812 
U. Callies et al.: Surface drifters in the inner German Bight 
Ocean Sci., 13, 799-827, 2017 
www.ocean-sci.net/13/799/2017/ 
circulation becomes cyclonic for 1 day. Under transi 
tional conditions, directional errors are particularly high 
in TRIM (Fig. lOd). 
Days 21-26 (17-22 June): Residual circulation gradually 
changes from anticyclonic to cyclonic (Fig. 3). During 
days 21-23, considerable errors in both BSHcmod + W 
and TRIM simulations resemble each other to a surpris 
ing degree (e.g. Fig. 9a and e). Except for drifter no. 9, 
drift directions are typically rotated to the left of obser 
vations (Fig. lOd). From about day 22 onward, drifter 
nos. 6 and 8 start separating again (Fig. 5). Expectedly, 
neither model reproduces sub-grid-scale differences in 
speed (day 22; SM3) or direction (day 23; Fig. 9a and e). 
Starting on about day 22, fast movements mostly in line 
with prevailing wind directions (e.g. Fig. 9a and e) and 
greatly exceeding simulated counterparts (Fig. 6) sug 
gest that drifter no. 9 experienced some problem with 
its drogue. 
Days 27-28 (23-24 June): On day 27 (Fig. 9b and f), strong 
winds from the north-west give rise to southern trans 
ports. Substantial differences between speeds of neigh 
bouring drifter nos. 6 and 8 (unresolved in simulations) 
imply a short period of their fast convergence (Fig. 5). 
BSHcmod + W simulations greatly benefit from the in 
clusion of windage (Fig. 9b and SM2), while TRIM 
simulations are more consistent even without windage 
(Fig. 9f). On day 28, winds abate. 
Days 29-33 (25-29 June): This is a period with variable 
wind directions. Drifter displacements are generally un 
derestimated (Figs. 6 and 10a), observed northward 
transports (e.g. for drifter no. 8; Fig. 4e) are not rea 
sonably reproduced based on BSHcmod + W (Fig. A2e) 
and even less based on TRIM (Fig. A4e). 
Day 34 (30 June): Drifter nos. 6 and 8 converge quickly 
(Fig. 5), caused by a fast west-northwest movement of 
drifter no. 8, not shared by drifter nos. 5 and 6 (SM3). 
No model resolves these substantial differences. 
Days 35-38 (1-4 July): Drifter nos. 5, 6 and 8 all move 
quickly into northern or north-western directions 
(Fig. 4). Fargest drifter displacements occur on day 35 
(see Figs. 9c, g and 6) with strong winds from the south 
east. Drifter no. 8 moving faster and more aligned with 
wind direction than its companion drifters could possi 
bly indicate problems with the drogue. 
On day 36, TRIM (but not BSHcmod) assumes the wind 
to persist (Fig. 3 or SM3), which results in a substantial 
overestimation of drifter displacements (Fig. 10a). Ac 
cording to observations at Heligoland (Fig. 3), winds 
used by BSHcmod + W seem more realistic. 
Under low wind conditions on day 37, BSHcmod + W 
(to a lesser degree also TRIM) very much underesti 
mates drift speeds (Fig. 6). On day 38, the process of 
drifter nos. 6 and 8 coming to rest is well reproduced in 
both models (SM3). 
Days 39-41 (5-7 July): Freshening south-westerly winds 
strengthen a cyclonic circulation (Fig. 3). The extremely 
fast movement of drifter no. 8 is remarkable in reac 
tion to this forcing (Figs. 4e and 6c). Simulations for 
drifter nos. 5 and 6 perform well, while the behaviour 
of drifter no. 8 cannot be reproduced. 
Days 42-43 (8-9 July): The wind turning from south-west 
to north-west implies a fast transition from a cyclonic to 
an anticyclonic residual current regime (Fig. 3). Mod 
els perform well for drifter no. 6, while simulations for 
drifter no. 8 are again very poor (Figs. 6c and 9d, h). 
Days 44-53 (10-19 July): Only drifter nos. 5 and 6 are left; 
both of them are already located in coastal waters. 
Extra large differences between wind velocities used 
in BSHcmod and TRIM occur (Fig. 10b). Effects of 
a sudden reversal of the mean wind direction between 
days 50 and 51 are reasonably reflected in both models. 
4 Discussion 
The model validation study suggests the assumption that 
inclusion of either wind drag or Stokes drift compensates 
insufficient vertical resolution (5 m) of surface currents in 
archived BSHcmod output. Magnitudes of TRIM surface 
currents, representative of a layer of 1 m depth, were gen 
erally similar to those observed (Fig. 10a). In many cases, 
however, 25 h simulations based on BSHcmod + W outper 
formed those based on TRIM, in other cases (e.g. days 13- 
16) TRIM simulations were in better agreement with obser 
vations (Figs. 8c, g or 10). 
In several other studies (e.g. Gastgifvars et ah, 2006; 
Kjellsson and Dôôs, 2012; De Dominicis et al., 2012), simu 
lated marine surface currents were found to be too small, pos 
sibly also due to insufficient resolution of the marine surface 
layer. As a side effect, predictions may be particularly good 
when marine currents and winds are nearly parallel (Gastgif 
vars et ah, 2006). The drift component most underestimated 
based on just BSHcmod Eulerian currents was a displace 
ment towards the east, along the most frequent wind direc 
tions (compare Figs. 4 and Al). This deficiency could very 
effectively be remedied by adding direct effects of winds 
or waves. However, during periods when anticyclonic resid 
ual currents prevail (along with winds from the north-west, 
for instance), currents will generally not be in the direction 
of winds (e.g. day 18; Fig. 8d and h), unlike the situation 
with south-westerly winds driving a cyclonic circulation (e.g. 
day 3; Fig. 7a and e). Erroneous residual surface currents in