32 
Callies & Schwichtenberg - Drift modelling 
decision making with regard to the application of chemical dispersants is supposed to be 
very complex. 
TWO 7°30'0"0 8°0'0"O 8 0 30'0"0 9WO 
Undispersed oil released at every low tide in 2008 
Figure 6.2: The left panel shows simulated distributions of pollutants five days after a hypothetical oil 
release took place at the location indicated by the ship symbol. Two examples assuming oil being re 
leased on Jan 7 or Nov 5 in the year 2008 are colour coded in blue and red, respectively. Both of the two 
hypothetical accidents were simulated assuming a) that the oil remained untreated (large dots) and b) 
that the oil was fully dispersed right after its release (small dots). Time series on the right hand side 
summarize results with (bottom panel) and without (top panel) chemical dispersion for a whole ensemble 
of simulations started at each low tide within the year 2008. Both of the two graphs show percentages of 
pollutant that enter sensitive Wadden Sea areas (green areas in the left panel) at any time within the first 
five days after the hypothetical accident took place. 
The time series on the right hand side of Figure 6.2 summarize for all simulations the per 
centages of pollutant that would have entered sensitive German Wadden Sea areas with and 
without the application of a 100 % effective chemical dispersant. A substantial reduction of 
the probability that the pollutant would enter the Wadden Sea by means of an effective dis 
persion can clearly be recognized. Of course, the results shown hold just for the specific lo 
cation selected. Schwichtenberg et al. (2016) extended this kind of analysis to produce a 
probability map covering the whole German Bight area (see below). 
With regard to the behaviour of untreated oil it should be noted that our simulations did not 
take into account the process of beaching so that simulated tracer particles can move along 
the coastline until they enter a tidal inlet. 
Variability in space 
To study the spatial variability of benefits from using dispersants, Schwichtenberg et al. 
(2016) introduced a regular grid made up by 636 cells of about 5x5 km 2 , covering the whole 
area of the inner German Bight. Considering each of these grid cells as a hypothetical source 
of oil pollution, they initialized corresponding simulations every 28 hours in the years 2008- 
2014. Based on the outcomes of the resulting 2190 simulations per grid cell (integration time: 
seven days) they compared the times after which the first oil reached any sensitive coastal 
area with and without application of a chemical dispersant. A substantial increase of the 10 th 
percentiles of travel time caused by dispersion becomes evident comparing the two panels in