Die Kuste, 81 (2014), 255-271 
266 
and E-wards of Helgoland) in die afternoon of 08.01.2012 (about 58 h after die accident). 
Since we cannot verify diat the objects were double sightings, we do not compare these 
observations with drift simulations. But diis illustrates, that aldiough the containers have 
been dropped off probably very close to each other (or even at the same position) at 
nearly the same time, the objects-over-board have already taken very different trajectories. 
The increasing distances between the containers observed 4 h, 33 h and 58 h after the 
accident (1/2 nm, 8 nm and 13 nm) illustrate the different behavior of drifting object de 
pending on size, drowning, shape, etc.. Witiiout knowing any of tiiese properties die un 
certainties of a drift simulation are very high. Additionally the turbulence and other pro 
cesses of different scales in wind, waves and currents put a random forcing on the ob 
jects, which is not possible to predict deterministically. 
Table 1: Observation of a container package made by GS Neuwerk 06.01.2012. 
Time in UTC 
Latitude 
Longitude 
17:18 
54° 18.2‘ N 
007° 11.6‘E 
17:30 
54° 18.1‘N 
007° 11.9‘E 
17:45 
54° 18.0‘ N 
007° 12.4‘ E 
18:00 
54° 17.8‘ N 
007° 12.9‘ E 
18:15 
54° 17.7‘ N 
007° 13.4‘ E 
18:30 
54° 17.6‘ N 
007° 13.9‘E 
18:45 
54° 17.5‘ N 
007° 14.4‘E 
19:00 
54° 17.3‘ N 
007° 14.9‘ E 
19:30 
54° 17.3‘ N 
007° 15.5‘ E 
19:45 
54° 17.T N 
007° 16.4‘ E 
20:00 
54° 16.9‘ N 
007° 16.9‘ E 
20:15 
54° 16.8‘ N 
007° 17.4‘ E 
20:30 
54° 16.8‘ N 
007° 17.9‘ E 
20:45 
54° 16.8‘ N 
007° 18.3‘ E 
21:00 
54° 16.8‘ N 
007° 18.6‘ E 
23:30 
54° 17.5‘ N 
007° 21. 4‘ E 
Further the objects drifting characteristics may change over time. In the evening two 
connected containers, which were probably die ones reported east of Helgoland in die 
afternoon, were sighted by GS Neuwerk soutiiwest of Helgoland. Again die positions 
were radar plotted, but in die morning of 09.01.2012 the containers broke off, and one 
deteriorated distributing wooden planks and a package of wood. Later die remaining blue 
container was salvaged. In die subsequent 6 days wood, wood packages and fragments of 
containers were found in the Elbe estuary and at die Nordi Frisian coast near the Eider 
estuary. 
We apply SeatrackWeb with BSHcmod/LME forcing simulating two connected con 
tainers drifting on 06. January 2012 between 17:18 UTC and 23:30 UTC and compare die 
results witii observations (given in the Tab. 1). Fig. 5 shows the trajectory of the drift 
simulation in blue and of the observation in magenta. At die beginning die simulated 
trajectory follows nicely the observations, later the containers move more soutiiwards 
than in the simulation. At the end of die simulation the distance between observed and