Additional Project Selections for FireSense Technology 2023 Solicitation

Seven Projects Awarded in Total under the FireSense Technology Program

12/18/2024 – NASA’s Science Mission Directorate, NASA Headquarters, Washington, DC has selected seven proposals (three of which were previously announced in Summer 2024) under the 2023 solicitation of the FireSense Technology Program (FIRET-23), (A.59 of the Research Opportunities in Space and Earth Sciences onmibus announcement). FIRET is a technology development program managed by the Earth Science Technology Office (ESTO) that seeks new, innovative Earth system observation capabilities to predict and manage wildfires and their impacts. These projects aim to develop new tools for monitoring extreme fire events and forecasting wildfire spread.

NASA received 49 proposals in response to this NRA and selected 7 for funding. Abstracts for the seven awards, which have a total dollar value of approximately $14.4M over three years, are as follows (abstracts published here as provided by proposers):

 —-

Back to Top

—

Subcanopy UAS Development for Watershed-Scale Surface and Ladder Fuel Quantification
Jonathan Greenberg, University of Nevada, Reno

Monitoring pre- and post-fire landscapes at scales matching management action is challenged by an inability to characterize subcanopy structure, namely surface and ladder fuels. Uncertainty in subcanopy fuel conditions is limiting management actions concerning fuel treatments, active fire management, and mitigating impacts to air and water quality and ecosystem health. Transformational sensing capabilities will be key to characterizing the subcanopy, and unmanned aircraft systems (UASs) afford new scaling approaches. Yet, there has been limited use of UAS to provide subcanopy conditions. To address this, we need drones with sensor packages that can navigate through (vs over) a forest quickly and automatically return subcanopy data. To date, little effort has gone into such data collection.

We propose to develop a fleet of autonomous drones with the capability to fly through the forest subcanopy, following a pre-set flight pattern with autonomous avoidance of trees and other obstacles. These drones will be outfitted with LiDAR and a multispectral sensor package to assess three-dimensional subcanopy fuels, bulk characteristics (living vs. dead), and moisture content. Our objectives are: 1) develop and test a UAS system capable of operating in the subcanopy during active fuel treatments in topographically complex regions, 2) use data collected from the subcanopy UAS to estimate surface and ladder fuel volumes, living/dead status, and size classes. We will demonstrate our system at the Nevada UAS Test Site and then at UNR’s Whittell Forest & Wildlife Area. Whittell will be undergoing fuel treatment and field data collection during the timeline of the proposed work.

This research is critical for local, state, and federal managers needing detailed fuel data for pre-fire and active-fire management. With confirmed support from CAL FIRE and the California Air Resources Board (CARB), our approach promises to revolutionize the characterization of subcanopy fuels, providing near-real-time data at various scales and engaging stakeholders at multiple levels.

 

Back to Top

—

Hot Spot: High-Resolution Real-Time Wildfire Detection, Mapping, and Communication Relay System with Persistent Broad-Area Coverage
Jared Leidich, Urban Sky Theory Inc.

As wildfires continue to surge in intensity and frequency, threatening communities and ecosystems, there is a pressing need for advanced detection and monitoring solutions. As wildfires change, so to must the technology solutions we use to sense and fight them. At the heart of this proposal lies the central objective: the development, advancement, and deployment of a cutting-edge sensor system for the overall management of wildfires.

Envision this system as a vigilant sentinel, floating high above in the stratosphere for days or weeks at a time, casting an expansive gaze over our forests. With its sharp eyes, it can detect the faintest glimmers of a nascent fire and monitor its every move, sending real-time data back to those who can combat the fire.

Our approach is anchored by two primary technical components:
High-Resolution Imaging: Our sensor’s distinct capability to capture vast areas at a rapid rate, approximately 3,000 acres per minute at a 3.5m resolution, ensures comprehensive monitoring. With its persistent operational ability, it can stay airborne for days, even weeks, either hovering over an active fire, moving between fires, or scouting areas at high risk for potential outbreaks.

