Assessing Lightning and Wildfire Hazard by Land Properties and Cloud to Ground Lightning Data with Association Rule Mining in Alberta, Canada

Monitoring Lightning Location Based on Deep Learning Combined with Multisource Spatial Data

Lightning is an important cause of casualties, and of the interruption of power supply and distribution facilities. Monitoring lightning locations is essential in disaster prevention and mitigation. Although there are many ways to obtain lightning information, there are still substantial problems in intelligent lightning monitoring. Deep learning combined with weather radar data and land attribute data can lay the foundation for future monitoring of lightning locations. Therefore, based on the residual network, the Lightning Monitoring Residual Network (LM-ResNet) is proposed in this paper to monitor lightning location. Furthermore, comparisons with GoogLeNet and DenseNet were also conducted to evaluate the proposed model. The results show that the LM-ResNet model has significant potential in monitoring lightning locations. In this study, we converted the lightning monitoring problem into a binary classification problem and then obtained weather radar product data (including the plan position indicator (PPI), composite reflectance (CR), echo top (ET), vertical integral liquid water (VIL), and average radial velocity (V)) and land attribute data (including aspect, slope, land use, and NDVI) to establish a lightning feature dataset. During model training, the focal loss function was adopted as a loss function to address the constructed imbalanced lightning feature dataset. Moreover, we conducted stepwise sensitivity analysis and single factor sensitivity analysis. The results of stepwise sensitivity analysis show that the best performance can be achieved using all the data, followed by the combination of PPI, CR, ET, and VIL. The single factor sensitivity analysis results show that the ET radar product data are very important for the monitoring of lightning locations, and the NDVI land attribute data also make significant contributions.

Understand Daily Fire Suppression Resource Ordering and Assignment Patterns by Unsupervised Learning

Wildland fire management agencies are responsible for assigning suppression resources to control fire spread and mitigate fire risks. This study implements a principle component analysis and an association rule analysis to study wildland fire response resource requests from 2016 to 2018 in the western US to identify daily resource ordering and assignment patterns for large fire incidents. Unsupervised learning can identify patterns in the assignment of individual resources or pairs of resources. Three national Geographic Area Coordination Centers (GACCs) are studied, including California (CA), Rocky Mountain (RMC), and Southwest (SWC) at both high and low suppression preparedness levels (PLs). Substantial differences are found in resource ordering and assignment between GACCs. For example, in comparison with RMC and SWC, CA generally orders and dispatches more resources to a fire per day; CA also likely orders and assigns multiple resource types in combination. Resources are more likely assigned to a fire at higher PLs in all GACCs. This study also suggests several future research directions including studying the causal relations behind different resource ordering and assignment patterns in different regions.

Applicability of Remote Sensing-Based Vegetation Water Content in Modeling Lightning-Caused Forest Fire Occurrences

In this study, our aim was to model forest fire occurrences caused by lightning using the variable of vegetation water content over six fire-dominant forested natural subregions in Northern Alberta, Canada. We used eight-day composites of surface reflectance data at 500-m spatial resolution, along with historical lightning-caused fire occurrences during the 2005–2016 period, derived from a Moderate Resolution Imaging Spectroradiometer. First, we calculated the normalized difference water index (NDWI) as an indicator of vegetation/fuel water content over the six natural subregions of interest. Then, we generated the subregion-specific annual dynamic median NDWI during the 2005–2012 period, which was assembled into a distinct pattern every year. We plotted the historical lightning-caused fires onto the generated patterns, and used the concept of cumulative frequency to model lightning-caused fire occurrences. Then, we applied this concept to model the cumulative frequencies of lightning-caused fires using the median NDWI values in each natural subregion. By finding the best subregion-specific function (i.e., R2 values over 0.98 for each subregion), we evaluated their performance using an independent subregion-specific lightning-caused fire dataset acquired during the 2013–2016 period. Our analyses revealed strong relationships (i.e., R2 values in the range of 0.92 to 0.98) between the observed and modeled cumulative frequencies of lightning-caused fires at the natural subregion level throughout the validation years. Finally, our results demonstrate the applicability of the proposed method in modeling lightning-caused fire occurrences over forested regions.