Assessing groundwater storage in the Kairouan plain aquifer
using a 3D lithology model (Central Tunisia)
Received: 6 March 2017 / Accepted: 26 April 2018 / Published online: 22 May 2018
Saudi Society for Geosciences 2018
The aquifer of the semi-arid Kairouan plain has been exploited for decades to supply the growing irrigated agriculture and the
need of drinking water. In parallel, the major hydraulic works drastically changed the natural groundwater recharge processes.
The continuous groundwater level drop observed since the 1970s naturally raises the question of groundwater storage sustain-
ability. To date, hydrogeological studies focused on groundwater fluxes, but the total amount of groundwater stored in the aquifer
system has never been fully estimated. This is the purpose of the present paper. A complete database of all available geological,
hydrogeological and geophysical data was created to build a 3D lithology model. Then, the lithological units were combined with
the hydraulic properties to estimate the groundwater storage. Over the 700 km
of the modelled area, the estimated storage in
2013 was around 18 × 10
(equivalent to 80 times the annual consumption of 2010) with a highly variable spatial distribution.
In 45 years (1968–2013), 12% of the amount of groundwater stored in the aquifer has been depleted. According to these results,
individual farms will face strong regional disparities for their access to groundwater in the near future.
Keywords Groundwater storage
In semi-arid areas where surface water is available only over
short periods of time, groundwater usually represents a long-
term water supply, even during severe droughts when river
fluxes are drastically reduced. Basically, this ability to sustain
a water supply relies on the groundwater storage (i.e. the vol-
ume of water that can be drained from the aquifer) and can be
threatened by the groundwater exploitation: a long-term de-
clining water table level breaches the groundwater resilience
capacity (e.g. Richey et al. 2015). For thick aquifers, the threat
must be related to the total groundwater storage (Steward et al.
2009), rather than the amplitude of the decline alone, i.e. the
storage variation (Massuel and Riaux 2017), which is much
easier to assess (Healy and Cook 2002), even from space
(Wahr et al. 2006).
The groundwater storage capacity of an unconfined aquifer is
directly linked to the volume of the aquifer and its specific yield.
Many methods were developed to estimate the specific yield, but
the most common are based on the drainage measurements of
aquifer materials and pumping tests (Song and Chen 2010). In
many studies, the groundwater storage capacity is estimated
using an average of the specific yield (e.g. Wahyuni et al.
2008;Vishnuetal.2013). This approximation becomes critical
in case of heterogeneous aquifers with complex geometry and
highly contrasted hydraulic properties like alluvial deposits
(Teles et al. 2004). The variation in hydraulic properties is tied
to the lithological units and building a 3D lithology model can
improve the assessment of the groundwater storage capacity and
its spatial distribution. Interpretations of subsurface figures in
three dimensions—including lithology—have been widely used
in hydrogeological studies around the world using Geographic
Information System software (e.g. Gogu et al. 2001;Rossetal.
2005). It facilitates the integration and the analysis of large
* Hamza Jerbi
Institut National Agronomique de Tunisie (INAT), Tunis, Tunisia
Institut de Recherche pour le Développement (IRD), UMR GEAU,
Arabian Journal of Geosciences (2018) 11: 236