Predicting fires for policy making: improving accuracy of fire brigade allocation in the Brazilian Amazon

Publicado en Ecological Economics

Thiago Fonseca Morello (a), Rossano Marchetti Ramos (b), LianaO. Anderson (c), Nathan Owen (d), Thais Michele Rosan (e), LaraSteil (b)

Año de publicación 2019


Federal University of ABC, Brazil, Alameda da Universidade, S/N, Bairro Anchieta, São Bernardo do Campo/SP 09606-045, Brazil
National Centre for Prevention and Suppression of Forest Fires (PREVFOGO/IBAMA), SCEN Trecho 2, Edifício Sede, Brasília, DF, Brazil
Brazilian Centre for Monitoring and Early Warnings of Natural Disasters (Cemaden), Brazil, Estrada Doutor Altino Bondensan. Eugênio de Mello, 12247016 - São José dos Campos, SP, Brazil
Land, Environment, Economics and Policy Institute, University of Exeter Business School, Xfi Building, Rennes Drive, Exeter EX4 4PU, UK
Brazilian Institute for Space Research (INPE), Av. dos Astronautas, 1.758 - Jardim da Granja, São José dos Campos, SP, Brazil


Proyecto SGPHW-016


• Amazon fires returned to levels observed when deforestation was twofold higher.

• Efficient spatial allocation of fire brigades requires accurate fire prediction.

• Accuracy is affected by unobservable time-fixed factors and spatial clustering.

• Top fire municipalities were detected twice as accurately as in current practice.

• Brigade allocation should consider the multidimensionality of fire predictors.


The positioning of federal fire brigades in the Brazilian Amazon is based on an oversimplified prediction of fire occurrences, where inaccuracies can affect the policy's efficiency. To mitigate this issue, this paper attempts to improve fire prediction. Firstly, a panel dataset was built at municipal level from socioeconomic and environmental data. The dataset is unparalleled in both the number of variables (48) and in geographical (whole Amazon) and temporal breadth (2008 to 2014). Secondly, econometric models were estimated to predict fire occurrences with high accuracy and to infer statistically significant predictors of fire. The best predictions were achieved by accounting for observed and unobserved time-invariant predictors and also for spatial dependence. The most accurate model predicted the top 20% municipal fire counts with 76% success rate. It was over twice as accurate in identifying priority municipalities as the current fire brigade allocation procedure. Of the 47 potential predictors, deforestation, forest degradation, primary forest, GDP, indigenous and protected areas, climate and soil proved statistically significant. Conclusively, the current criteria for allocating fire brigades should be expanded to account for (i) socioeconomic and environmental predictors, (ii) time-invariant unobservables and (iii) spatial autocorrelation on fires.