Beyond just detection and mapping, our system will be outfitted with a mesh networked communication repeater and transmitter. This addition enables firefighters on the ground, often working in regions with sparse communication, to maintain connectivity, ensuring efficient coordination during firefighting operations. This communication will also provide the ability to send shape files outlining a fires location, generated by the imager, to firefighters in the field directly, providing a cutting-edge opportunity for firefighters to use a sensor tens of kilometers above them in the stratosphere to see the fire they are fighting in near real time.

Over 6 flights have been flown with early versions of the system to date, including comprehensive imaging of the Pass Wildfire in New Mexico. Our preliminary system has already demonstrated an ability to map and relay real-time data, emphasizing its potential when fully realized.

Addressing the objectives of the solicitation, our proposed system offers a transformative approach to wildfire detection, monitoring, and communication. As the climate crisis escalates and wildfires become an even more prominent challenge, timely and accurate data, coupled with effective communication, will be critical.

Our proposed system aligns with NASA’s commitment to harnessing advanced technologies to safeguard our planet and communities. With its unique blend of rapid, high-resolution imaging and communication capabilities, this system doesn’t just offer a solution; it promises a paradigm shift in how we confront and manage the wildfire challenge. Moreover, the system’s affordability and scalability make it a feasible solution for wide-scale deployment, amplifying its impact and offering a new paradigm where the entire expanse of at-risk or on-fire areas are monitored continuously.

In sum, as NASA continues its legacy of pushing the boundaries of what technology can achieve, our proposal stands as a beacon of innovation, directly addressing a pressing global challenge and exemplifying NASA’s broader vision for a safer, better-connected world.

 

Back to Top

—

Sediment Plumes and Blooms: Using Earth Observations and Modeling to Forecast Post-Fire impacts to Reservoir Water Quality and Quantity
Mary Miller, Michigan Technological University

Our primary goal is to improve forecasting and long-term monitoring tools for watershed managers dealing with post-fire hydrological impacts on critical watersheds and reservoirs.

Within the continental US, 67 of the 100 largest cities obtain their drinking water solely from surface sources. Thousands of smaller communities with limited budgets also rely on surface water. The majority of these headwater catchments are forested, some are in rangelands and grasslands ecosystems, and are subject to wildfires with droughts and past fire management policies causing an abundance of fuels. When wildfires occur, there is a high likelihood of impaired water quality (excess nitrogen, carbon and phosphorous), high sediment loads, increased stream temperatures, and suspended ash particles that are transported to water intakes and reservoirs. Dramatic increases in post-fire runoff, erosion and sedimentation is well documented. The loss of vegetation and forest litter results in decreased evapotranspiration and surface cover. Post-fire peak flows can be as high as 300 m3 s-1 km-2 resulting in catastrophic floods. Water utilities in watersheds recently impacted by wildfire are spending millions of dollars treating water supplies and dredging post-fire sediments that reduce vital water storage capacity. The cost of replacing the water treatment plant impacted by the Hermits Peak-Calf Canyon Fire in New Mexico is projected to be 145 million dollars.

In this step 1 proposal we will leverage MTRI’s hydrology, fire, and water sensing expertise to forecast and monitor threats from wildfire to water quality and quantity in the Western US. We propose three primary objectives. (1) We will identify reservoirs and watersheds at potential risk by merging fire detections and burn scars within reservoir watersheds. This analysis will be carried out for historical and current fires. Hydrological modeling of fire effects typically occurs shortly after the fire, however there is a growing interest and need for modeling watershed recovery as well. Field studies have shown the amount of surface cover after a wildfire is a dominant control on post-fire erosion rates under a given climatic regime. We will leverage both process-based models such as the Water Erosion Prediction Project in conjunction with the NASA developed Rapid Response Erosion Database along with empirical curve number models used by Burned Area Emergency Response teams on larger watersheds. Last summer our team collaborated with CALFIRE to create an ESRI toolbox capable of rapidly creating inputs to over a dozen empirical post-fire hydrology models frequently used by Watershed Emergency Response Teams (WERT) teams in California. We are proposing to incorporate these models into an online watershed database along with easy-to-follow instructions for verifying the assumptions and identifying the best models. (2) We will leverage NASA-developed remote sensing tools for mapping sediment plumes in conjunction with monitoring vegetation recovery in order to advance capabilities for monitoring and forecasting hydrological recovery. After a fire the increased influx of sediments, nutrients and metals threaten both water quality and quantity. Low water inflows due to drought, elevated temperatures and increased nutrients elevate risks for Harmful Algal Blooms (HABs). (3) Finally, we will adapt existing algorithms for detecting algal blooms developed for the Great Lakes to these smaller watersheds so that reservoir managers have additional monitoring tools for protecting their communities from sedimentation, water quality issues and HABs.

