Protocol for ignition risk assessment

The occurrence of wildfires requires from the presence a heat source and specific fuel conditions to trigger an ignition or starting point were a heat source will enable the combustion of nearby fuel and the subsequent spread of the fire. Although a limited number of ignitions evolve into fires of significant size, the knowledge about their spatial and temporal distribution provides information for assessing potential risk of fire that can be used to allocate resources for early attack fires on the more risky areas (Carmel et al., 2009). For example, when combined with Fire Danger Rating Systems (FDRS) the results of models for predicting ignition occurrence with FDRS predictions we would include those socioeconomic factors behind human‐caused ignitions to the weather and fuel moisture components that basically define FDRS. The number and spatial distribution of ignitions can also be integrated into fire spread models to generate spatially continuous information on probability of fire occurrence for landscape planning purposes when reducing the negative impact of fires is a goal (Gonzalez‐Olabarria and Pukkala 2011; Gonzalez‐Olabarria et al., 2012a). Finally, by understanding the human behaviour that triggers human‐caused ignitions, it is possible to implement measures for reducing the number of ignitions and the subsequent fires. However, it has to be mentioned that predicting where and when they will the ignitions take place implies a high degree of uncertainty, as both natural events and human activities leading to their occurrence are often difficult to predict or even to measure.