Data-based wildfire risk model for Mediterranean ecosystems – case study of the Concepción metropolitan area in central Chile

Abstract

Abstract. Wildfire risk is latent in Chilean metropolitan areas characterized by the strong presence of wildland–urban interfaces (WUIs). The Concepción metropolitan area (CMA) constitutes one of the most representative samples of that dynamic. The wildfire risk in the CMA was addressed by establishing a model of five categories (near zero, low, moderate, high, and very high) that represent discernible thresholds in fire occurrence, using geospatial data and satellite images describing anthropic–biophysical factors that trigger fires. Those were used to deliver a model of fire hazard using machine learning algorithms, including principal component analysis and Kohonen self-organizing maps in two experimental scenarios: only native forest and only forestry plantation. The model was validated using fire hotspots obtained from the forestry government organization. The results indicated that 12.3 % of the CMA's surface area has a high and very high risk of a forest fire, 29.4 % has a moderate risk, and 58.3 % has a low and very low risk. Lastly, the observed main drivers that have deepened this risk were discussed: first, the evident proximity between the increasing urban areas with exotic forestry plantations and, second, climate change that threatens triggering more severe and large wildfires because of human activities.

Challenges in Developing Wildfire Understanding from Wildfire Information through Spatial Planning Processes

“…In both cases, there are settlements at high risk of exposure to wildfires. Victoria's rural-urban fringe areas are among the most vulnerable to wildfires worldwide [46], and recent wildfires have increased the awareness of the fact that most populated areas of southern-central Chile are exposed to wildfire risks [47,48]. Furthermore, there is preliminary evidence of general awareness about the need to address wildfires via spatial planning and of the initiatives to change the spatial system to better integrate wildfire considerations [49,50].…”

Wildland–Urban Interface: Definition and Physical Fire Risk Mitigation Measures, a Systematic Review

“…At a landscape level, fuel discontinuity was proposed for risk mitigation with land cover fragmentation, favouring land cover with low fuel load (e.g., cultivated land) amid fire-prone vegetation (e.g., coniferous forest) [8,119,129,[135][136][137]152,154,[161][162][163][164][165][166][167][168][169]. As for landscaping measures, conservation of a target fuel load requires periodical maintenance [89,98,108,109,147,149,167].…”

“…In addition, constructions on certain topographical features (e.g., ridge tops) were discouraged as they are more exposed to heat and smoke, although those locations are preferred by residents for aesthetic qualities [5,89,92,101,102,120,175,176]. Special mention was given to the road network, considered as a source of ignitions [160,168,169,[178][179][180] but also an obstacle to fire propagation [108,159,179].…”

“…A risk mitigation measure found to be effective for slowing the spread of fire across a landscape was land cover fragmentation, for example, achieved by alternating wildland parcels with agrarian fields [154,162]. This pattern is easy to maintain if it is already present in the landscape, whereas if it requires land use change, it is unlikely to be accepted by stakeholders or by the society as a whole [168].…”

“…At the landscape level, maintenance of a fragmented land cover reduces wildfire severity in the same way as fuel discontinuity around assets. Agrarian fields, grasslands, and deciduous forests are less prone to fire than grasslands, plantations, or coniferous forests [8,119,129,136,152,154,[161][162][163][164][165][166]168,169,188]. Mixed broadleaf-conifer forests may have a lower risk of fire than pure forests [135].…”