Citation

Bakke SJ, Wanders N, van der Wiel K, Tallaksen LM. Nat. Hazards Earth Syst. Sci. 2023; 23(1): 65-89.

Copyright

(Copyright © 2023, Copernicus GmbH on behalf of the European Geosciences Union)

DOI

10.5194/nhess-23-65-2023

PMID

unavailable

Abstract

Wildfires are recurrent natural hazards that affect terrestrial ecosystems, the carbon cycle, climate and society. They are typically hard to predict, as their exact location and occurrence are driven by a variety of factors. Identifying a selection of dominant controls can ultimately improve predictions and projections of wildfires in both the current and a future climate. Data-driven models are suitable for identification of dominant factors of complex and partly unknown processes and can both help improve process-based models and work as independent models. In this study, we applied a data-driven machine learning approach to identify dominant hydrometeorological factors determining fire occurrence over Fennoscandia and produced spatiotemporally resolved fire danger probability maps. A random forest learner was applied to predict fire danger probabilities over space and time, using a monthly (2001-2019) satellite-based fire occurrence dataset at a 0.25∘ spatial grid as the target variable. The final data-driven model slightly outperformed the established Canadian Forest Fire Weather Index (FWI) used for comparison. Half of the 30 potential predictors included in the study were automatically selected for the model. Shallow volumetric soil water anomaly stood out as the dominant predictor, followed by predictors related to temperature and deep volumetric soil water. Using a local fire occurrence record for Norway as target data in a separate analysis, the test set performance increased considerably. This demonstrates the potential of developing reliable data-driven models for regions with a high-quality fire occurrence record and the limitation of using satellite-based fire occurrence data in regions subject to small fires not identified by satellites. We conclude that data-driven fire danger probability models are promising, both as a tool to identify the dominant predictors and for fire danger probability mapping. The derived relationships between wildfires and the selected predictors can further be used to assess potential changes in fire danger probability under different (future) climate scenarios.

Language: en