Synoptic weather patterns conducive to lightning-ignited wildfires in Catalonia

Abstract. Wildfires cause substantial losses to socio-economic and natural assets, especially in Mediterranean climate regions. Despite human activity being the main cause of wildfires in Mediterranean European countries, lightning-ignited wildfires should also be considered a major disruptive agent as they can trigger large fires. In addition, recent studies on the potential climate change effects on wildfires pointed out that lightning-ignited wildfires may gain relevance in Mediterranean areas in the years to come. The present study analyses the synoptical weather patterns favouring lightning-ignited wildfires in Catalonia (NE Iberian Peninsula). Being able to identify areas with an elevated lightning-ignition survival at daily timescales would be of great assistance to wildfire management agencies, i.e. locating ignitions and potential holdover fires, preparing for days with multiple ignitions or routing detection flight paths. It is worth noticing that one of the reasons that lightning-caused wildfires are difficult to manage is that they can survive for several days after the ignition, emerging days later once surface vegetation becomes dry enough to support sustained combustion. For this reason, in a first step, a reliable lightning–wildfire association is needed to properly identify the date and time of the fire starter for each wildfire. Afterwards, the circulation types on the days of ignition are analysed. The study relies on a dataset of 870 lightning-ignited wildfires, gathered by the Forest Protection Agency of the autonomous government of Catalonia between 2005 and 2020. Lightning data were provided by the Lightning Location System operated by the Meteorological Service of Catalonia. Results show that lightning-ignited wildfires in Catalonia are related to a typical synoptic weather pattern dominated by a short-wave trough at 500 hPa, with three distinct associations: an Iberian thermal low (51 % of the fires), a northern flow (24 %) and prefrontal convection (13 %).