Application of a Numerical Statistical Model to Estimate Potential Oil Spill Risk - Mathematical Modelling and Numerical Simulation of Oil Pollution Problems - page 125

to describe the mechanisms governing the fate of oil slick has also been developed
[ 5 , 11 ]. The principal aim of oil spill numerical model is for oil spill response
strategy to reduce environmental and economic impacts when spills happen. A lot of
deterministic models have evolved from two-dimensional trajectory-type models to
three-dimensional models that include transport and fate processes [ 1 , 13 ]. Forecasts
of the oil trajectory and distribution are very useful to help make decisions to control
and clean up spilled oil, but most oil spill models consider the oil-slick movement as
a deterministic mechanism [ 2 , 9 ]. When it comes to the problem of forecasting oil
spill risk, the deterministic approach seems inadequate. Since the oil spill impacts
vary according to a range of factors—from spill size, meteorological and hydrolog-
ical conditions, and physicochemical properties of the oil, to the characteristics of
the affected areas, statistical methods need to be employed to present a risk map of
oil impacting. There is uncertainty on not only the amount and the position of oil
that would be released but also the site and time dependent environmental condi-
tions. Compared with the direct usage of hydrological and meteorological data in
a deterministic model, adopting multiple scenarios run during a long period seems
appropriate more to represent the feature of a specified domain. The progressive
advancement of computer speed in the last decade makes the implementation of
vast test simulation computations to investigate oil spill risk possible. The statistical
model should evaluate not only the probability of a region to be influenced by spilled
oil, but should also predict the time taken for the oil to reach the location.
What calls for special attention is that the degree of the vulnerability associated
to oil spills for different regions varied obviously. For example, the marine protected
areas are more sensitive for contaminants than harbour area. A quantitative method-
ology, by introducing the index accounting for both ecological and socioeconomic
dimensions of vulnerability, to assess the spatial distribution and the degree of coastal
vulnerability to oil spills was developed by [ 7 ].
6.2 The Basics of Oil Spill Modelling
6.2.1 Environmental Dynamical Factors
It benefits us little to have excellent results for spilled oil distribution if we are
unable to accurately describe the surrounding conditions. Linkage to environmen-
tal dynamical model is of paramount importance for oil spill modelling since flow
regimes are necessary for transport process. A high-resolution flowmodel is required
to provide the complex hydrodynamics information of sea waters of the influence of
atmospheric forcing, tide, surface wave, and river discharge.
Not only do the winds directly drift oil slick horizontally, they drive the breaking
waves spilt the surface slick into droplets and then propel them into the water col-
umn, even as a crucial factor affecting evaporation process. The wind data can be
obtained from field measurements for forecast/hindcast oil spills. Since the in-situ