Publications

Xu, WD; Wooster, MJ; Kaneko, T; He, JP; Zhang, TR; Fisher, D (2017). Major advances in geostationary fire radiative power (FRP) retrieval over Asia and Australia stemming from use of Himarawi-8 AHI. REMOTE SENSING OF ENVIRONMENT, 193, 138-149.

Abstract
Characterising the highly variable temporal dynamics of landscape-scale fire activity is best achieved using geostationary satellites, and the Himawari-8 Advanced Himawari Imager (AHI) now provides views of Asian and Australian fires at an unprecedented 10 min temporal resolution and 2 km nadir thermal channel spatial resolution. We here develop the first processing system to identify active fires and retrieve their fire radiative power (FRP) from AHI data, based on the geostationary Fire Thermal Anomaly (FTA) algorithm and FRP retrieval method originally developed for use with Meteosat SEVIRI over Africa and Europe. This scheme detects active fires covering as little as 10(-3) to 10(-4) of an AHI pixel, and we compare performance to the same scheme applied to data from the forerunner geostationary MTSAT imager and the FengYun-2 (FY-2) Stretched Visible and Infrared Spin Scan Radiometer (S-VISSR), and also to 1 km (at nadir) polar-orbiting MODIS active fire data. We find major benefits of Himawari-8 AHI over both MTSAT and FY-2, being able to detect a substantially greater proportion of fire activity and with little impact from sensor saturation. AHI-derived FRP retrievals of detected fires show a very strong agreement and a low (3 MW) bias with respect to near-simultaneous MODIS retrievals, though fires having FRP 5 40 MW are undercounted by AHI due to its 4x larger pixel area (at nadir) than MODIS. Large parts of Asia are characterised by smaller/lower FRP fires associated with e.g. agricultural residue burning, meaning many are at or below this AHI minimum FRP detection limit, and during June 2015 AHI fails to detect around 66% of the hotspots that MODIS detects when both sensors view the same area simultaneously. However, AHI provides 144 observation opportunities per day compared to 4 typical observations from MODIS, and shows a low (8%) active fire detection error of commission. We demonstrate the unique value of the geostationary FRP retrievals made from AHI data for full fire diurnal cycle assessment and for Fire Radiative Energy (FRE) calculations. We conclude that these FRP data demonstrate major benefits for studies of active fires over Asia and Australia, and expect them to become an important component of the global geostationary active fire observation system. (C) 2017 The Author(s). Published by Elsevier Inc.

DOI:
10.1016/j.rse.2017.02.024

ISSN:
0034-4257