2017 ESA Annual Meeting (August 6 -- 11)

COS 161-4 - Building a more comprehensive natural hazards dataset: Incorporating information mined from the US National Incident Command System (ICS) into the analysis of large fires and natural hazard events

Thursday, August 10, 2017: 2:30 PM
B116, Oregon Convention Center
Lise St Denis, Earth Lab, University of Colorado, Boulder, CO, Jennifer K. Balch, Earth Lab, University of Colorado Boulder, Boulder, CO, Nathan Mietkiewicz, Geography, University of Colorado, Boulder, CO and Mollie Buckland, Geography Earth Lab, University of Colorado, Boulder, CO
Background/Question/Methods

As the societal impacts of large-scale catastrophic fires has risen in recent years, the inclusion of both physical and societal variables in the analysis of these events is increasingly important. To this end, we present a dataset that links existing fire datasets with an expanded dataset mined from the United States National Incident Management System (NIMS) Incident Command System (ICS) Historical ICS-209 Reports (fam.nwcg.gov). Only about one percent of all wildland fires require the management and oversight of an incident management team, but these fires account for approximately 85% of fire suppression costs (Williams et al, 2005). In addition, this dataset comprises virtually all wildland fires where significant social disruption occurs.

The ICS-209 Situation Report, used by all levels of incident response, captures the best available information for each operational period during an incident. These situation reports provide valuable information about evolving fire characteristics, weather conditions, accruing costs, damages, fatalities, details about response resources, levels of social disruption, values at risk, and important contributing factors throughout the active response phase. We are interested in using this information to explore the inter-relationships between important physical and socio-economic measures related to large-scale events.

Results/Conclusions

The ICS-209 dataset contains profiles for 22,588 all-hazards incidents and related situation reports from 2002-2013. Over 98% of the incidents in this dataset are wildland fires and prescribed burns. Although the fires within this dataset comprise only 1.6% of the total fires in the United States, the estimated suppression costs account for 85% of the total annual suppression costs reported by the National Interagency Fire Center (NIFC) and 57% of the area burned (196,761 km2) over this same time period.

Highlights from multiple ongoing research efforts illustrate some of the potential uses for this newly mined information. In the first we look at how geospatial information, suppression costs and ignition source provided insight into where fire suppressions costs are dominated by human-caused fires versus lightning ignitions. A second looks at the analysis of physical, emergency response, and socio-economic variables related to mega-fires for insight into how these extreme events differ from the larger population of fires. Careful curation of this data offers new research opportunities and preliminary use of this data within Earth Lab, a new earth science synthesis center at CU Boulder, illustrates the potential value of this new dataset for evaluating complex relationships between physical and societal variables.