 

Back to Top

—

Near Real-Time Updated Wildfire Risk Map Model Informed by Powerline Fault Status
Hamidreza Nazaripouya, Oklahoma State University

This project aims to enhance the pre-fire situational awareness for wildfires caused by power lines. The project represents a collaborative endeavor, uniting investigators from an academic institution and key stakeholder entities. These stakeholders encompass a fire department, the state forestry service, an electric utility, and an insurance company. Our team boasts expertise spanning diverse fields, including power system fault analysis, fire experimentation, ecology and wildfire modeling, wildland fire management, earth science and remote sensing, risk assessment, and emergency and protective services. Furthermore, the team is bolstered by a network of collaborators drawn from targeted stakeholders, featuring another electric utility company, fire departments, and state and federal agencies involved in fire management, all of whom will contribute their valuable support to the project.

Failure in electrical infrastructure has regularly been ranked among the top identified causes of wildfires, and thus, effective wildfire risk assessment will increasingly depend upon systematically understanding the triggering mechanism of wildfires caused by electrical infrastructure. This project will develop an operational approach for determining the risk of electrical wildfires and updating wildfire danger rating in near-real time, via integration of ignition source data and higher resolution of fire danger biophysical factors (fuel conditions, fire weather, vegetation, and topography), all within a geospatial framework afforded by fine-resolution earth observing data. To this end, the project will 1) evolve the wildland fire potential map to satisfy the level of details needed for risk assessment of electrical wildfire. In particular, one of the novelties of this project is to improve existing fuel models by using fine spatial and temporal resolution multispectral data from PlanetScope CubeSats. Simultaneously, this project seeks to create novel AI-based fire danger indices that has the capacity to incorporate spatial information of fuel types as well as other high-resolution fuel properties, such as vegetation biomass, greenness, etc., 2) identify the ignition probability of vegetation electric faults. Another novelty in this project is to combine experimental fire ignition tests with power-hardware-in-the-loop grid simulation to understand the ignition dynamics and propagation patterns of ignition and non-ignition electrical faults across the gird under different ignition and environmental conditions. The unique and valuable dataset generated as part of these experiments will be used to identify the ignition probability as a new data product, leveraging machine learning techniques, and finally 3) advance wildfire risk mapping through an operational field demonstration, in collaboration with project collaborators, including utility companies and forest services, contributing to the FireSense field campaign mission.

This project advances biophysical data precision via fine-resolution PlanetScope data and deep learning, aligning with NASA’s “Science 2020-2024: A Vision for Scientific Excellence” and “Decadal Survey.”, emphasizing diverse data integration for wildfire solutions. The use of advanced machine learning on the uniquely generated vegetation electric faults dataset to obtain ignition probability and development of a custom wildfire potential model will help to fulfill NASA’s Earth Science Division’s mission to enhance US wildfire prediction and management via novel observations. The field demonstration in an operational environment aims to aid first responders with rapid, precise wildfire threat intel, reducing risks and costs, aligning with airborne field campaigns and the capstone mission, and NASA’s 2022 Strategic Plan for innovation in national challenges.

