Identifying the density of grassland fire points with kernel density estimation based on spatial distribution characteristics

Understanding the risk of grassland fire occurrence associated with historical fire point events is critical for implementing effective management of grasslands. This may require a model to convert the fire point records into continuous spatial distribution data. Kernel density estimation (KDE) can be used to represent the spatial distribution of grassland fire occurrences and decrease the influences historical records in point format with inaccurate positions. The bandwidth is the most important parameter because it dominates the amount of variation in the estimation of KDE. In this study, the spatial distribution characteristic of the points was considered to determine the bandwidth of KDE with the Ripley’s K function method. With high, medium, and low concentration scenes of grassland fire points, kernel density surfaces were produced by using the kernel function with four bandwidth parameter selection methods. For acquiring the best maps, the estimated density surfaces were compared by mean integrated squared error methods. The results show that Ripley’s K function method is the best bandwidth selection method for mapping and analyzing the risk of grassland fire occurrence with the dependent or inaccurate point variable, considering the spatial distribution characteristics.

Application of Grassland Fire Monitoring Based on Himawari-8 Geostationary Meteorological Satellite Data

<p>Himawari-8 is the next-generation geostationary meteorological satellite, which is developed by JMA and was been launched in October,2014. As the successor to the MTSAT series,Its spatial resolution, observation frequency and position accuracy are much better than the last generation, so it has large advantage in grassland fire monitoring. In this paper, we presentthe method of fire monitoring self-adaptive threshold based on Himawari-8 data, and takean example of using Himawari-8 data to monitor dynamically the grassland fire located near the border of China in April of 2016. The monitoring results show that the fire lasted about 22 hours, the size of burned area were large than 1500 km<sup>2</sup>, the longest duration of a fire pixel was about 6 hours. Through analyzing a series fire information from successive  Himawari-8 10 minutes frequency observation,the result shows that the expanding speed of the fire is 5.4 km in the direction from west to east during some duration, which is up to the extent of fast speed fire type,. Using this method, analyzed the dynamic monitoring in the next day and other scattered fire point in different areas, which indicate that this method is universality in fire monitoring and Himawari-8 can be well used to monitor the fire dynamically changing, get the location, area and temperature of the fire, evaluate the expanding speed, estimate the trends of fire development and raise the ability of grass land fire monitoring and early warning.</p>

Effects of Live Fuel Moisture Content on Wildfire Occurrence in Fire-Prone Regions over Southwest China

Previous studies have shown that Live Fuel Moisture Content (LFMC) is a crucial driver affecting wildfire occurrence worldwide, but the effect of LFMC in driving wildfire occurrence still remains unexplored over the southwest China ecosystem, an area historically vulnerable to wildfires. To this end, we took 10-years of LFMC dynamics retrieved from Moderate Resolution Imaging Spectrometer (MODIS) reflectance product using the physical Radiative Transfer Model (RTM) and the wildfire events extracted from the MODIS Burned Area (BA) product to explore the relations between LFMC and forest/grassland fire occurrence across the subtropical highland zone (Cwa) and humid subtropical zone (Cwb) over southwest China. The statistical results of pre-fire LFMC and cumulative burned area show that distinct pre-fire LFMC critical thresholds were identified for Cwa (151.3%, 123.1%, and 51.4% for forest, and 138.1%, 72.8%, and 13.1% for grassland) and Cwb (115.0% and 54.4% for forest, and 137.5%, 69.0%, and 10.6% for grassland) zones. Below these thresholds, the fire occurrence and the burned area increased significantly. Additionally, a significant decreasing trend on LFMC dynamics was found during the days prior to two large fire events, Qiubei forest fire and Lantern Mountain grassland fire that broke during the 2009/2010 and 2015/2016 fire seasons, respectively. The minimum LFMC values reached prior to the fires (49.8% and 17.3%) were close to the lowest critical LFMC thresholds we reported for forest (51.4%) and grassland (13.1%). Further LFMC trend analysis revealed that the regional median LFMC dynamics for the 2009/2010 and 2015/2016 fire seasons were also significantly lower than the 10-year LFMC of the region. Hence, this study demonstrated that the LFMC dynamics explained wildfire occurrence in these fire-prone regions over southwest China, allowing the possibility to develop a new operational wildfire danger forecasting model over this area by considering the satellite-derived LFMC product.