Surface-Air Temperature Difference Anomaly (SATDA) from Himawari-8 geostationary satellite and meteorological grids for detecting vegetation drought stress

********************************** 1. Please always use the latest version of the Zenodo repository for the latest data and codes. 2. Please check the GitHub repository https://github.com/dejuncai/SATDA-VegStress/ for future updates of the codes. ************************************ Surface-Air Temperature Difference Anomaly (SATDA) is a metric for tracking vegetation drought stress using the cumulative sub-diurnal difference between land surface temperature (Ts) from the Himawari-8 geostationary satellite and hourly air temperature (Ta) from meteorological grids. For each grid cell on each date, the cumulative sub-diurnal difference between hourly Ts and Ta was computed over the late-morning to early-afternoon window to target at the ‘mid-day depression’ in plant stomatal conductance, incorporating a more representative drought stress signal. Due to cloud-induced gaps in satellite Ts, the grid cell-wise daily cumulative Ts-Ta time series were temporally composited using a 4-week backward moving average. SATDA was then computed as the z-score standardised anomalies of the composited daily cumulative Ts-Ta with respect to the seasonal baseline estimated based on observations that fell within the 31-day (i.e., +/- 15 days) baseline window surrounding each day-of-year (DoY). SATDA showed good utility in tracking the spatio-temporal patterns of the 2017-2019 Tinderbox Drought in southeast Australia and identified the flash drought event showing rapid intensification of vegetation stress conditions at weekly timescales. SATDA also provided effective forecasts for drought-induced vegetation greenness decline in semi-arid and sub-humid regions, with the forecast correlation (r) above 0.5 at 32-day lead time. It demonstrated evident advantage in vegetation greenness forecasts over the conventional water availability-based indices (i.e., precipitation and soil moisture anomalies) in woody ecosystems like woodland, open forests and closed forests. When compared with existing satellite Ts-based indices in literature including the Temperature Condition Index (TCI) and Temperature Rise Index (TRI), SATDA showed overall better vegetation greenness forecast skill with the greatest improvement in woody vegetation. SATDA is provided as GeoTiffs with a daily temporal resolution and 0.02° spatial resolution in the World Geodetic System 1984 (WGS84) longitude-latitude geographic projection (EPSG:4326). The temporal coverage of SATDA is 28/Aug/2015 to 31/Dec/2022. The current version of SATDA covers the southeast Australia region (138°E to 153.62°E, 24°S to 39.5°S), while the all-Australia version is in preparation. To reduce the number of files, the daily SATDA rasters from the same year and month are stacked into a monthly raster stack file. For instance, the file ‘SATDA_SEAus_201901_2km.tif’ is a 31-layer stack of daily SATDA rasters from 01/Jan/2019 to 31/Jan/2019. R codes for developing SATDA are also provided on this repository. As SATDA is parsimonious with small data inputs and computationally efficient formulations, this workflow has the potential to be extended to other regions covered by Himawari satellite’s observation area (e.g., East and Southeast Asia) or to other new-generation geostationary satellites (e.g., GOES-R series over the Americas, Meteosat Second Generation series over Europe and Africa). If you implement SATDA, please let the lead author know via email ([e-mail]). For full details about the background, methodology, and results for SATDA in the early detection of vegetation drought stress, check the publication at the Remote Sensing of Environment: Cai, D., McVicar, T. R., Van Niel, T. G., Donohue, R. J., Yamamoto, Y., Stewart, S. B., Ichii, K. & Stenson, M. P. 2025. Using sub-diurnal surface-air temperature difference anomaly derived from Himawari-8 geostationary satellite and meteorological grids for early detection of vegetation drought stress: Application to Australia's 2017–2019 Tinderbox Drought. Remote Sensing of Environment, 327, 114768. https://doi.org/10.1016/j.rse.2025.114768