Najmi et al. (2026) Evaluation of Precipitation-Based Drought Indices Under Future Climate Change Scenarios: Integration of PERSIANN-CCS-CDR and Dynamically Downscaled Regional Climate Models for the Tensift Watershed

Identification

• Journal: Earth Systems and Environment
• Year: 2026
• Authors: Adam Najmi, Imane El Bouazzaoui, Brahim Igmoullan, Mustapha Namous, Youssef Bammou, Mustapha Raougua, Yassine Ait Brahim
• DOI: 10.1007/s41748-025-00994-4

Short Summary

This study evaluates future meteorological drought in the Tensift watershed, Morocco, by integrating PERSIANN-CCS-CDR data with elevation-stratified bias-corrected MED-CORDEX models. Results project a statistically significant, basin-wide shift toward persistent moderate-to-severe drought conditions by the late 21st century under both RCP4.5 and RCP8.5 scenarios.

Objective

• To evaluate the evolution of meteorological drought (Standardized Precipitation Index, SPI-3 and SPI-9) in the Tensift watershed, Morocco, under RCP4.5 and RCP8.5 scenarios through 2100.
• To generate high-resolution, bias-corrected precipitation series by integrating PERSIANN-CCS-CDR observations with dynamically downscaled MED-CORDEX regional climate models using an elevation-stratified merging strategy.
• To identify statistically significant spatial and temporal trends in drought frequency, duration, and intensity across the watershed using Mann-Kendall analysis.

Study Configuration

• Spatial Scale: Tensift watershed, central Morocco ($\sim 20,000 \text{ km}^2$), analyzed at a spatial resolution of $0.04^\circ$ ($\sim 4 \text{ km}$). The analysis uses an elevation threshold of $500 \text{ m}$ to stratify model performance.
• Temporal Scale: Full simulation period (1983–2100). Historical baseline (1983–2005) and future horizons: Near-term (2015–2035), Mid-term (2045–2065), and Late-term (2075–2100).

Methodology and Data

• Models used: Dynamically downscaled MED-CORDEX Regional Climate Models (RCMs): CNRM-ALADIN63 (used for pixels $< 500 \text{ m}$) and HadREM3-GA7-05 (used for pixels $\ge 500 \text{ m}$). The CMhyd tool (Power transformation method) was used for bias correction. Drought indices calculated were the Standardized Precipitation Index (SPI) at 3-month (SPI-3) and 9-month (SPI-9) timescales. Trends were analyzed using the non-parametric Mann-Kendall (MK) test.
• Data sources: Observational reference: PERSIANN-CCS-CDR (daily precipitation, $0.04^\circ$ resolution, 1983–present). Topography: Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM). Climate Scenarios: Representative Concentration Pathway (RCP) 4.5 and 8.5.

Main Results

• Bias correction using the Power transformation method reduced the Root Mean Square Error (RMSE) of the RCM outputs by $25\%$ to $40\%$ compared to the PERSIANN-CCS-CDR reference data.
• Mean annual precipitation is projected to decrease by approximately $60\%$ under RCP4.5 and nearly $70\%$ under RCP8.5 by the late century (2075–2100) relative to the historical baseline (1983–2005).
• The watershed is projected to shift from historically near-normal conditions (mean SPI-3 $\approx +0.05$) to persistent moderate drought (mean SPI-3 $\approx -0.3$ to $-0.4$) by late century. Under RCP8.5, the entire plain area is projected to reach SPI values below $-1.2$ (severe drought) by 2100.
• Mann-Kendall trend analysis showed statistically significant negative (drying) trends (p < 0.05) across $89\%$ (RCP4.5) to $97\%$ (RCP8.5) of the watershed for both SPI-3 and SPI-9. SPI-9 Z-scores reached as low as $-17.31$ under RCP8.5, indicating extremely strong long-term drying trends.
• Seasonal analysis revealed that winter and spring (March to May), the critical periods for precipitation and agriculture, experience the largest absolute shifts toward drying, with spring SPI-3 declining from $+0.1$ (historical) to $-0.6$ (late century).

Contributions

• Provides the first comprehensive, high-resolution drought projection specifically focused on the Tensift watershed, quantifying drought evolution at multiple temporal scales and across elevation gradients.
• Implements a novel elevation-stratified model merging strategy, retaining CNRM outputs for lowlands ($< 500 \text{ m}$) and HadGEM outputs for highlands ($\ge 500 \text{ m}$), optimizing the representation of precipitation dynamics in complex topography.
• Integrates the satellite-derived PERSIANN-CCS-CDR dataset as a robust observational benchmark for bias correction in a data-sparse mountainous region.
• Documents the transformation of the regional hydroclimate from episodic historical droughts to persistent, basin-wide moderate-to-severe drought conditions, providing critical information for long-term water resource management and agricultural adaptation planning.

Funding

• None reported. The authors stated they did not receive support from any organization for the submitted work.

Citation

@article{Najmi2026Evaluation,
  author = {Najmi, Adam and Bouazzaoui, Imane El and Igmoullan, Brahim and Namous, Mustapha and Bammou, Youssef and Raougua, Mustapha and Brahim, Yassine Ait},
  title = {Evaluation of Precipitation-Based Drought Indices Under Future Climate Change Scenarios: Integration of PERSIANN-CCS-CDR and Dynamically Downscaled Regional Climate Models for the Tensift Watershed},
  journal = {Earth Systems and Environment},
  year = {2026},
  doi = {10.1007/s41748-025-00994-4},
  url = {https://doi.org/10.1007/s41748-025-00994-4}
}

Generated by BiblioAssistant using gemini-flash-latest (Google API)