Hydrology and Earth System Sciences (Nov 2022)

Influence of intensive agriculture and geological heterogeneity on the recharge of an arid aquifer system (Saq–Ram, Arabian Peninsula) inferred from GRACE data

  • P. Seraphin,
  • J. Gonçalvès,
  • B. Hamelin,
  • T. Stieglitz,
  • P. Deschamps

DOI
https://doi.org/10.5194/hess-26-5757-2022
Journal volume & issue
Vol. 26
pp. 5757 – 5771

Abstract

Read online

This study assesses the detailed water budget of the Saq–Ram Aquifer System (520 000 km2) over the 2002–2019 period using satellite gravity data from the Gravity Recovery And Climate Experiment (GRACE). The three existing GRACE solutions were tested for their local compatibility to compute groundwater storage (GWS) variations in combination with the three soil moisture datasets available from the land surface models (LSMs) of the Global Land Data Assimilation System (GLDAS). Accounting for groundwater pumping, artificial recharge, and natural discharge uniformly distributed over the Saq–Ram domain, the GRACE-derived mass balance calculation for water yields a long-term estimate of the domain-averaged natural recharge of (2.4±1.4) mm yr−1, corresponding to (4.4±2.6) % of the annual average rainfall (AAR). Beyond the regional-scale approach proposed here, spatial heterogeneities regarding the groundwater recharge were identified. The first source of heterogeneity is of anthropogenic origin: chiefly induced by irrigation excess over irrigated surfaces (about 1 % of the domain), artificial recharge corresponds to half of the total recharge of the aquifer. The second source of recharge heterogeneity identified here is natural: volcanic lava deposits (called harrats on the Arabian Peninsula) which cover 8 % of the Saq–Ram aquifer domain but contribute to more than 50 % of the natural recharge. Hence, in addition to this application on the Arabian Peninsula, this study strongly indicates a major control of geological context on arid aquifer recharge, which has been poorly discussed hitherto. Due to large lag times of the diffuse recharge mechanism, the annual analysis using this GRACE–GLDAS approach in arid domains should be limited to areas where focused recharge is the main mechanism, while long-term analysis is valid regardless of the recharge mechanism. Moreover, it appears that about 15 years of GRACE records are required to obtain a relevant long-term recharge estimate.