Event Response

Risk Reduction Category

Technology Description

Seasonal wildfire poses a substantial risk to private citizens who may be forced to evacuate as fires encroach on residential areas. Exacerbating the issue is continued housing construction near fire-prone areas. For first responders, the potential for fast-moving burns demands situational awareness in near-real-time. The objective is not only timely response, but prioritized response.

Multiple sensors located in potential wildfire zones may provide any or all of this information which may be organized on a geospatial and interactive map similar to that found at the National Interagency Fire Center map as illustrated below [1]. A number of layers with corresponding information may be seen at left.

Figure source [1]

Because the speed and direction of fire spread are mainly driven by fuel and by weather, responders need geographically and temporally resolute data. There is currently no single source of truth for fuel density and what fuel data exists can be either outdated or spatially sparse. Similarly, weather may be obtained from the National Weather Service, but more timely updates and more weather stations reporting are helpful for predicting fire spread.

Other layers of intelligence that are needed for timely and prioritized response may include factors that affect access, such as road congestion, smoke (for aerial support), downed powerlines, and other hazards. Additional intelligence may include automatic warnings of wildfire proximity to critical infrastructure and a priority marker for infrastructure feeding critical services such as hospitals, food supply, etc.

Technology goals for improving the future state of event response would include:

Single source of truth for fuel condition
A software platform to ingest data from multiple disparate data sources
Visualizations that improve decision making

Technical Readiness (Commercial Availability)

Fire Spread Modeling

Pacific Northwest National Lab (PNNL) is studying a different approach to identifying the potential path of wildfires came from atmospheric scientists: two new models employ twenty-eight “wildfire predictors” to project current wildfire behavior. Used with climate change modelled estimates, these two models may project future wildfire behavior. Several variables such as atmospheric moisture levels, vegetation dryness, density of nearby population and others may better determine wildfire likelihood, the extent of the burn, and the amount of smoke sent into the atmosphere. [2]

Data-driven Wildfire Analytics

Sandia National Labs (SNL) is modeling wildfires using accurate and current characterization of vegetation fuel along with existing work from the Resilient Energy Systems-funded Lab Directed Research and Development (LDRD). Use weather station data and satellite imagery to generate machine learning (ML)- derived characterization of vegetation. Thus, utilities may better assess, plan, and adapt to wildfires. Outputs from the model include burn probability, energy release component, and wildfire behavior. Sandia proposes to run simulations with active fire perimiters as inputs. [3]

In a separate effort SNL is working on near-real-time determination of wildfire risk with respect to critical infrastructure including wildfire impacts leading to cascading failure. Determine near-real-time fuel moisture by applying machine learning to weather station data. Identify component damage by using wildfire-spread software and Sandia grid modeling and interactive map. [3]

Sandia is also studying how visualizations impact decision making. Various representations of uncertainty may result in differing decision patterns—regarding when or if homes should be evacuated, for instance, or when a Public Safety Power Shutoff (PSPS) may be necessary. Sandia hopes to support optimal decision making regarding grid operations through better understanding of how those decisions may be affected by visualization. [3]

Implementations / Deployments

Innovations as of Mid 2023

Potential Enrichment Work Opportunity

References

[1] https://maps.wildfire.gov/sa/#/%3F/%3F/38.7471/-87.3725/6

[2] https://www.pnnl.gov/news-media/taming-tomorrows-wildfires

[3] https://energy.sandia.gov/programs/electric-grid/wildfire-electric-grid-resilience/