Background: Accurate simulation of wildfires can benefit pre-ignition mitigation and preparedness, and post-ignition emergency response management. Aims: We evaluated the performance of Weather Research and Forecast-Fire (WRF-Fire), a coupled fire-atmosphere wildland fire simulation platform, in simulating a large historic fire (2018 Camp Fire). Methods: A baseline model based on a setup typically used for WRF-Fire operational applications is utilised to simulate Camp Fire. Simulation results are compared to high-temporal-resolution fire perimeters derived from NEXRAD observations. The sensitivity of the model to a series of modelling parameters and assumptions governing the simulated wind field are then investigated. Results of WRF-Fire for Camp Fire are compared to FARSITE. Key results: Baseline case shows non-negligible discrepancies between the simulated fire and the observations on rate of spread (ROS) and spread direction. Sensitivity analysis results show that refining the atmospheric grid of Camp Fire’s complex terrain improves fire prediction capabilities. Conclusions: Sensitivity studies show the importance of refined atmosphere modelling for wildland fire simulation using WRF-Fire in complex terrains. Compared to FARSITE, WRF-Fire agrees better with the observations in terms of fire propagation rate and direction. Implications: The findings suggest the need for further investigation of other possible sources of wildfire modelling uncertainties and errors.

Accurate estimation of fuels is essential for wildland fire simulations as well as decision-making related to land management. Numerous research efforts have leveraged remote sensing and machine learning for classifying land cover and mapping forest vegetation species. In most cases that focused on surface fuel mapping, the spatial scale of interest was smaller than a few hundred square kilometers; thus, many small-scale site-specific models had to be created to cover the landscape at the national scale. The present work aims to develop a large-scale surface fuel identification model using a custom deep learning framework that can ingest multimodal data. Specifically, we use deep learning to extract information from multispectral signatures, high-resolution imagery, and biophysical climate and terrain data in a way that facilitates their end-to-end training on labeled data. A multi-layer neural network is used with spectral and biophysical data, and a convolutional neural network backbone is used to extract the visual features from high-resolution imagery. A Monte Carlo dropout mechanism was also devised to create a stochastic ensemble of models that can capture classification uncertainties while boosting the prediction performance. To train the system as a proof-of-concept, fuel pseudo-labels were created by a random geospatial sampling of existing fuel maps across California. Application results on independent test sets showed promising fuel identification performance with an overall accuracy ranging from 55% to 75%, depending on the level of granularity of the included fuel types. As expected, including the rare—and possibly less consequential—fuel types reduced the accuracy. On the other hand, the addition of high-resolution imagery improved classification performance at all levels.

On 30 December 2021, the Marshall Fire devastated the Boulder, Colorado region. The fire initiated in fine fuels in open space just southeast of Boulder and spread rapidly due to the strong, downslope winds that penetrated into the Boulder Foothills. Despite the increasing occurrence of wildland‐urban interface (WUI) disasters, many questions remain about how fires progress through vegetation and the built environment. To help answer these questions for the Marshall Fire, we use a coupled fire‐atmosphere model and Doppler on Wheels (DOW) observations to study the fire's progression as well as examine the physical drivers of its spread. Evaluation of the model using the DOW suggests that the model is able to capture general characteristics of the flow field; however, it does not produce as robust of a hydraulic jump as the one observed. Our results highlight limitations of the model that should be addressed for successful WUI simulations.

Following the destructive Lahaina Fire in Hawaii, our team has modeled the wind and fire spread processes to understand the drivers of this devastating event. The results are in good agreement with observations recorded during the event. Extreme winds with high variability, a fire ignition close to the community, and construction characteristics led to continued fire spread in multiple directions. Our results suggest that available modeling capabilities can provide vital information to guide decision-making and emergency response management during wildfire events.

The rising severity and frequency of wildfires in recent years in the United States have raised numerous concerns regarding the improvement in wildfire emergency response management and decision-making systems, which require operational high temporal and spatial resolution monitoring capabilities. Satellites are one of the tools that can be used for wildfire monitoring. However, none of the currently available satellite systems provide both high temporal and spatial resolution. For example, GOES-17 geostationary satellite fire products have high temporal (1–5 min) but low spatial resolution (≥2 km), and VIIRS polar orbiter satellite fire products have low temporal (~12 h) but high spatial resolution (375 m). This work aims to leverage currently available satellite data sources, such as GOES and VIIRS, along with deep learning (DL) advances to achieve an operational high-resolution, both spatially and temporarily, wildfire monitoring tool. Specifically, this study considers the problem of increasing the spatial resolution of high temporal but low spatial resolution GOES-17 data products using low temporal but high spatial resolution VIIRS data products. The main idea is using an Autoencoder DL model to learn how to map GOES-17 geostationary low spatial resolution satellite images to VIIRS polar orbiter high spatial resolution satellite images. In this context, several loss functions and DL architectures are implemented and tested to predict both the fire area and the corresponding brightness temperature. These models are trained and tested on wildfire sites from 2019 to 2021 in the western U.S. The results indicate that DL models can improve the spatial resolution of GOES-17 images, leading to images that mimic the spatial resolution of VIIRS images. Combined with GOES-17 higher temporal resolution, the DL model can provide high-resolution near-real-time wildfire monitoring capability as well as semi-continuous wildfire progression maps.

Coupled fire‐atmosphere models often struggle to simulate important fire processes like fire generated flows, deep flaming fronts, extreme updrafts, and stratospheric smoke injection during large wildfires. This study uses the coupled fire‐atmosphere model, WRF‐Fire, to examine the sensitivities of some of these phenomena to the modeled total fuel load and its consumption. Specifically, the 2020 Bear Fire and 2021 Caldor Fire in California's Sierra Nevada are simulated using three fuel loading scenarios (1X, 4X, and 8X LANDFIRE derived surface fuel), while controlling the fire rate of spread using observations. This approach helps isolate the fuel loading and consumption needed to produce fire‐generated winds and plume rise comparable to radar observations of these events. Increasing fuel loads and corresponding fire residence time in WRF‐Fire leads to deep plumes in excess of 10 km, strong vertical velocities of 40–45 m s⁻¹, and combustion fronts several kilometers in width (in the along wind direction). These results indicate that LANDFIRE‐based surface fuel loads in WRF‐Fire likely under‐represent fuel loading, having significant implications for simulating landscape‐scale wildfire processes, associated impacts on spread, and fire‐atmosphere feedbacks.