Fire weather and likelihood: Characterizing climate space for fire occurrence and extent in Puerto Rico

VAN BEUSEKOM, ASHLEY E.; HENAREH KHALYANI, AZAD; ANDRADE NÚÑEZ, MARÍA JOSÉ; GOULD, W.; QUIÑONES, MAYA; GONZÁLEZ, GRIZELLE; MONMANY, A. CAROLINA; FAIN, STEPHEN J.

CLIMATIC CHANGE

SPRINGER

Lugar: Berlin; Año: 2017

0165-0009

Assessing the relationships between weather patterns and the likelihood of fireoccurrence in the Caribbean has not been as central to climate change research as intemperate regions, due in part to the smaller extent of individual fires. However, thecumulative effect of small frequent fires can shape large landscapes, and fire-proneecosystems are abundant in the tropics. Climate change has the potential to greatlyexpand fire-prone areas to moist and wet tropical forests and grasslands that havebeen traditionally less fire-prone, and to extend and create greater variability in fireseasons. We built a machine learning random forest classifier to analyze therelationship between climatic, socioeconomic, and fire history data with fire occurrenceand extent for the years 2003-2011 in Puerto Rico, nearly 35,000 fires. Usingclassifiers based on climate measurements alone, we found the climate space is areliable associate, if not a predictor, of fire occurrence and extent in this environment.We found a strong relationship between occurrence and a change from averageweather conditions, and between extent and severity of weather conditions. Theprobability that the random forest classifiers will rank a positive example higher than anegative example is 0.8-0.89 in the classifiers for deciding if a fire occurs, and 0.64-0.69 in the classifiers for deciding if the fire is greater than 5 ha. Future climateprojections of extreme seasons indicate increased potential for fire occurrence withlarger extents.