Development of Satellite Vegetation Indices to assess Grassland Curing across Australia and New Zealand
Grasslands cover 75% of Australia and constitute 70% of vegetation cover in New Zealand. Depending on a grass’s growth stage, certain curing characteristics (such as water content and degree of senescence) determine the vulnerability of grass to ignite / propagate a fire. A Bushfire CRC project aims to investigate both remote sensing and grass growth models as approaches to improve the assessment and prediction of grassland curing across Australia and New Zealand. This paper presents an overview of methods aimed at identifying and developing vegetation indices from AVHRR and MODIS for the assessment and prediction of grassland curing. NOAA AVHRR, which has not been utilised outside of Victoria, has been used operationally to produce a satellite curing index since the 1980s across southeastern Australia, and this will extend across the country on the basis of extensive field data collected by CRC. The EOS MODIS system can monitor vegetation more accurately than AVHRR, and has the potential to provide enhanced curing estimates due to its greater number of spectral bands. The research questions for this project include: How may satellite measurements in several spectral bands be best combined to quantitatively estimate the degree of grassland curing? How do factors such as grassland type, soil type, topography and amount of tree cover influence the curing algorithm? What is the accuracy of the satellite curing estimates? Do very cloudy regions/seasons require a modification to the algorithm to accommodate less frequent satellite views? Approaches from literature, and an understanding of their limitations, will guide the development of a satellite – based method for curing assessment that is robust and validated across Australia and New Zealand. Methods will vary regionally due to climatic and topographic variability, and variability in grassland type.