Assessment of Fire Danger for the Italian Power Utilities using the MERIDA HRES Reanalysis

Climate change is causing a dramatic increase in the frequency and intensity of wildfires around the world. These fires have a significant impact on the power grid, with transmission and distribution lines being often de-energized or de-activated for safety purposes and infrastructures getting damaged, which leads to power outages and, consequently, to economic losses and disruptions to the population. Fire prevention plays a pivotal role in protecting environment, communities, and infrastructures. In Italy, between 25 to 45 % of the total fire events are associated with at least one line de-activation, increasing the demand from grid operators to develop new solutions that can prevent fire danger in proximity of the overhead lines.

To support power utilities, a research activity is ongoing, aiming for the development of an effective way to assess fire danger over the Italian peninsula, and to prevent and mitigate impacts on the power system. In particular, the Fire Weather Index (FWI), developed by the Canadian Forest Fire Danger Rating System (CFFDRS), was used to assesses the potential fire spread in case of ignition using only meteorological variables, such as temperature, precipitation, relative humidity, and wind speed. Initially, this was done using the MERIDA HRES OI meteorological reanalysis dataset obtained through an ERA5 downscaling with WRF-ARW model. The dataset covers the Italian domain with a 4 km resolution and adopts the Optimal Interpolation (OI) technique on the 2-m temperature and precipitation fields reconstructed by WRF using data recorded by the observing meteorological stations of the Regional Agencies for Environmental Protection (ARPA) distributed throughout the national territory.

Daily FWI values have been calculated for the period 2005-2020, obtaining 15-years of distributed FWI maps over the Italian territory that were used to characterize the areas that are more exposed to fire danger. Results highlighted that, both for the 15-year average and for single years, the areas of the southern part of mainland Italy and of the West coast, together with the South-East coast are the ones that are more exposed to fire danger. In addition to that, daily values of FWI were validated using the burned area dataset from the European Forest Fire Information System (EFFIS), which contains the number and extent of fires, derived from both satellite observations and records kept by each European State Member. These analyses showed a very high correlation between the calculated FWI value and fire extension, associating highest values of FWI to largest burnt area and highest danger classes (very high). Finally, a case study for the October 2017 wildfire in the western part of the Italian Alps (Val di Susa) was analyzed, confirming the potential of these methodology even at the single event scale.

Once the methodology for calculating fire danger starting from weather variables was validated, it was exported to define a prototype of a forecast system using numerical weather prediction models instead of reanalyses to calculate the FWI daily values over Italy, giving information to power utilities on the most appropriate solutions to adopt to mitigate the danger.