 

Back to Top

—

UAS-Mounted Canopy Penetrating Radar-Tag System for Understory Fuel Sensing
Elahe Soltanaghai, University of Illinois, Urbana-Champaign

Pre-fire assessments and interventions, such as long-term fire projections, fuel treatments, and prescribed burning operations, rely heavily on accurately characterizing forest fuels across large areas. However, existing remote sensing approaches offer limited vertical resolution to characterize understory fuels obscured by the canopy layer, resulting in errors sometimes even more than 100% in predicting fire perimeter and burned areas. This proposal addresses this fundamental vertical sensing resolution challenge in forest environments by introducing an unmanned aerial system (UAS)-mounted radar system that leverages radio-frequency (RF) tags as ground references, similar in concept to NISAR calibration corner reflectors, but in a much more scalable setting. One of the core challenges with UAS in forest sensing is the ability to quickly and effectively sense, characterize, and map biomass, given the limited power budget of UASs. To address this, we minimize the power requirements of individual parts and create intelligent path planning schemes that optimize the UAS path to obtain the most relevant fuel information. The proposed technology innovations enable unprecedented sensing penetration while disaggregating understory and canopy signal echoes.

The overarching objective of this proposal is to commoditize a canopy penetrating radar system for highly reliable in-field testing and understory fuel sensing. The technology development efforts under this proposal include designing a UAS-mounted C-band polarimetric radar and tag system, as well as developing physics-informed models for generalizing RF signatures of different understory fuel types in diverse forestry sites. We will also develop a framework for integrating our sparse but accurate sensing data with landscape-scale orbital data (e.g., UAVSAR, NISAR) to deliver wall-to-wall fuel maps. This framework will facilitate the application of mixed-model statistical analysis in future stakeholder management and fire-fuel research activities.

FireTech Significance: This project dramatically improves our ability to estimate fuel moisture across large areas involved in pre-fire fuel treatments and prescribed burning operations. The fuel characteristics resulting from this project will be used as inputs for the models and projections (e.g., BEHAVE, FOFEM, FVS) used by burn managers. Finally, the proposed RF tags will provide a new data type complementary to (sub)orbital SAR data and augment existing NASA data products with accurate ground truthing. The proposed technology presents a leap in spatial measurement capability of fuel moisture, attesting to the game-changing potential for this work in fire ecology. The interdisciplinary team spans expertise in radar sensing and communication (Soltanaghai), physics-informed data analytics (Alipour), antenna and RFIC design (Rebeiz), control systems (Chowdhary), remote sensing (An), and fire ecology (Watts).

 

Back to Top

—

WAMI for Support for Wildland Fire Science, Management, and Disaster Mitigation
Christopher Stellman, Logos Technologies LLC

We propose to adapt and demonstrate a low-swap infrared wide area motion imagery (WAMI) system for support of the mission of fighting active fires. WAMI consists of low-rate, high-pixel-count video in the mid-wave infrared (MWIR) over large areas. While its applications have been primarily for defense and security domains, based on our conversations with USFS wildland fire management stakeholders in this effort, it is exceptionally well suited to fire detection and monitoring even in its current, defense-focused, design and that the capability would improve the safety and effectiveness of firefighting efforts. It provides a new form of data to emerging analysis capabilities that can increase the efficiency and capacity of fire managers. The data provided would also the wildfire modeling community.

BlackKite is a podded sensor that includes processing, storage, and communications in a compact, lightweight package. The system is controlled and utilized in flight through a digital data link, with real-time imagery and products available on demand through cell or ad hoc networks. Once on the ground, all data collected in the flight (up to 6 hours continuous) are available for download and further processing.

Features of BlackKite support of the Fire Services Mission:

WAMI systems image large areas at medium resolution and at less than full video frame rates. The combination of image resolution and frame rate is designed to serve important functions in automated processing systems like tracking vehicles or monitoring the progress of fires. In these cases, higher resolution and frame rates are not critical to the function. The resulting system allows a broad access to the area of interest by multiple users to look at images and products either from the present or from prior times in the flight.

* Data Acquisition and Early Utility Analysis. Execute e 100 hours of data acquisition time covering the hot phases of fire-fighting campaigns. We will provide direct support to operations and will use data to develop products and tools. We propose that the sensor be mounted on a USFS (or contractor) aircraft, so that it operates for the fire season to expand the data set.

