Wang et al. (2025) Satellite observations reveal ecosystem resistance and resilience to short-term water stress driven by dominant vegetation along a rainfall gradient in Australia

Identification

• Journal: Remote Sensing of Environment
• Year: 2025
• Date: 2025-10-07
• Authors: Huanhuan Wang, Qiaoyun Xie, Sally Thompson, Caitlin E. Moore, David L. Miller, Erik J. Veneklaas, Richard Silberstein, Xing Li, Jingfeng Xiao, Belinda E. Medlyn, William K. Smith
• DOI: 10.1016/j.rse.2025.115046

Research Groups

• School of Engineering, The University of Western Australia
• Centre for Water and Spatial Science, The University of Western Australia
• School of Agriculture and Environment, The University of Western Australia
• School of Integrative Plant Science, Cornell University
• School of Biological Sciences, and Institute of Agriculture, The University of Western Australia
• School of Science, Edith Cowan University
• School of Geography and Planning, Sun Yat-sen University
• Earth Systems Research Center, Institute for the Study of Earth, Oceans, and Space, University of New Hampshire
• Hawkesbury Institute for the Environment, University of Western Sydney
• School of Natural Resources and the Environment, University of Arizona

Short Summary

This study quantified ecosystem resistance and resilience to short-term water stress (2000-2018) along Australia's North Australian Tropical Transect using satellite-derived GPP and flux tower data. It found that resistance and resilience patterns are primarily driven by dominant vegetation types along the rainfall gradient, with semi-arid grasslands exhibiting low resistance but high resilience, while mesic woody savannas and arid shrublands show higher resistance but lower resilience.

Objective

• To quantify ecosystem resistance and resilience to short-term water stress (<1 year) across complex ecosystems using satellite-derived GPP.
• To examine how vegetation type influences the resistance and resilience of complex ecosystems under short-term water stress.
• To determine if responses to short-term water stress vary between dry and wet seasons in highly seasonal ecosystems.

Study Configuration

• Spatial Scale: North Australian Tropical Transect (12°S to 23°S; 128°E to 138°E), spanning a 1600 mm annual rainfall gradient; 0.05° spatial resolution for satellite data.
• Temporal Scale: 2000 to 2018; 8-day temporal resolution for GOSIF GPP (interpolated to daily); daily for climate data and EC GPP; short-term water stress events (<1 year).

Methodology and Data

• Models used:
  • SOLO ANN (Self-Organising Linear Output Artificial Neural Network) for gap-filling EC data.
  • FAO Penman-Monteith equation for reference evapotranspiration (ETref).
  • Ordinary Least Squares (OLS) regression for statistical analysis and residual calculation.
  • Locally Weighted Scatterplot Smoothing (LOWESS) for showing residual patterns.
  • Kruskal-Wallis test for comparing group distributions.
• Data sources:
  • Satellite: Global Orbiting Carbon Observatory-2 Solar-Induced Fluorescence (GOSIF) Gross Primary Productivity (GPP) (0.05°, 8-day); MODIS GPP (MOD17A2H); MODIS Near-Infrared Reflectance of Vegetation (NIRv); Advanced Very High-Resolution Radiometer (AVHRR) Long-Term Data Record (LTDR) Burned Area (FireCCILT11) product (0.05°, monthly).
  • Observation: Eddy covariance (EC) flux tower data from six OzFlux sites (AU-How, AU-DaS, AU-Dry, AU-Stp, AU-TTE, AU-ASM) for GPP, evapotranspiration (ET), and potential evapotranspiration (PET).
  • Reanalysis/Climate: Morton's daily actual evapotranspiration (ET) and potential evapotranspiration (PET) from SILO Data Drill (0.05°); Mean Annual Precipitation (MAP) from Australian Government Bureau of Meteorology (0.05°).
  • Land Cover: Dynamic Land Cover Dataset Version 2 (DLCDv2.1) (0.05°, yearly from 2001-2015).

Main Results

• Satellite-derived GOSIF GPP showed strong correlation with flux tower GPP (R² values 0.58-0.73, KGE 0.67-0.79, RMSE 0.10-0.17), outperforming MODIS GPP. GOSIF GPP-based resistance and resilience metrics aligned well with EC-based metrics (RMSE 0.25 for resistance, 0.29 for resilience).
• Ecosystem resistance to water stress was lowest in semi-arid regions (0.78 ± 0.15) but higher in both arid (0.81 ± 0.14) and mesic regions (0.82 ± 0.13).
• Ecosystem resilience showed the opposite pattern, with highest values in semi-arid regions (0.38 ± 0.24) and lower values in mesic (0.26 ± 0.19) and arid regions (0.27 ± 0.22).
• These spatial patterns were consistent regardless of seasonality and were primarily associated with dominant vegetation type:
  • Woody savanna-dominated mesic regions: highest resistance (0.82 ± 0.13) and lowest resilience (0.26 ± 0.19).
  • Shrublands in arid areas: intermediate resistance (0.81 ± 0.14) and resilience (0.27 ± 0.22).
  • Grasslands in semi-arid regions: low resistance (0.78 ± 0.15) and high resilience (0.38 ± 0.24).
• The likelihood of full recovery (exceeding baseline after one year) was highest in mesic regions during the wet season (75.0%), compared to the dry season (56.4%). Arid regions showed a lower likelihood of full recovery (57.0%).

Contributions

• Developed and validated a remote sensing framework using satellite-derived GOSIF GPP to quantify ecosystem resistance and resilience to short-term water stress across large, complex, and transitioning ecosystems.
• Demonstrated that ecosystem resistance and resilience patterns along a significant rainfall gradient are primarily driven by dominant vegetation functional types (woody savanna, shrubland, grassland), rather than linear relationships with precipitation, highlighting the importance of species-specific drought adaptations.
• Provided insights into seasonal variations in recovery likelihood, showing that mesic ecosystems have a higher probability of full recovery during the wet season, likely due to energy limitation rather than water stress.
• Offered a robust framework for analyzing the effects of short-term water stress across diverse environmental contexts, applicable to other regions and leveraging new high-resolution SIF data.

Funding

• University of Western Australia Research Collaboration Award project “Piloting a novel geospatial approach to assess climate change impacts on grasslands”.
• National Science Foundation (Macrosystem Biology and NEON-Enabled Science program: DEB-2017870) for J. X.
• National Aeronautics and Space Administration (Carbon Cycle Science program: 80NSSC23K0109) for W. K. S.
• Terrestrial Ecosystem Research Network (TERN) infrastructure (Australian Government’s National Collaborative Research Infrastructure Strategy, NCRIS).

Citation

@article{Wang2025Satellite,
  author = {Wang, Huanhuan and Xie, Qiaoyun and Thompson, Sally and Moore, Caitlin E. and Miller, David L. and Veneklaas, Erik J. and Silberstein, Richard and Li, Xing and Xiao, Jingfeng and Medlyn, Belinda E. and Smith, William K.},
  title = {Satellite observations reveal ecosystem resistance and resilience to short-term water stress driven by dominant vegetation along a rainfall gradient in Australia},
  journal = {Remote Sensing of Environment},
  year = {2025},
  doi = {10.1016/j.rse.2025.115046},
  url = {https://doi.org/10.1016/j.rse.2025.115046}
}