2518 
F. Große et al.: Looking beyond stratification 
Biogeosciences, 13, 2511-2535, 2016 
www.biogeosciences.net/13/2511/2016/ 
Table 1. Average critical quantities (2000-2012) characterising the O2 dynamics in the four different 4 x 4-regions (see Fig. 2, yellow boxes). 
Fluxes are cumulated from 1 April to 30 September and relate to a surface layer of thickness Z) re f = 25 m. 
Region 
A-SNS 
B-SCNS 
C - NCNS 
D-NNS 
PPmld 
gem“ 2 
169.0 
147.8 
134.6 
148.1 
ADH or g j n 
gCm- 2 
95.9 
109.2 
92.3 
111.0 
ADH 0 rg,out 
gem“ 2 
89.2 
107.6 
92.7 
114.4 
EXPorg 
gCm“ 2 
23.4 
17.5 
16.2 
18.4 
MIXq 2 
g0 2 m -2 
116.1 
66.7 
13.7 
18.2 
initial O2 
mgC>2 L -1 
10.1 
9.9 
9.5 
9.5 
final O2 
m g0 2 L -1 
7.7 
7.9 
8.0 
8.3 
t strat 
days 
80 
151 
220 
226 
Dmld 
m 
11.4 
14.7 
23.1 
25.7 
^bot 
m 
39.6 
43.5 
93.0 
113.4 
area 
km 2 
7643.1 
7454.3 
7108.9 
6677.2 
^sub 
km 3 
111.3 
138.0 
483.3 
590.5 
different regions encompassing 2x2 grid cells (see Fig. 2, 
regions 3-5). First, mass balances for the entire volume be 
low the thermocline (hereafter “sub-MLD”) in region 3 are 
compared with the corresponding bottom layer mass bal 
ances to identify differences between the bottom layer dy 
namics and the dynamics within the entire sub-MLD volume. 
This is done for 2 years, 2002 and 2010, to analyse variations 
between these years. Region 3 was chosen as it shows the 
lowest bottom O2 concentrations within the entire model do 
main, with the overall minimum in 2002 and relatively high 
concentrations in 2010. The daily resolved MLD defines the 
upper integration limit for the sub-MLD mass balances, i.e., 
the integration depth may vary during the stratified period. 
The daily MLD is defined as the vertical level of the model 
grid which is closest to the daily average MLD of the 2x2 
region according to Eqs. (1) and (2). 
Second, we compare the O2 mass balances of the bottom 
layer for regions 4 and 5 in 2002 with that of region 3 to 
unveil regional differences. In a last step the mass balance 
analysis is applied to interpret the O2 evolution observed at 
North Dogger (see Fig. 2, region 2). 
The O2 concentrations and saturation concentrations 
shown in the different mass balances represent the average 
values within the analysed volume. Values of O2 satura 
tion concentrations were calculated according to Benson and 
Krause (1984) using simulated T and S. The fluxes presented 
are cumulative changes in the O2 concentrations of the con 
sidered volume, i.e., the values at the end of the stratified pe 
riod reflect the total net change of the O2 concentrations due 
to the corresponding physical or biological process. Positive 
and negative values at the end of the stratification period in 
dicate net gain and loss, respectively. The slope of each line 
represents the intensity of the corresponding flux at the spe 
cific moment in time, i.e., a steep positive (negative) slope 
implies a strong gain (loss) effect. 
3 Results and discussion 
3.1 Model validation 
3.1.1 Temporal evolution of bottom O2 
Figure 3 shows the comparison of simulated bottom O2 
against time series data at the Cefas station North Dogger for 
the years 2007 (a) and 2008 (b) and the MARNET station 
Ems during 2010 (c) and 2011 (d). The indicated stratifica 
tion period was derived from the simulated temperature fields 
using Eq. (1). 
At North Dogger, observed and simulated bottom O2 con 
centrations show a steady decrease after the onset of stratifi 
cation. While stratification according to Eq. (1) starts a bit 
earlier compared to that described by Greenwood et al. 
(2010), the beginning of the decrease in bottom O2 concen 
trations coincides well. The simulated and observed O2 con 
centrations at this time are in good agreement. 
Some small-scale fluctuations in the observations are not 
fully reproduced by the simulation, however, the general evo 
lution is represented well by the model. The average O2 re 
duction in the simulation is slightly less than in the observa 
tions, visible in the difference between the concentrations at 
beginning and end of the stratified period. Stratification ends 
a bit earlier in the simulation, with the result that simulated 
bottom O2 starts to recover while the observed concentra 
tions continue to decline. The observed O2 concentration at 
the end of the stratified period is about 6.8 mg02L _1 , while 
the simulation results in about 7.4mg02 L _1 . 
In 2008, we can see a similar slight overestimation of sim 
ulated O2 concentrations, but less than in 2007. Some minor 
fluctuations in the observations are again not represented by 
the model, but the general evolution of bottom O2 is repre 
sented well. It should be noted, that the different depths of the 
time series (76 m for simulation, 85 m for observation) may