Hydro-climatic Projections and Computational Framework for the Maamora Aquifer: SPI, SPEI, GRDI, and De Martonne Aridity Index Datasets using CORDEX and WRF
Description
Data Acquisition: The climate data for this study was retrieved from the ESGF MetaGrid. We utilized a multi-model ensemble consisting of four downscaled and bias-adjusted regional climate models from the CORDEX framework: CNRM-CM5, EC-EARTH, IPSL-CM5A-MR, and MPI-ESM-LR. These were complemented by a regionally-configured Weather Research and Forecasting (WRF) model to capture climate data specific to Morocco. The variables extracted include monthly precipitation and minimum/maximum temperatures, which are essential for calculating climate indices with a 12-month lead time to assess long-term drought trends. Meteorological Drought Assessment: We employed three distinct indices to characterize atmospheric water deficits: Standardized Precipitation Index (SPI): Calculated by fitting annual precipitation data to a Gamma distribution to quantify standardized anomalies. Standardized Precipitation-Evapotranspiration Index (SPEI): Calculated using the Hargreaves method for Potential Evapotranspiration (PET) to integrate temperature-driven moisture demand. De Martonne Aridity Index: A classic climatological measure used to classify the shift toward regional aridification based on annual temperature and precipitation. Hydrological Drought Assessment Future aquifer fluctuations and groundwater stress were evaluated through: Groundwater Recharge Estimation: Applying established infiltration coefficients to precipitation data across three distinct aquifer zones (Coastal, Central, and Eastern). Groundwater Recharge Drought Index (GRDI): A standardized metric conceptsually similar to SPI but applied to annual recharge time series. To ensure statistical robustness, recharge data were fitted to six candidate probability distributions (Gamma, Lognormal, Weibull, Extreme Value, Rician, and Nakagami), with the Kolmogorov–Smirnov (K–S) test used to identify the best-fit model for each climate scenario.
Files
Steps to reproduce
This dataset supports the study titled "A Multi-Index Assessment of Future Meteorological and Hydrological Drought in the Maamora Aquifer, Morocco, Using CORDEX and WRF Projections." It contains processed results, calculation templates, and computational scripts for evaluating climate-driven drought under RCP4.5 and RCP8.5 scenarios. Raw CORDEX and WRF datasets, exceeding 50 GB, were retrieved from the ESGF MetaGrid and are cited in the manuscript . The repository is organized into the following categories: 1. Meteorological Drought Assessment Standardized Precipitation Index (SPI): A MATLAB script ("SPI_with_Gamma_fitting.m") automates calculations using the original Gamma distribution fitting method for multiple locations and scenarios . An Excel template is provided for manual calculation based on the simple standardization approach, which is valid if the precipitation series exhibits a Gaussian (normal) distribution . Standardized Precipitation-Evapotranspiration Index (SPEI): An R-studio script ("SPEI R-studio.txt") calculates the 12-month SPEI using the Hargreaves method for potential evapotranspiration (PET) . This method accounts for atmospheric water demand driven by temperature variability. De Martonne Aridity Index: An Excel template facilitates the calculation of regional aridity based on annual precipitation and temperature . This classification identifies shifts toward arid conditions independently of the standardized indices. Climate Visualization: A MATLAB script ("Evolutions_of_P_and_T.m") plots temporal trends for temperature and precipitation projections . 2. Hydrological Drought Assessment Groundwater Recharge: The "Recharge Volume.m" MATLAB script estimates aquifer recharge by applying zone-specific infiltration coefficients (Coastal, Central, and Eastern) to precipitation data . Groundwater Recharge Drought Index (GRDI): The "GRDI_Calculation.m" MATLAB script evaluates hydrological drought by fitting recharge data to six candidate distributions (Gamma, Lognormal, Weibull, Extreme Value, Rician, and Nakagami). It utilizes the Kolmogorov-Smirnov (KS) test to identify the best-fit model, ensuring the standardized index accurately represents the statistical behavior and skewness of the recharge series
Institutions
- Universite Mohammed V de Rabat Ecole Mohammadia d'Ingenieurs