Innovative methodological approach for the modeling of seawater intrusion in the coastal plain of Muravera (Sardinia, Italy)

Many coastal areas, especially low-lying deltaic areas, have high density populations and are often important economic areas. At the same time, their aquifers are often overexploited and subject to seawater intrusion phenomena. Climate change, including sea-level rise, will exacerbate the vulnerability of these sensitive areas, so it is critical to provide the complex water resource management with tools capable to consider all the variables involved to predict appropriate scenarios for action, balancing water demand of the population and the natural water equilibrium. Hydrogeological models are as such powerful instruments though they need to be calibrated with enough reliable hydrogeological data. This study is applying an innovative hydrogeological modelling approach to simulate seawater intrusion in the coastal plain of Muravera (south-east, Sardinia). This area has been studied since the 1950s due to seawater intrusion phenomena that have led to various socio-economic and environmental problems. Thanks to the geological, hydrogeological, and geochemical data collected in the recent years, it has been possible to develop a 3D numerical groundwater model of the plain. To simulate density dependent groundwater flow and coupled solute transport, by Deltares developed, iMOD-WQ software is used. iMOD-WQ consists of the widely used software SEAWAT, which enables density dependent fresh-salt groundwater modelling (Langevin & Guo, 2006), and the software RT3D, for reactive transport modelling, which in not applied in this study. iMOD is a source code adapted for parallellization and thus speeding up the computation up to two order of magnitudes depending on the available computer codes (Verkaik et al., 2021) and it allows fast, flexible, and consistent sub-domain modelling techniques. Unlike other groundwater modelling software, data with different cover extension can be implemented within the iMOD without a pre-processing phase for fitting model-grid resolution but it regrids the datasets automatically to the selected resolutions. The software is freely available and open source, can be implemented either using the Graphical User Interface or with Python. In this study, the latter was preferred due to its ability to facilitate working with groundwater models and for large datasets. Based on the conceptual geological model described in Arras et al. (2019), the plain was divided into six geological layers with different hydraulic conductivity values, which was then voxelized to a model with 45 model layers. The recharge values were derived from the application of Soil Water Balance code (Porru et al., 2021). The identification of the fresh water and saltwater interface was derived from direct measurements carried out with a multiparameter probe in deep wells (20/30 m deep) in 2021. The model was then validated using piezometric heads, electrical conductivity values and chemical and multi-isotopes data collected between 2020 and 2021. The implemented 3D model is a preliminary valuable tool to support groundwater management. It can also be used to simulate the effects of future sea water intrusion under different climate change driven sea-level and natural recharge scenarios. Arras C., Balia R., Buttau C., Cau P., Da Pelo S., Funedda A., Ghiglieri G., Loi A., Lorrai M., Melis M.T. & Testa M. (2019) - Hydrogeological characterisation of the Flumendosa plain. Flowpath 2019, National Meeting on Hydrogeology, 6. Langevin C.D. & Guo W. (2006) - MODFLOW/MT3DMS–based simulation of variable-density ground water flow and transport. Groundwater, 44(3), 339-351. Porru M.C., Da Pelo S., Westenbroek S., Vacca A., Loi A., Melis M.T., Pirellas A., Buttau C., Arras C., Vacca S., Mason M.L., Lorrai M., Testa M. & Botta P. (2021) - A methodological approach for the effective infiltration assessment in a coastal groundwater. Italian Journal of Engeneering Geology and Environment, 1, 183-193. Verkaik J., Van Engelen J., Huizer S., Bierkens M.F.P., Lin H.X. & Oude Essink G.H.P. (2021) - Distributed memory parallel computing of three-dimensional variable-density groundwater flow and salt transport. Adv. Water Resour., 154, 103976