| Popis: |
Modelling wildfire spread from data collected on various slopes and across different fuel types is an important way to understand how different parameters influence the speed at which these fires travel. Fitting this data to statistical models could be used to provide accurate fire spread predictions and can influence the way in which fire fighters may combat a wildfire. The objective of this thesis is to establish a model that describes the influence of slope on a fire's resulting rate of spread in three different fuel types. We present the analysis of data from both small scale experimental fires and test burns conducted in the field in a natural landscape. We use non-linear regression and generalized additive models to create this model, using current rate of spread models as our foundation. An analysis on the influence of slope on the rate of spread travelling downhill at the heel end of a fire is also provided. |
