that were tailored for impact analysis. With such scenarios, it is possible to evaluate, in a tailored context, the range of impacts without simulating the entire set of scenarios. In conclusion, challenges still exist in the hydrological climate change impact investigations. Statistical downscaling, bias removal, and probabilistic scenarios are some of the areas that merit further research. But it is worthwhile to emphasize the necessity for hydrological impact analysts to interact with other experts such as climate scientists, statisticians, and policy makers for a better guidance on how to synthesize the dimensions of climate change. This will lead to more valuable studies of the hydrological changes linked to climate change. |
