PUBLICATIONS
Published works
A model-data fusion framework for estimating fuel properties, vegetation growth, carbon storage and the water balance at hillslope scale
Title | A model-data fusion framework for estimating fuel properties, vegetation growth, carbon storage and the water balance at hillslope scale |
Publication Type | Report |
Year of Publication | 2015 |
Authors | van Dijk, A, Yebra, M, Cary, GJ |
Date Published | 06/2015 |
Abstract | This progress report is an output from the Bushfire and National Hazards CRC, Project A1 ‘Mapping Bushfire Hazard and Impacts’. It documents the first steps in the development of a model-data fusion framework to provide estimates on historic fire impacts on landscape values, as well as potentially real-time estimates of current fuel load and flammability. Model-data fusion refers to the blending of observations with model predictions by considering the error in both. A case study area was undertaken to analyse the value of alternative airborne and remote sensing observations in a model-data fusion framework. The data were used to set up a spatial forest growth, water use and carbon uptake model at high (25 m) spatial resolution and daily time step. The model shares a common heritage with the Australian Water Resources Assessment Landscape (AWRA-L) water balance model used by the Bureau of Meteorology, and includes coupled models of water, carbon and biomass (fuel) dynamics that can be applied at high resolution. The model-data fusion framework is provisionally referred to here as the High-resolution Fire Risk and Impact (HiFRI) framework. The case study comprised the period 2000-2010 for the western ACT, including native forests, plantation forests and grasslands. Much of the area was burnt in 2003. The data integrated within HiFRI included high-resolution Landsat imagery, a digital elevation model and daily climate grids. The case study demonstrated that it is feasible to produce estimates of water and carbon balance variables. The results show strong slope effects associated with solar irradiance. Integrating satellite observations also showed the expected influence from vegetation regrowth after the fire. |