University of Florence, Italy
May 22, 2009
Mesa Laboratory- Damon Room
Understanding and Extending Climate Change in Hydrology
The climate change issues are becoming everyday more and more central, not only for scientists and specialists but also for a large part of the public opinion. An important branch of climate change science is covered by modifications inherent the hydrologic cycle, since floods, droughts and water resources management are thorny problems all around the world. Two paths of climate change studies are presented together with applications. The first is the analysis of long-range data series in order to ascertain whether signals of climatic change can already be detected. The second is the application of downscaling techniques to use General Circulation Models (GCMs) realizations, to forecast climate evolution in the next decades or centuries by means of simulations. In the first a spatial and local analysis of changes in precipitation regime and air temperature in the Tuscany Region (Italy) is proposed. In this framework, are discussed the principal problems in time series trend detection as the hypothesis test choice, the multiple test issue and the uncertainty arisen when pronounced stochastic behaviors are present in the data. Recent studies have, in fact, demonstrated that the stochastic behavior of a time series can sensibly change the statistical significance of a trend, especially if the time series exhibits long range dependence. In the second is presented a "proof-of-concept" framework to downscaling climate disturbances as inferred, from General Circulation Models (GCM), to use in eco-hydrologic modeling.
A stochastic downscaling technique is proposed together with a weather generator to generate future climate scenarios. The weather generator is designed to reproduce climate statistics over a wide range of temporal scales, from high-frequency extremes, to low-frequency inter-annual variability. In simulating future climate, it includes the possibility to modify precipitation and temperature, on the basis of statistical information inferred from a GCM scenario.