Abstract
Wildfire, a natural part of many ecosystems, has also resulted in significant disasters impacting ecology and human life in Australia. This study proposes a prototype of fire propagation prediction as an extension of preceding research; this system is called “Cloud computing based bushfire prediction”, the computational performance of which is expected to be about twice that of the traditional client-server (CS) model. As the first step in the modelling approach, this prototype focuses on the prediction of fire propagation. The direction of fire is limited in regular grid approaches, such as cellular automata, due to the shape of the uniformed grid, while irregular grids are freed from this constraint. In this prototype, fire propagation is computed from a centroid regardless of grid shape to remove the above constraint. Additionally, the prototype employs existing fire indices, including the Grassland Fire Danger Index (GFDI), Forest Fire Danger Index (FFDI) and Button Grass Moorland Fire Index (BGML). A number of parameters, such as Digital Elevation Model (DEM) and forecast weather data, are prepared for use in the calculation of the indices above. The fire study area is located around Lake Mackenzie in the central north of Tasmania where a fire burnt approximately 247.11 km 2 in January 2016. The prototype produces nine different prediction results with three polygon configurations, including Delaunay Triangulation, Square and Voronoi, using three different resolutions: fine, medium and coarse. The Delaunay Triangulation, which has the greatest number of adjacent grids among three shapes of polygon, shows the shortest elapsed time for spread of fire compared to other shapes. The medium grid performs the best trade-off between cost and time among the three grain sizes of prediction polygons, and the coarse size shows the best cost-effectiveness. A staging approach where coarse size prediction is released initially, followed by a medium size one, can be a pragmatic solution for the purpose of providing timely evacuation guidance.
Highlights
This study has three aims: (1) it identifies limitations in the development of fire prediction models using cellular automata; (2) it suggests a solution to these limitations and implements an example prototype as a preliminary approach, concentrating on the functionality of fire propagation prediction in detail using geographical information system (GIS) software; and (3) it proposes an efficient strategy of evacuation guidance from wildfires
The fire was assumed to ignite at 6 am on 19th January 2016 in local Tasmanian time, the coordinate is (x = 439,700, y = 5,387,000) in GDA94 MGA zone 55 mentioned in the configuration (Table 8; section in Appendix: Results)
Because Delaunay has the greatest number of adjacent grids, averaging approximately 12.37 compared to Square (7.94) and Voronoi (5.96), each neighbor has approximately 12 chances to recalculate the tentative elapse at maximum
Summary
This study has three aims: (1) it identifies limitations in the development of fire prediction models using cellular automata; (2) it suggests a solution to these limitations and implements an example prototype as a preliminary approach, concentrating on the functionality of fire propagation prediction in detail using geographical information system (GIS) software; and (3) it proposes an efficient strategy of evacuation guidance from wildfires. Poor-quality guidance should not be released, even if the prediction is calculated rapi ly. To address this problem, various predictions with different geometries and sizes of polygons are developed in this study by employing various Fire Danger Indices (FDIs) and ingesting parameters, including the Digital Elevation Model (DEM), forecast weather data and vegetation, storing the predictions in a database. The limitations of this study and possibilities for future work are mentioned
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
