Evaluation of management scenarios for potable water supply using script-based numerical groundwater models of a freshwater lens

Vincent E.A. Post, Sandra C. Galvis, Peter J. Sinclair, Adrian D. Werner

Research output: Contribution to journalArticlepeer-review

15 Citations (Scopus)
1 Downloads (Pure)

Abstract

Challenges in balancing freshwater demands and the long-term availability of freshwater from small island aquifers warrants responsive management, whereby groundwater conditions guide decisions about pumping rates to avoid well salinization. We evaluate responsive freshwater lens management for the first time, through transient, three-dimensional, dispersive modelling of Bonriki Island (Kiribati). Both responsive- and fixed-management scenarios are explored, including a novel pumping redistribution strategy. Modelling results reveal that responsive management offers superior lens protection, particularly during droughts. Pumping redistribution produced lower salinities but greater lens depletion. All scenarios indicate that the Bonriki lens will continue to decline, consistent with previous shorter-timeframe projections. Lower lens storage losses are attainable by abstracting groundwater at the maximum acceptable salinity, contrary to traditional strategies of seeking the lowest available salinities. The methodology developed in this research provides a blueprint for investigating responsive, “monitor-and-react” management scenarios, which we advocate as best practice for balancing freshwater demands with long-term lens security.

Original languageEnglish
Pages (from-to)843-855
Number of pages13
JournalJournal of Hydrology
Volume571
DOIs
Publication statusPublished - Apr 2019

Keywords

  • Atoll island
  • Drought
  • Saltwater intrusion
  • Sustainable yield
  • Wellfield operation

Fingerprint

Dive into the research topics of 'Evaluation of management scenarios for potable water supply using script-based numerical groundwater models of a freshwater lens'. Together they form a unique fingerprint.

Cite this