Regional climate projections, precipitation changes and flooding: the connection between science and practice

CAMILLONI, INÉS

Sao Jose dos Santos

Workshop; IPCC Workshop of WG I on Regional Climate Projections and their Use in Impacts and Risk Analysis Studies; 2015

IPCC

Climate change projections indicate alterations in precipitation regimes in terms of intensity and frequency and an increase in temperaturewhich maylead to runoff shortages or excesses and consequent water availability changesin many regions. The impacts ofclimate change on water resources also include modifications in the frequencyand intensity of floods and droughts. Quantitative estimation of thehydrological effects of climate change particularly at basin levelis essentialfor water managers and decision makersto formulate adaptation strategies tocope with the negative impacts on hydrology.In some cases, developing projections of extremes maybe necessary to investigate whether infrastructure will still provide anadequate level of protection in future or for the design of new infrastructure.Consequently, design floods and design rainfalls shows should explicitly address climate change with adjustment factors to be applied to current designestimates and may depend on design return period and projection horizon. The increased reliability in climate outputs derivedboth from Global Climate Models (GCMs) promote many studies to quantify theimpacts of climate change on the hydrology(e.g. Saurral 2010,Montroull et al,2013, Saurral et al. 2013,Mourato et al. 2015) and hydrological extremes (e.g. Tayeet al. 2011, Dang Tri et al. 2012; Camilloni et al. 2013) of different catchments. However, climate change impact assessments on water resources requirethe consideration of the sources and relative magnitude of associated uncertaintiessuch as climate and hydrology modeling and downscaling techniques. For example,GCMs provide credible estimates of climate change at continental and largerscales but they have significant errors at smaller scales and in the simulation of some large scale features. One important source of error is the parameterization of sub-grid scale processes. Due to these model deficiencies,GCM projections may have significant biases and cannot, in general, be directly applied for impact modelling.Statistical bias correction of the systematic errors of climate models produce long-term time series with a statistical distribution close to that of the observations making them applicable as inputfor hydrology models (e.g.Piani et al. 2010,Hagemann et al. 2011, Teutschbeinand Seibert 2012, Chen et al. 2013).Both downscaling and statistical correctionrequire historical data at spatial scales appropriate for input to impactmodels and are key procedures when considering extremes. In particular, assessmentof climate change impacts on flood frequency and duration due to projectedchanges in extreme precipitation requires a methodology considering a physicallybased approach that incorporates bias corrected meteorological information derived from climate models and a hydrologic model.For the remaining uncertainties associated with projecting the future in the context of climate change, a variety of approaches to decision making can be considered: scenario analysis,classical decision analysis and robust decision-making.In particular, the robustdecision making uses a rational approach to identify conditions under which alternatives are likely to fail. Consequently, this information can then be used by water resources planners and managers to detect and design options that are less vulnerable to failure.