Influence of aquifer heterogeneity on sea level rise-induced seawater intrusion: A probabilistic approach

Abstract

Seawater intrusion (SWI) is influenced by a variety of coastal phenomena, such as sea level rise (SLR), inundation of low-lying coastal regions, coastal storms, recharge rate variations, and pumping-induced saltwater upconing. Quantification of the influence of heterogeneity in the hydraulic conductivity field on SWI combined with SLR, land-surface inundation, and recharge rate variations in an unconfined aquifer is the main objective of the present study. The principal SWI indicators used in this study are length of the SWI wedge, seawater volume, and weighted average transition zone width. Characterized by the hydraulic conductivity field variance (σlnk2), the longitudinal correlation length (λx), the type of SLR (gradual or instantaneous SLR), the land-surface inundation consideration, and the recharge rate variations, 72 scenarios have been introduced, and for each of them, 50 sets of heterogeneous hydraulic conductivity fields have been generated. Based on two approaches, namely ensemble Monte-Carlo and a Bayesian framework, it is demonstrated that: (1) the land-surface inundation consideration increases the SWI wedge length and the seawater volume regardless of the type of SLR, while it decreases the weighted average transition zone width in gradual SLR scenarios; (2) λx has a more significant impact on SWI characteristics compared to σlnk2; (3) increasing the degree of aquifer heterogeneity results in larger effective dispersion values; (4) Numerical bootstrapping suggests that the introduced Bayesian framework could be adopted as an alternative to computationally demanding methods such as bootstrapping for stochastic analysis of SWI; (5) Reliability analysis indicates the general belief that considering the heterogeneity decreases the SWI wedge length and the seawater volume, while increases the transition zone width compared to the homogeneous modeling is associated with huge amounts of uncertainty proportional to the aquifer heterogeneity itself; and (6) the results show that the impact of heterogeneity on the SWI indicators is similar under different recharge rates.