Fire simulations have greatly improved the way fire agencies and others approach wildland fire management. These tools allow the prediction of how a fire will spread and behave, giving firefighters valuable information for strategic decision-making. However, even with the help of these advanced simulations, there are still limitations and uncertainties that can affect accuracy.
To address this, Technosylva led a 2019 peer-reviewed study, published in the academic journal, Ecological Modelling, demonstrating an innovative method implemented in Wildfire Analyst™ to adjust fire simulations in real-time. The method determines the adjustment factors needed for the optimal rate of spread by fuel model to minimize the arrival time error between the simulated fire and a set of control points where the arrival time of the observed (real) fire is known. The results showed a significant reduction in error and a better fit between the simulated fire growth and the actual fire spread.
”This is a unique simulation mode we have in Wildfire Analyst™ that fire agencies and others can utilize in operational environments. You compare your simulated perimeter with the real fire perimeter coming from satellite platforms, available cameras, or field information. The analysis proved that we can better predict the rate of spread when conditions like convective or erratic wildfire behavior go beyond the modeling by teaching the fire spread model with a mathematical approach to incorporate these unexpected variables.
The advancement is that, while before you needed to calculate this manually through several interactions, now you can do this process automatically. Additionally, you can analyze both error and bias of all this process, therefore improving the process, especially if you are doing it in an operational environment,
Adrián Cardil, Ph. DLead Author & Senior Fire Researcher
Insight from the Research
Fire agencies have turned to fire simulation and modeling as crucial tools to help predict how wildfires will spread and behave under various conditions. This insight aids in the development of strategies and tactics for firefighting and prevention efforts. Numerous fire simulation software tools have been developed over the years, including Farsite, Nexus, FlamMap, BehavePlus, ForeFire, and Wildfire Analyst (WFA).
While these simulations are powerful tools, their accuracy is vital for effective decision-making during wildfire incidents. However, several factors can introduce uncertainties and errors into fire simulations, including limitations and assumptions in the models and uncertainties in input data such as weather, fuel types, and topography. These uncertainties can make simulations deviate from actual fire spread, which can be problematic in real-time firefighting situations.
A Novel Method of Approach
A groundbreaking method implemented in Wildfire Analyst allows for the adjustment of fire simulations in real-time. The objective of this method is to determine adjustment factors that optimize the rate of fire spread according to the specific fuel model, minimizing the error in arrival times compared to observed control points.
The analysis reviews two case studies from Catalonia, Spain, where this method was applied. The study area’s climate is characterized by hot and dry summers with low winter precipitation, making it prone to wildfires. The selected fires, Castell d’Aro and San Llorenç Savall, were closely monitored by firefighters during suppression operations, providing valuable data for testing the adjustment method.
Research Findings
The heart of the method lies in computing Rate of Spread (ROS) adjustment factors for various fuel models involved in fire simulations. These factors are specific to each fuel model and are used to adjust the initial ROS. The adjustment process is based on a least squares approach, aiming to minimize the error between simulated fire growth and real-world observations. The method allows for real-time adjustments and requires minimal input data.
The results of the case studies demonstrate the effectiveness of this approach. Without adjustment, the initial simulations significantly underestimated the actual ROS, particularly at the head of the fires. However, after applying the adjustment factors, the simulated fire spread closely matched the observed fire’s behavior, reducing both time and area errors.
The implications of this method are significant. Firefighters and managers can now have access to more accurate and consistent fire simulations, especially during real-time wildfire situations where quick decisions are crucial. This tool can be used to reconstruct past fires, analyze fire behavior, and support operational planning and decision-making.
While the adjustment method presented in this study is a game-changer for enhancing the accuracy of fire simulations, there are several considerations to keep in mind. It primarily adjusts the fuel models included in the adjustment mode, so the presence of additional fuel models in forward-time simulations may require further adjustments. Factors such as local winds, not considered in initial simulations, can also influence ROS differently in different areas.
Next Steps from the Research
The study’s innovative method for real-time adjustment of fire simulations is a valuable addition to the firefighting toolkit. It addresses the critical need for accurate fire spread predictions in the face of increasingly frequent and severe wildfires. By reducing uncertainties and errors in simulations, it empowers firefighting agencies to make more informed decisions and better protect lives, property, and the environment in the battle against wildfires.
Learn more about how this science is put into practice.
