Multi-Temporal Analysis of Forest Fire Probability Using Socio-Economic and Environmental Variables

Abstract

As most of the forest fires in South Korea are related to human activity, socio-economic factors are critical in estimating their probability. To estimate and analyze how human activity is influencing forest fire probability, this study considered not only environmental factors such as precipitation, elevation, topographic wetness index, and forest type, but also socio-economic factors such as population density and distance from urban area. The machine learning Maximum Entropy (Maxent) and Random Forest models were used to predict and analyze the spatial distribution of forest fire probability in South Korea. The model performance was evaluated using the receiver operating characteristic (ROC) curve method, and models’ outputs were compared based on the area under the ROC curve (AUC). In addition, a multi-temporal analysis was conducted to determine the relationships between forest fire probability and socio-economic or environmental changes from the 1980s to the 2000s. The analysis revealed that the spatial distribution was concentrated in or around cities, and the probability had a strong correlation with variables related to human activity and accessibility over the decades. The AUC values for validation were higher in the Random Forest result compared to the Maxent result throughout the decades. Our findings can be useful for developing preventive measures for forest fire risk reduction considering socio-economic development and environmental conditions.

Assessment of Agricultural Drought Considering the Hydrological Cycle and Crop Phenology in the Korean Peninsula

“…Although, for SPI, there are various types such as SPI-1, 3, 6, 12, and 24 depending on the range of application of precipitation data, in this study, we applied SPI-3, which has been shown to be related to agriculture and most widely used [30,31]. A detailed description of the SPI can be found in McKee et al [32] and in similar studies [11,12]. In the case of the SPI, which is calculated monthly, the annual average SPI-3 was calculated separately for comparison with the SADI derived from annual data.…”

“…Many drought indexes, such as the Standardized Precipitation Index (SPI), the Palmer Drought Severity Index (PDSI), and the Standardized Precipitation Evapotranspiration Index (SPEI), have been proposed and widely used in agriculture, hydrology, and forestry [9][10][11]. Although the reliability of existing meteorological drought indexes has been proven in many studies, they are not highly relevant to the effects on vegetation, such as agriculture and forests [11,12]. They are limited to considering the characteristics of a specific agriculture, forest and ecosystem because of the use of time series precipitation data or the consideration of evapotranspiration in calculating the drought index.…”

Developing an Adaptive Pathway to Mitigate Air Pollution Risk for Vulnerable Groups in South Korea

“…According to the Sustainable Development Report published by the Sustainable Development Solutions Network (SDSN), air pollution in South Korea is considered a "significant challenge" [7]. Air pollution policies should consider both social and environmental aspects in the face of intense economic development in South Korea [8].…”

Mapeamento da probabilidade de incêndio e de cicatrizes de dano como suporte ao manejo florestal

“…Altas elevações apresentam ar rarefeito e temperatura mais baixa, por isso, locais situados em menores altitudes possuem maior risco de incêndio (SOARES;BATISTA, 2007). Além do fator climático, florestas situadas em maiores elevações são menos acessadas pela atividade humana e, consequentemente, possuem menor risco de incêndio (KIM et al, 2019). A exposição do terreno está relacionada à quantidade de calor recebida pela superfície, que provocam reações em cadeia e originam condições distintas ao risco de fogo (SOARES;BATISTA, 2007).…”

“…Em ciências ambientais, observa-se crescente aumento no uso de "Inteligência Artificial" (IA) em estudos de modelagem (CHEN et al, 2008), fato também verificado na área florestal (LAGERQUIST et al, 2017) pelo uso de técnicas de Aprendizado de Máquinas (em inglês, Machine Learning -ML). Estudos recentes mostraram que a performance de algoritmos de ML foram eficientes para predição da probabilidade da ocorrência de eventos climáticos indesejáveis às florestas (HART et al, 2019;KIM et al, 2019;LEUENBERGER et al, 2018;NGUYEN et al, 2018;LE;HOANG, 2018;YU et al, 2017). Esses algoritmos são capazes de obter bons desempenhos ao relacionar fenômenos não-lineares e complexos, assim como as variáveis que influem as causas de ignição e propagação dos incêndios florestais.…”