05.04.2013 Views

Appendices 5-13 - Nautilus Cares - Nautilus Minerals

Appendices 5-13 - Nautilus Cares - Nautilus Minerals

Appendices 5-13 - Nautilus Cares - Nautilus Minerals

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

dispersion is modelled using a random walk scheme (Bear and Verruijt, 1987), with the<br />

magnitudes scaled by horizontal and vertical diffusion coefficients (Okubo, 1971). The mixing<br />

parameters were conservative estimates of deep water conditions.<br />

2.2 Stochastic Modelling<br />

Originally CHEMMAP was designed to simulate specific spill incidents and/or ongoing discharges<br />

for evaluating impacts and damages (French et al. 1996). More recently, the model has been set<br />

up in a probabilistic stochastic configuration, allowing evaluation of risks of consequences and<br />

statistical computations (French McCay and Isaji, 2004). While a few chemical spill models exist<br />

that can simulate transport and physical fate of single events (Lunel, 1991; Shen et al., 1995; Rusin<br />

et al., 1996), CHEMMAP is unique in being able to evaluate biological impacts, in its stochastic<br />

implementation, and in its interconnection with hydrodynamic models, geographical information<br />

systems, and its graphical user interface. In the stochastic mode, CHEMMAP can be used to<br />

predict the fate of multiple or continuous releases that occur under a random selection of prevailing<br />

conditions (also known as stochastic modelling). The stochastic model performs a large number of<br />

sample simulations for a given release site, randomly varying the sample time frame, so that the<br />

transport, concentration and dilution of each particle representing the plume mass and<br />

concentration are subject to a different set of prevailing current conditions and water properties.<br />

During each simulation, the model records the grid cells that were contacted by the plume, as well<br />

as the amount of time that had elapsed prior to the contact or exposure.<br />

Once the stochastic modelling is complete, the results are compiled from each of the sample<br />

trajectories to provide a statistical weighting to the likelihood of exposure of grid cells. Stochastic<br />

results can be summarised as:<br />

1. The probability, frequency or risk that a grid cell may be exposed to the<br />

plume; and<br />

2. The maximum expected (averaged) concentrations of the plume in each grid<br />

cell.<br />

The stochastic modelling approach provides an objective measure of the possible outcomes of a<br />

release, as well as the means of quantifying the likelihood of a given outcome. The most commonly<br />

occurring conditions would be selected most often while conditions that are more unusual can also<br />

be represented.<br />

Page 6 of 24

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!