CMStatistics 2017: Start Registration
View Submission - CMStatistics
Title: Exploring fire incidence in Portugal using generalized additive models for location, scale and shape Authors:  Antonia Turkman - FCiencias.ID (Portugal) [presenting]
Ana Sa - Universidade de Lisboa (Portugal)
Abstract: Wildfires are responsible for large environmental, ecological and socio-economic impacts. Portugal has a +40-year fire atlas of burnt perimeters, allowing the assessment of the fire most prone areas. Burnt area (1975-2009), topography, climate, vegetation and population data were compiled into a 2km grid cell size overlaying mainland Portugal. Fire incidence (FI), defined as the ratio between the total burned area during the 35-year period and the area of each cell, was modelled as a function of those predictors, using generalized additive models for location, scale and shape. By exploring the drivers of fire incidence dispersion and asymmetry parameters it is expected to further understand the forces that promote regional fire incidence and thus to contribute with relevant information for fire modelers and fire management decisions.