Transience of seawater intrusion in response to sea level rise

Abstract

Understanding seawater intrusion (SWI) induced by sea level rise (SLR) is important for the future management of many coastal aquifers. Only simplified steady state sharp interface analyses of generalized SLR‐SWI exist in the literature, and the important issue of associated time scales has been neglected. We employ numerical modeling in order to explore the transience of dispersive SLR‐SWI in common unconfined coastal aquifer settings. An instantaneous SLR is adopted to compare with the instantaneous sea level drop (SLD) case of a previous SLD‐SWI analysis. Temporal asymmetry between the SWI responses to SLR and SLD is observed. A SLR‐SWI simulation series indicates that toe “representative response times” (time to reach 95% of new steady state) range from decades to centuries for a 1 m SLR. Significant discrepancies between the representative response times of various SWI quantitative indicators (e.g., toe position, wedge center‐of‐mass) are observed. This demonstrates that the indication of steady state SWI conditions depends upon the monitoring approach and thus holds implications for studies reporting that SWI steady state has been attained. We adopt 100 years as a typical “planning time frame” and compare 100 year and steady state SLR‐SWI. As expected, the simplified steady state sharp interface solution overpredicts the 100 year landward toe shift in most cases. However, some simulations exhibit temporary “overshoot” of the steady state interface position: this is in contradiction to the presumption that steady state SWI is the worst case. Steady state sharp interface estimates appear to be at best useful as initial approximations of SLR‐SWI, given that they span 40%–250% of 100 year dispersive interface results for the cases considered.