Calculating exchange times in a Scottish fjord using a two-dimensional, laterally-integrated numerical model

Authors
Citation
Pa. Gillibrand, Calculating exchange times in a Scottish fjord using a two-dimensional, laterally-integrated numerical model, EST COAST S, 53(4), 2001, pp. 437-449
Citations number
25
Categorie Soggetti
Aquatic Sciences
Journal title
ESTUARINE COASTAL AND SHELF SCIENCE
ISSN journal
02727714 → ACNP
Volume
53
Issue
4
Year of publication
2001
Pages
437 - 449
Database
ISI
SICI code
0272-7714(200110)53:4<437:CETIAS>2.0.ZU;2-S
Abstract
In order to assess the potential impact of pollutants, particularly soluble wastes discharged by the mariculture industry, on the fjordic sea loch env ironment in Scotland, simple management models have been developed which es timate steady-state concentrations based on the quantities of effluent rele ased and the residence time of such material within a loch. These models ma ke various simplifications about the hydrodynamic characteristics of Scotti sh sea lochs, the most important of which is the concept of an exchange tim e which parametrizes the rate at which pollutants are removed from the syst em. Exchange times for individual lochs are calculated using the tidal pris m method, which has some well-known shortcomings. In this paper, a two-dime nsional laterally-integrated circulation model is used to investigate the e xchange characteristics of Loch Fyne and its sub-basins. By simulating the transport of a passive, conservative tracer, the turnover times for the loc h, two sub-basins and various depth layers are calculated. By varying the s tarting time of the tracer simulations, the variability in the exchange tim es is examined. The results from the circulation model are compared with th e estimates given by the tidal prism method. The results show that the tida l prism method consistently underestimates the exchange times, although the predicted times tend to lie within the range of the simulated times. Inclu ding a simple return flow factor into the tidal prism estimate leads to sig nificant improvements in the comparison.