Incorporating soil moisture dynamics for dead fine fuel moisture estimates

Dead fuel moisture is an important variable for wildland fire forecast and management. Accurate information on dead fuel moisture is necessary for bushfire management. Although oven-drying of field samples is the most direct and accurate way to measure dead fuel moisture, it is time-consuming and labour intensive. In addition, it only provides limited value when spatially and temporally continuous dead fuel moisture information is needed for effective wildfire management. Therefore, models ranging from empirical to physics-based have been developed to estimate dead fuel moisture. Dead fuel moisture changes in response to metrological variables, such as temperature, relative humidity, wind speed, radiation and rainfall. Soil moisture has also been previously proposed to affect this variable, but the explicit role of soil moisture in influencing dead fuel moisture has been poorly studied, or modelled. The research presented in this thesis aims to investigate the role of soil moisture in influencing dead fuel moisture. That investigation should lead to an advancement in the understanding of processes that govern dead fuel moisture dynamics and an improvement of dead fuel moisture estimates. The first part of the research analysed the effect of soil moisture on dead fuel moisture estimates by incorporating vapour flux from the soil in physics-based dead fuel moisture modelling at five Australian sites. The model used here is the first fully physics-based model (Koba model) in Australia. Results showed that the importance of soil moisture in determining dead fuel moisture estimates depends on soil moisture status, with minor influence on dead fuel moisture when soil moisture was low. The second part investigated the role of soil moisture in influencing dead fuel moisture through controlled factorial field experiments at a dry and wet site in Canberra, Australia. The results demonstrated that the effect of soil moisture on dead fuel moisture occurs through two processes: vapour flux and capillary rise from the soil. With the identified moisture movement from the soil to litter from the field experiment, the third part of the research explored soil moisture coupling in the physics-based modelling through soil vapour flux and capillary rise when estimating dead fuel moisture. The model was tested at the dry (with a shallow litter bed) and wet (with a thicker litter bed) experimental sites. Results showed that a model accounting for vapour flux and capillary rise improved dead fuel moisture estimations, especially under moist conditions. The final part of the research evaluated the prediction capability of the extended model to estimate dead fuel moisture under a wide range of environments and at a regional scale. The model was firstly applied at five Australian sites with in-situ observations, to test the robustness of the extended model under various environmental conditions, and then across the whole state of Victoria, Australia with gridded input data, to test the operational capability of the model at a regional scale. Results showed that the extended model is robust to estimate litter FMC across moist and dry environmental conditions, especially compared with the original model. The extended model can be driven by downscaled gridded weather variables under wide environmental conditions, from dry to wet, in Victoria. The application of the extended model in Victoria showed that the model has the potential to map dead fuel moisture at large spatial scales. In summary, this thesis has investigated the influence of soil moisture on dead fuel moisture dynamics and estimates. The analysis demonstrates that soil moisture can influence dead fuel moisture through two processes, vapour flux and capillary rise. The influence is limited during dry and warm periods but larger under moist conditions. Consequently, models ignoring the influence of soil moisture will tend to underestimate dead fuel moisture estimates when soil is moist