Preceding this coming bushfire season, a team of researchers at RMIT University in collaboration with Geoscience Australia will release a new hotspot product that makes observations available every 10 minutes, day and night, customised for Australian conditions.
The new product – an algorithm called BRIGHT/AHI (Biogeographical Region and Individual Geostationary HHMMSS Threshold/Advanced Himawari Imager) – uses images from the Himawari-8 satellite from the Japanese Meteorological Agency to provide automated, near-continuous and near-real-time surveillance of potential fire activity across Australia. This research is funded as part of the Australian federal government’s SmartSat CRC, and formerly by the Bushfire and Natural Hazards CRC and Natural Hazards Research Australia. Prior to this, polar-orbiting earth observation satellites were limited to providing a new image (and potential hotspots) over Australia once or twice a day. The addition of geostationary satellites to the ‘toolbox’ brings new opportunities for how we observe fire in the landscape from space, enhancing the ability for persistent continental monitoring.
Fire detection algorithms distinguish hotspots based on the difference between a candidate pixel temperature and some form of reference value. Traditional fire detection techniques identify hotspots by comparing the candidate pixel to its neighbouring pixels, using fixed threshold values to trigger a detection. As a technique, this can be particularly problematic when fires occur in complex landscapes, or where cloud and/or smoke obscure or partially obscure pixel values. For continents as large and diverse in fire regimes as Australia, this poses additional complexity.
The BRIGHT/AHI algorithm processes earth observation data and detects fire activity in a new way. Enabled by the geostationary imager (AHI Himawari-8) that takes a new scan of Australia every ten minutes, day and night, the algorithm creates bioregion-specific control values and associated thresholds based on a pixel’s temperature history, optimised for reducing errors of omission in near-real time – specifically, the 30-day history of temperature of a region for that specific time of day or night (Engel et al., 2020, 2021, 2022). Upon receiving the satellite imagery, the algorithm takes less than 45 seconds to process the whole of Australia, leading to timely bushfire information (Engel et al. 2021). These detections have been compared to existing polar-orbiting, satellite-based fire detection products (Chatzopoulos-Vouzoglanis et al., 2022).
The availability of the new BRIGHT/AHI hotspot product to Australian fire management agencies and the public is timely, since the two polar-orbiting fire detection products (the Moderate Resolution Imaging Spectrometer or MODIS, and the Visible Infrared Imaging Radiometer Suite or VIIRS) are approaching the end of their operational lifecycles. The MODIS instruments, launched in 1999 (Terra satellite) and 2002 (Aqua satellite), are both well beyond their designed operational lifecycle of six years. Continuity of earth observations is provided by the Suomi NPP VIIRS, however this was compromised with the satellite being placed in safe mode in July of 2022 and only recently coming back online.
The BRIGHT/AHI algorithm and geostationary satellites form part of a new generation of near-real time earth observation products that are enabling new insights into fire activity, regimes, impacts and recovery for Australia in a systematic and replicable way. Recent developments and testing have led to estimates of Fire Radiative Power (FRP) to be provided alongside hotspots, with associated products of fire severity and combustion completeness also being developed by the RMIT team. The challenge of persistent surveillance and tracking of wildfire across Australia is complex. No single technology or solution can claim to provide us with robust answers to all our questions, but rather a suite of technologies and techniques is needed. BRIGHT/AHI contributes to this ongoing challenge through the provision of hotspots for Australia via the Geoscience Australia Digital Earth Australia Hotspots website (https://hotspots.dea.ga.gov.au/).
This research is funded by the Australian federal government’s SmartSat CRC. For more details please see: https://smartsatcrc.com/research-programs/real-time-fire-analytics/
Acknowledgements
The support of the Commonwealth of Australia, through the Bushfire and Natural Hazards Cooperative Research Centre and Natural Hazards Research Australia, is gratefully acknowledged for funding of the original inception of the algorithm. The Flying Hellfish team in Geoscience Australia are acknowledged for the implementation and maintenance of the algorithm and delivery of hotspots via the Geoscience Digital Earth Australia.
Written by Professor Simon Jones, A/Prof Karin Reinke, and Dr Chermelle Engel, School of Geospatial Sciences, RMIT University
Originally published in Fire Australia Magazine