[A-CG36] Terrestrial monitoring using new-generation geostationary satellites

[ACG36-01] Improving the understanding of vegetation dynamics using Advanced Himawari Imager in Australian semi-arid ecosystems

★Invited Papers

*Ankur Srivastava¹, Mauro E. Holzman^1,2, Paras Sidiqui¹, Yi Qin³, Jamie Cleverly⁴, Alfredo Huete¹ (1.Faculty of Science, University of Technology Sydney, Sydney, NSW 2007, Australia, 2.Consejo Nacional de Investigaciones Científicas y Técnicas, Instituto de Hidrología de Llanuras “Dr. Eduardo Jorge Usunoff” (IHLLA), Rep. Italia 780, Azul B7300, Argentina, 3.CSIRO, Oceans and Atmospheres, Canberra , 2601, Australia, 4.College of Science & Engineering,, James Cook University, Cairns, QLD, 4878, Australia)

Keywords:Himawari, BRDF, vegetation indices, GPP, semi-arid ecosystems

Arid and semiarid landscapes are characterized by high rainfall and soil moisture variations which in turn drive vegetation response and distribution. The Advance Himawari Imager (AHI) onboard the Japanese geostationary satellites Himawari-8 and 9, allows for high frequency monitoring of landscapes at 10-minute intervals. They have not yet been tested for their ability to characterise the rapid spatiotemporal responses of arid and semiarid vegetation to wetting and drying events. We analysed and evaluated daily AHI data from a flux tower sites in semi-arid regions of Australia to better understand the temporal response of vegetation to wet pulses and their influence on productivity.
To better understand the temporal response of vegetation to rainfall pulses, we utilised Bidirectional Reflectance Distribution Function (BRDF)- corrected surface reflectances, derived as part of a multi-institution Terrestrial Ecosystem Research Network (TERN) project, in six visible, near-infrared (NIR) and shortwave infrared bands (SWIR). Further, vegetation indices (NDVI, EVI2) and a plant moisture index (NDWI – combination of NIR and SWIR bands) were computed at nadir view and fixed solar zenith angle of 450. We then compared our AHI prototype vegetation products with in-situ daily tower flux measures of gross primary productivity (GPP), evapotranspiration, water use efficiency, and soil moisture.
Our research revealed that the seasonal dynamics of vegetation indices are consistent with the plant responses to wet periods. The AHI data are shown to be useful to monitor vegetation productivity response to wet pulses at daily scale, considering the strong relationship of OzFlux tower GPP measurements among vegetation indices. The AHI results further highlighted the limitations of the current low earth orbiting satellites in capturing these high productivity periods by exhibiting phenology driven hysteresis pattern and more well-characterized vegetation greenness lags with flux tower GPP. These results will contribute to a better understanding of the vegetation dynamics of semiarid ecosystems in Australia and enable accurate monitoring of both dynamic wet and dry vegetation patterns.