EconPapers    
Economics at your fingertips  
 

Global data-driven prediction of fire activity

Francesca Di Giuseppe (), Joe McNorton (), Anna Lombardi and Fredrik Wetterhall
Additional contact information
Francesca Di Giuseppe: ECMWF, European Centre for Medium-range Weather Forecast
Joe McNorton: ECMWF, European Centre for Medium-range Weather Forecast
Anna Lombardi: ECMWF, European Centre for Medium-range Weather Forecast
Fredrik Wetterhall: ECMWF, European Centre for Medium-range Weather Forecast

Nature Communications, 2025, vol. 16, issue 1, 1-12

Abstract: Abstract Recent advancements in machine learning (ML) have expanded the potential use across scientific applications, including weather and hazard forecasting. The ability of these methods to extract information from diverse and novel data types enables the transition from forecasting fire weather, to predicting actual fire activity. In this study we demonstrate that this shift is feasible also within an operational context. Traditional methods of fire forecasts tend to over predict high fire danger, particularly in fuel limited biomes, often resulting in false alarms. By using data on fuel characteristics, ignitions and observed fire activity, data-driven predictions reduce the false-alarm rate of high-danger forecasts, enhancing their accuracy. This is made possible by high quality global datasets of fuel evolution and fire detection. We find that the quality of input data is more important when improving forecasts than the complexity of the ML architecture. While the focus on ML advancements is often justified, our findings highlight the importance of investing in high-quality data and, where necessary create it through physical models. Neglecting this aspect would undermine the potential gains from ML-based approaches, emphasizing that data quality is essential to achieve meaningful progress in fire activity forecasting.

Date: 2025
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.nature.com/articles/s41467-025-58097-7 Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:16:y:2025:i:1:d:10.1038_s41467-025-58097-7

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-025-58097-7

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().

 
Page updated 2025-05-10
Handle: RePEc:nat:natcom:v:16:y:2025:i:1:d:10.1038_s41467-025-58097-7