engleski

Wildfire risk in the Mediterranean under future climate

Regarding the projected temperature increase accompanied by the precipitation decrease, an increased wildfire risk may be expected in the Mediterranean in the forthcoming decades. The present projection of the future wildfire risk in the Mediterranean is based on using the (Canadian) Fire-Weather Index (FWI) implemented in the M&Rfi Weather Generator (WG), which is calibrated with the present-climate surface weather data and then perturbed using a set of GCM-and RCM-based climate change scenarios. The projection will be made for whole Mediterranean with a stress on several regions (Sardinia, Apulia and Croatia) and will be presented in maps showing selected statistics of FWI (e.g. probability of FWI threshold exceedance). The contribution will address several aspects of the applied methodology and resulting wildfire risk projection: (1) Validity of WG for FWI calculations. An advantage of using the WG for wildfire risk forecasting and/or assessing climate change impacts on wildfire risk emerges from the fact that this approach may be applied also for ungauged locations (by using an interpolated WG) and allows one to make a probabilistic projection (by using multiple and/or long synthetic weather series). The WG validity will be tested by comparing the FWI statistics obtained with the synthetic weather series vs observed weather series. (2) The WG will be calibrated using the surface weather data from two sources: Firstly, the main focus will be put on the three regions of interest for which the weather station data will be available. Secondly, to project the future wildfire risk for the whole Mediterranean area, the WG will be calibrated using the bias-corrected RCM outputs (simulations from the ENSEMBLES project). (3) Uncertainty in future FWI projection will be assessed based on a set of climate change scenarios (used to perturb the WG parameters) derived from available global climate models. The set of applied scenarios will represent uncertainties in climate sensitivity, emission scenario and modelling (GCM). The FWI projection will be presented using the probabilistic maps showing multi-GCM means of the future FWI statistics together with the inter-GCM variability.

Fire-Weather Index ; M&Rfi Weather Generator ; Mediterranean area

nije evidentirano