* Software products and compatibility – Modify our ESRI ARC-GIS-based tools and software to be more specifically adapted to USFS needs and will adapt USFS tools and methods for product development to maximize the utility of the system. We will add a high dynamic range mode which will provide larger dynamic ranges to avoid blooming while maintaining high sensitivity.

* Quantification of Benefit — Use collected data to quantify the value of high-dynamic range imaging and temporal and spatial resolution in fire detection, tracking, and prediction.

* Machine learning – Apply machine learning techniques to the collected data, along with other associative data, to provide tactical predictions for firefighting managers.

 

Back to Top

—

Fuel-Driven Wildfire Risk Mapping Over CONUS to Guide Targeted Resources Allocation
Kiley Yeakel, Massachusetts Institute of Technology/Lincoln Laboratory

Dynamic factors such as precipitation, fuel accumulation, and live fuel moisture content all play a role in determining the risk of any given wildland area to fire potential, while evolving on differing timescales. Fuel compositions within a landscape may evolve slowly over decades or dramatically change within weeks depending on the fuel category (i.e., trees versus grasses). Likewise, LFMC can vary substantially within differing vegetation types at the same site due to plant physiological adaptions to water stress. Despite the dynamic nature of fuels, most operational wildland fire management systems are dependent on in-situ sampling or CONUS-scale maps updated at a nominal 5-year cadence, which fails to capture both the broad-scale and high temporal resolution needed for CONUS-scale fire surveillance. Recent large-scale fires have demonstrated how quickly fuel conditions can change and manifest into explosive fire conditions, emphasizing the even greater need for up-to-date fuels information. To address this operational need, we propose the development of three ML algorithms using existing remote sensing assets and ground-based measurements. The proposed project will address the pre-fire life-cycle stage by combining multiple remote sensing and ground-based measurements with machine learning (ML) to provide rapid updates (daily to monthly) at fine spatial scale (~30-meter) for three key fuels intelligence gaps in the operational landscape: (1) fuel category (i.e., grasses, mixed forests, etc.), (2) canopy structure, and (3) live fuel moisture content (LFMC). These fuels intelligence products will be produced at CONUS-scale and will serve as input to a daily extreme wildfire risk analysis based on ensemble 3D fire behavior simulations, meteorological conditions and the climatology of risk at local scale. Together, the new fuels maps and wildfire risk index will provide a comprehensive view of how fuel conditions are changing and the subsequent implications for enhanced risk, enabling better coordination amongst wildland land and fire management at regional and national scale.

All three proposed ML products will leverage various NASA, USGS and ESA remote sensing assets in conjunction with existing ground measurements to generate broad-coverage maps. The fuel category algorithm will ingest monthly accumulated observations from ESA’s Sentinel-1A C-band synthetic aperture radar (SAR) and USGS Landsat multispectral imagery to estimate fuel category. To estimate canopy structure, we will first post-process ICESat-2 measurements to produce a higher spatial-resolution and more accurate canopy height (CH) and fractional canopy coverage (CC) product. We will then train a ML algorithm to estimate on a per-pixel basis CC and CH using monthly accumulated SAR data from both Sentinel-1A (C-band) and NISAR (L-band) and Landsat imagery. Finally, we will use a time series of SAR (Sentinel-1A and NISAR), Landsat, VIIRS and SMAP measurements over the previous 90-days to estimate daily LFMC on a per pixel basis. By using both active and passive microwave imaging sensors as well as optical imagery, we will directly sense both the dry and wet mass of vegetation, and therefore enable more accurate estimation of LFMC. To translate dynamic fuel status maps into daily assessment of fire risk, we will perform ensembles of QUIC-fire, a new 3D fire behavior model, for a variety of fuel, weather and topography input conditions. 3D QUIC-fire simulations will then be postprocessed so that incident fuel and weather conditions can be mapped to an indicator of extreme fire behavior, and thus, the ML-generated fuel conditions maps can be quickly translated to fire risk at CONUS-scale. This project intends to enhance the science of mapping fuel conditions using remote sensing and ML, while also specifically addressing the needs of the operational community, with all proposed methods potentially extensible to a global, operational fire surveillance system.

 

Back to Top