Spatial analysis of wildfire incidence in the USA: the role of climatic spillovers

Abstract

Wildfires constitute a serious threat for both the environment and human well-being. The US fire policy aims to tackle this problem, devoting a sizeable amount of resources and resorting extensively to fire suppression strategies. The theoretical literature has established a link between climate conditions and wildfire incidence. Using state-level data from 2002 to 2013 for the USA, this work proposes a wildfire incidence indicator and runs a generalized spatial ordered probit model in order to test the findings of the previous literature empirically. Moreover, this article investigates the extent of spatial spillovers in the climatic covariates. The results highlight a significant impact of precipitation and temperature on fire incidence and provide some evidence of the role of spatial spillovers. In particular, transitions from lower to higher wildfire incidence levels are significantly encouraged by increases in local temperature and significantly discouraged by increases in both local precipitation and lagged precipitation. The present analysis complements the recent literature, confirming the previous findings with a solid empirical investigation and offering a policyoriented picture of wildfire risks all over the USA.

Developing the Assessment and Indicators for Local Institutions in Dealing with Forest Fire Dilemmas

“…The correlation between governance and local institutions was initially described by determining the significant and insignificant variables [49]. As posited, the relationship between local institutions and the proposed variables was significant.…”

Effects of climate change on health and wellbeing: A systematic review

“…Before applying the inclusion and exclusion criteria, an attempt was made first to check the relevance of articles by screening the article title and abstract. During this stage, articles regarding climate‐related disasters (e.g., sea level rise, storms, hurricanes, floods, and wildfires), considered to have been intensified by climate change, were excluded as they are usually limited to specific regions, such as coastal communities (Bukvic et al, 2018) or woodlands (Agovino et al, 2020). Since the objective of this review is to identify the health and wellbeing outcomes among human societies, studies considering the effects of climate change on the health of other living organisms, such as plants and animals, and studies focused on the non‐natural environment (e.g., social environments in the school and workplace) were also excluded.…”

Early Warning Scheme of COVID-19 related Internet Public Opinion based on RVM-L Model

“…To determine whether machine-learning extrapolation method extracted from the reference region can estimate soil classes in the interest region, the multinomial logistic regression was extracted from the reference region and applied into the interest region to estimate soil classes [37] . And multinomial logistic regression based schemes also were adopted to improve the prediction performance in spatial analysis of wildfire incidence in the USA [38] , financial risk tolerance [39] , and hyperspectral image classification [40] . However, for events with small amounts of associated data and burst feature, it is necessary not only to increase the reliability of prediction of such combined models but also to overcome the limitations imposed by the limited data.…”