Application of a rapid-assessment method for seawater intrusion vulnerability: Willunga Basin, South Australia

Leanne Morgan, Adrian Werner, M Morris, Michael Teubner

    Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

    12 Citations (Scopus)


    Seawater intrusion (SWI) causes degradation of water quality and loss of water security in coastal aquifers. Although the threat of SWI has been reported in all of the Australian states and the Northern Territory, comprehensive investigations of SWI are relatively uncommon because SWI is a complex process that can be difficult and expensive to characterise. The current study involves the application of a first-order method developed recently by Werner et al. (Ground Water 50(1):48–58, 2012) for rapidly assessing SWI vulnerability. The method improves on previous approaches for the rapid assessment of large-scale SWI vulnerability, because it is theoretically based and requires limited data, although it has not been widely applied. In this study, the Werner et al. (Ground Water 50 (1):48–58, 2012) method is applied to the Willunga Basin, South Australia to explore SWI vulnerability arising from extraction, recharge change and sea-level rise (SLR). The Willunga Basin is a multi-aquifer system comprising the uncon-fined Quaternary (Qa) aquifer, confined Port Willunga Formation (PWF) aquifer and confined Maslin Sands (MS) aquifer. Groundwater is extracted from the PWF and MS aquifers for irrigated agriculture. In the Qa aquifer, the extent of SWI under current conditions was found to be small and SWI vulnerability, in general, was relatively low. For the PWF, SWI extent was found to be large and SWI is likely to be active due to change in heads since pre-development. Anecdotal evidence from recent drilling in the PWF suggests a seawater wedge at least 2 km from the coast. A relatively high vulnerability to future stresses was determined for the PWF, with key SWI drivers being SLR (under head-controlled conditions, which occur when pumping controls aquifer heads) and changes in flows at the inland boundary (as might occur if extraction increases). The MS aquifer was found to be highly vulnerable because it has unstable interface conditions, with active SWI likely. Limitations of the vulnerability indicators method, associated with the sharp-interface and steady-state assumptions, are addressed using numerical modelling to explore transient, dispersive SWI caused by SLR of 0.88 m. Both instantaneous and gradual (linear rise over 90 years) SLR impacts are evaluated for the Qa and PWF aquifers. A maximum change in wedge toe of 50 m occurred within 40 years (for instantaneous SLR) and 100 years (for gradual SLR) in the Qa. In the PWF, change in wedge toe was about 410 and 230 m after 100 years, for instantaneous and gradual SLR, respectively. Steady state had not been reached after 450 years in the PWF. Analysis of SLR in the MS was not possible due to unstable interface conditions. In general, results of this study highlight the need for further detailed investigation of SWI in the PWF and MS aquifers. Establishing the extent of SWI under current conditions is the main priority for both the PWF and MS aquifers. An important element of this involves research into the offshore extent of these aquifers. Further, predictions of SWI in the PWF should consider future extraction and SLR scenarios in the first instance. A field-based investigation of the Willunga aquifer is ongoing, and the current study provides guidance for well installation and for future data collection.

    Original languageEnglish
    Title of host publicationGroundwater in the Coastal Zones of Asia-Pacific
    Number of pages21
    ISBN (Print)9789400756472
    Publication statusPublished - 2013

    Publication series

    NameCoastal Research Library
    ISSN (Print)2211-0577
    ISSN (Electronic)2211-0585


    Dive into the research topics of 'Application of a rapid-assessment method for seawater intrusion vulnerability: Willunga Basin, South Australia'. Together they form a unique fingerprint.

    Cite this