Midterm Fire Danger Prediction Using Satellite Imagery and Auxiliary Thematic Layers

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.

The Daily Fire Hazard Index: A Fire Danger Rating Method for Mediterranean Areas

Mediterranean forests are gravely affected by wildfires, and despite the increased prevention effort of competent authorities in the past few decades, the yearly number of fires and the consequent damage has not decreased significantly. To this end, a number of dynamical methods have been developed in order to produce short-term hazard indices, such as the Fire Probability Index and the Fire Weather Index. The possibility to estimate the fire hazard is based on the observation that there is a relationship between the characteristics of the vegetation (i.e., the fuel), in terms of abundance and moisture content, and the probability of fire insurgence. The density, type, and moisture content of the vegetation are modeled using custom fuel maps, developed using the latest Corine Land Cover, and using a number of indices such as the NDVI (Normalized Difference Vegetation Index), Global Vegetation Moisture Index (GVMI), and the evapotranspiration, derived from daily satellite imagery. This paper shows how the algorithm for the calculation of the Fire Potential Index (FPI) was improved by taking into account the effect of wind speed, topography, and local solar illumination through a simple temperature correction, preserving the straightforward structure of the FPI algorithm. The results were validated on the Italian region of Sardinia using official wildfire records provided by the regional administration.