Almalki et al. (2025) Quantifying Time-Lagged Vegetation Responses to Hydroclimatic Factors in Dam-Influenced Arid Regions Using VAR Modeling and Remote Sensing

Identification

• Journal: Environmental Management
• Year: 2025
• Date: 2025-12-23
• Authors: Raid Almalki, Mehdi Khaki, Patricia Saco, José F. Rodríguez
• DOI: 10.1007/s00267-025-02328-6

Research Groups

• Department of Geography, Umm Al-Qura University, Makkah, Saudi Arabia
• School of Environmental and Life Science, University of Newcastle, Callaghan, NSW, Australia
• School of Engineering, University of Newcastle, Callaghan, NSW, Australia
• School of Civil and Environmental Engineering, University of Technology Sydney, Ultimo, NSW, Australia

Short Summary

This study quantified time-lagged vegetation responses to hydroclimatic factors in four dam-influenced arid basins in southern Saudi Arabia using VAR modeling and remote sensing. It revealed that dam construction significantly increased vegetation response lags from 2–3 months in the pre-dam period to 4–5 months post-dam, highlighting the disruption of natural hydroclimatic-vegetation coupling.

Objective

• To quantify lagged vegetation responses in downstream areas to hydroclimatic forcings (precipitation, temperature, runoff, and total water storage) in dryland regions affected by dam construction.
• To identify the peak response month when vegetation exhibits the strongest reaction.
• To forecast vegetation changes using the VAR model and assess forecast accuracy by comparing them with observed data.

Study Configuration

• Spatial Scale: Four dam-impacted basins in southern Saudi Arabia: Hali, Baish, Qanuna, and Al-Ahsabah.
• Temporal Scale: 2003–2024, divided into pre-dam (2003–2009) and post-dam (2010–2024, with specific start years for each dam) periods.

Methodology and Data

• Models used: Vector Autoregression (VAR) model, Impulse Response Function (IRF), Augmented Dickey-Fuller (ADF) test.
• Data sources:
  • Normalized Difference Vegetation Index (NDVI): MODIS Vegetation Indices (MOD13Q1) version 6.1 (250 m resolution, 16-day).
  • Precipitation: Global Precipitation Measurement (GPM) (0.1-degree resolution).
  • Daytime Air Temperature: Atmospheric Infrared Sounder (AIRS) (1-degree resolution).
  • Total Water Storage (TWS): Catchment Land Surface Model (CLSM) product (GRACEDADMCLSM025GL7D v3.0) (weekly data).
  • Runoff: Calculated using a water budget equation (ΔS = P - E - R) from GPM (precipitation), GRACE (water storage change), and FLDAS (evapotranspiration) data.
  • Evapotranspiration (E): FLDAS reanalyzed data.
  • All monthly data aggregated to a uniform 1 km spatial resolution.

Main Results

• Dam construction significantly increased the time-lag of vegetation responses to hydroclimatic factors in the studied arid regions.
• Pre-dam period (2003–2009): The peak NDVI response to hydroclimatic conditions generally occurred within 1–2 months. Average time-lags were approximately 3 months for runoff and precipitation, and 2–3 months for TWS and temperature.
• Post-dam period (2010–2024): Average time-lags increased for all factors. Runoff and precipitation response times rose to approximately 4–5 months, while temperature showed the most notable increase, extending to around 5 months. TWS experienced a similar delay.
• Before dam construction, vegetation generally showed a positive response to TWS and precipitation (except Qanuna) and a negative response to runoff (except Al-Ahsabah).
• After dam construction, vegetation generally showed a positive response to runoff (except Baish), positive sensitivity to temperature and precipitation (except Hali for precipitation), and a negative response to TWS (except Hali).
• The VAR model demonstrated good predictive performance for NDVI, with R² values ranging from 0.72 to 0.84 in the pre-dam period, and 0.63 to 0.71 in the post-dam period, indicating a slight reduction in accuracy post-dam.

Contributions

• This study is novel in quantifying how dam construction specifically modifies lagged vegetation responses to hydroclimatic drivers in arid ecosystems, addressing a critical knowledge gap.
• It provides valuable insights into how dams alter the temporal course of vegetation recovery by comparing pre- and post-dam construction periods.
• It demonstrates the effectiveness of the VAR model combined with remote sensing data for monitoring and forecasting ecosystem responses in data-poor arid regions.
• The findings offer practical guidance for optimizing dam operations, improving irrigation planning, and supporting ecosystem monitoring to mitigate environmental impacts in dam-influenced arid regions.

Funding

• Umm Al-Qura University provided encouragement for this research.

Citation

@article{Almalki2025Quantifying,
  author = {Almalki, Raid and Khaki, Mehdi and Saco, Patricia and Rodríguez, José F.},
  title = {Quantifying Time-Lagged Vegetation Responses to Hydroclimatic Factors in Dam-Influenced Arid Regions Using VAR Modeling and Remote Sensing},
  journal = {Environmental Management},
  year = {2025},
  doi = {10.1007/s00267-025-02328-6},
  url = {https://doi.org/10.1007/s00267-025-02328-6}
}