Rethinking the focus on forest fires in federal wildland fire management: Landscape patterns and trends of non-forest and forest burned area

Promoting forest landscape dynamic prediction with an online collaborative strategy

Modeling and predicting forest landscape dynamics are crucial for forest management and policy making, especially under the context of climate change and increased severities of disturbances. As forest landscapes change rapidly due to a variety of anthropogenic and natural factors, accurately and efficiently predicting forest dynamics requires the collaboration and synthesis of domain knowledge and experience from geographically dispersed experts. Owing to advanced web techniques, such collaboration can now be achieved to a certain extent, for example, discussion about modeling methods, consultation for model use, and surveying for stakeholders' feedback can be conducted on the web. However, a research gap remains in terms of how to facilitate online joint actions in the core task of forest landscape modeling by overcoming the challenges from decentralized and heterogeneous data, offline model computation modes, complex simulation scenarios, and exploratory modeling processes. Therefore, we propose an online collaborative strategy to enable collaborative forest landscape dynamic prediction with four core modules, namely data preparation, forest landscape model (FLM) computation, simulation scenario configuration, and process organization. These four modules are designed to support: (1) voluntary data collection and online processing, (2) online synchronous use of FLMs, (3) collaborative simulation scenario design, altering, and execution, and (4) participatory modeling process customization and coordination. We used the LANDIS-II model as a representative FLM to demonstrate the online collaborative strategy for predicting the dynamics of forest aboveground biomass. The results showed that the online collaboration strategy effectively promoted forest landscape dynamic prediction in data preparation, scenario configuration, and task arrangement, thus supporting forest-related decision making.

Multiple social and environmental factors affect wildland fire response of full or less-than-full suppression

Wildland fire incident commanders make wildfire response decisions within an increasingly complex socio-environmental context. Threats to human safety and property, along with public pressures and agency cultures, often lead commanders to emphasize full suppression. However, commanders may use less-than-full suppression to enhance responder safety, reduce firefighting costs, and encourage beneficial effects of fire. This study asks: what management, socioeconomic, environmental, and fire behavior characteristics are associated with full suppression and the less-than-full suppression methods of point-zone protection, confinement/containment, and maintain/monitor? We analyzed incident report data from 374 wildfires in the United States northern Rocky Mountains between 2008 and 2013. Regression models showed that full suppression was most strongly associated with higher housing density and earlier dates in the calendar year, along with non-federal land jurisdiction, regional and national incident management teams, human-caused ignitions, low fire-growth potential, and greater fire size. Interviews with commanders provided decision-making context for these regression results. Future efforts to encourage less-than-full suppression should address the complex management context, in addition to the biophysical context, of fire response.

Estimation of potential wildfire behavior characteristics to assess wildfire danger in southwest China using deep learning schemes

Accurate estimation of potential wildfire behavior characteristics (PWBC) can improve wildfire danger assessment. However, wildfire behavior has been estimated by most fire spread models with immeasurable uncertainties and difficulties in large-scale applications. In this study, a PWBC estimation model (named PWBC-QR-BiLSTM) was proposed by coupling the Bi-directional Long Short-Term Memory (BiLSTM) and quantile regression (QR) methods. Multi-source data, including fuel, weather, topography, infrastructure, and landscape variables, were input into the PWBC-QR-BiLSTM model to estimate the potential rate of spread (ROS) and fire radiative power (FRP) over western Sichuan of China, and then to estimate the probability density of ROS and FRP. Daily ROS and FRP were extracted from the Global Fire Atlas and the MOD14A1/MYD14A1 product. The optimal PWBC-QR-BiLSTM model was determined using the Non-dominated Sorting Genetic Algorithm Ⅱ (NAGA-Ⅱ). Results showed that the PWBC-QR-BiLSTM performed well in estimating potential ROS and FRP with high accuracy (ROS: R2 > 0.7 and MAPE<30%, FRP: R2 > 0.8 and MAPE<25%). The modal PWBC values extracted from the estimated probability density were closer to the observed values, which can be regarded as a good indicator for wildfire danger assessment. The variable importance analysis also verified that fuel and infrastructure variables played an important role in driving wildfire behavior. This study suggests the potential of utilizing artificial intelligence to estimate PWBC and its probability density to improve the guidance on wildfire management.