Alternatives to the mathematical modeling of beaches

Andrew Cooper, OH Pilkey

Research output: Contribution to journalArticlepeer-review

16 Citations (Scopus)

Abstract

dDespite acknowledgement by scientists and modelers of the shortcomings of mathematical models of beach behaviour, they continue in widespread use for engineering design and environmental impact studies. In part, this is because of the assumed absence of a viable alternative approach to prediction. The objective of this paper is to suggest that a conceptual approach can succeed as an alternative to mathematical models for predictions with accuracy ranges useful for engineering and coastal management purposes. The most valuable information for prediction is past engineering experience on the beach in question or the experience on neighboring beaches. The next most useful may be the global experience on oceanographically similar beaches. Another approach is the use of geoindicators and field studies on the beach. The widely applied ``predict nothing'' approach to beach engineering may well work if fortified by common sense. Finally, a composite approach in which the assembled body of information on a stretch of coast is considered and used to provide an expert judgement on the likely course of future coastal behavior is considered the best option. In all cases a gradualism or go-slow approach is strongly recommended because no method can accurately predict the response to engineering of the complex natural coastal system. Answers from such approaches will be approximations with significant error bars expressed in qualitative terms but will be based on a reality that applied models with all their simplified assumptions and omissions are incapable of achieving.
Original languageEnglish
Pages (from-to)641-644
JournalJournal of Coastal Research
Volume20
Issue number3
Publication statusPublished (in print/issue) - 2004

Fingerprint

Dive into the research topics of 'Alternatives to the mathematical modeling of beaches'. Together they form a unique fingerprint.

Cite this