Remotely sensed vegetation phenology drives large fire spread in northwestern Europe

Summary

Background
Increasing frequency of large fires in northwest Europe, a region under-represented in fire studies, with different ecosystem processes from those most studied, indicates the need to understand the drivers of hazardous fire behavior.
Aims
This study characterizes rate of spread variation in the region and delves into vegetation and weather drivers through remote sensing.
Methods
For 58 large fires, we analyzed phenology (using the temporal variation of satellite-measured vegetation indices) and weather (using as the Canadian Fire Weather Index System). Their relations and capability of predicting fire spread rates were assessed.
Key results
Low vegetation greenness correlated non-linearly with high rate of spread, and fires in the growing season showed a drastic reduction in spread. Low levels of weather-related danger were correlated with high rate of spread.
Conclusions
In NW Europe, the integration of phenology into fire behavior analyses helps predict spread rate. Analyzing vegetation indices variation can help estimate times when ignition could generate fast-spreading fires. Contrary to expectations, high danger related to fire weather was associated with low spread.
Implications
This study highlights the need for including timing of vegetation greenness in wildfire risk modeling and for a fire weather index systems tailored to regional conditions that relate to high-hazard fire behavior.