Predicting US Wildfire Possibility and Severity
Abstract
Over the last few centuries forest landscapes across America have suffered from dramatic climate change. As result, many changes in landscape conditions have resulted in an increase in wildfires across the United States. Unquestionably many of the fires were caused by humans. Using data provided by the U.S. National Interagency Fire Center, we investigate the possibility of predicting future wildfire events based on fires from within the past 20 years. Specifically, Suppression cost which is included DOI Agencies and Forest Services varies from $239,943,000 to $2,274,000,000 from 1985 to 2020 respectively [1]. Data include the size of fire, topography, terrain, and cause would be used to help predict an effective evacuation and mitigation strategy. This could become an essential tool when planning evacuation strategies and wildfire prevention. This research will focus on data visualization of all mattered factors that cause U.S wildfire. Additionally, logistic regression machine learning algorithms will also be applied using related data in order to predict the possibility and severity of wildfire. References: [1] U.S National Interagency Fire Center. (2020). Suppression Cost. Federal Firefighting Costs (Suppression only). Retrieved from https://www.nifc.gov/fire-information/statistics/suppression-costs
Predicting US Wildfire Possibility and Severity
Volstorff A
Over the last few centuries forest landscapes across America have suffered from dramatic climate change. As result, many changes in landscape conditions have resulted in an increase in wildfires across the United States. Unquestionably many of the fires were caused by humans. Using data provided by the U.S. National Interagency Fire Center, we investigate the possibility of predicting future wildfire events based on fires from within the past 20 years. Specifically, Suppression cost which is included DOI Agencies and Forest Services varies from $239,943,000 to $2,274,000,000 from 1985 to 2020 respectively [1]. Data include the size of fire, topography, terrain, and cause would be used to help predict an effective evacuation and mitigation strategy. This could become an essential tool when planning evacuation strategies and wildfire prevention. This research will focus on data visualization of all mattered factors that cause U.S wildfire. Additionally, logistic regression machine learning algorithms will also be applied using related data in order to predict the possibility and severity of wildfire. References: [1] U.S National Interagency Fire Center. (2020). Suppression Cost. Federal Firefighting Costs (Suppression only). Retrieved from https://www.nifc.gov/fire-information/statistics/suppression-costs