Thesis for the Degree of Doctor of Philosophy - DTU Orbit

FIGURE 1. Sampling sites (the red solid circle) for August 2006 and April 2007 expeditions. Blue and turquoise arrows represent 129 I
transported from Sellafield and La Hague, respectively. Purple arrows refer to deep flow into the Baltic Proper while red arrows
mean surface flow. Ranges of salinity concentration, within the Skagerrak, Kattegat, and Baltic Proper, are labeled.
mized stratification). Furthermore, inventory estimates are
updated and a model is developed for future prediction,
involving the processes of seawater inflow and outflow,
freshwater tributary and sedimentation effects.
Sampling and Analytical Techniques
Depth-profiles of seawaters were collected from 16 sites in
August 2006 and 19 sites in April 2007 (Figure 1 and
Supporting Information Figure S-1). Sampling was carried
out by the research Vessel Argos, operated by the marine
division of the Swedish Meteorological and Hydrological
Institute. Water samples from different depths were collected
in Nansen bottles (Hydro Bios). Meanwhile, a CTD automatic
sampler (CTD- model SD204) was launched to measure
hydrologic parameters such as salinity, temperature, depth,
and dissolved oxygen. Salinity was also measured on board,
using an AEG MINISAL 2100 salinometer based on relative
conductivity. Oxygen was determined by Winkler titration
method using an automated potentiometric titration system
and detail of the procedure has been reported elsewhere (9).
Iodine isotopes were extracted from seawater by following
the method of Hou et al. (10, 11), and the details are presented
in the Supporting Information. To all samples, 0.2 mL of 200
Bq/ml 125 I - (NaI) as a chemical yield tracer and 2.0 mg of
stable iodine (Woodward Iodine Company, U.S.) as a carrier
were used. Iodine species were separated by anion exchange
chromatography and determination of 129 I and 127 I were
carried out using Pelletron (NEC machine) AMS and an X
Series II ICP-MS (Thermal Electron Corporation), under hot
plasma conditions, with the Xt interface system. Processing
blanks 129 I/ 127 I value was 1.5 ( 0.5 × 10 -13 , which is more
than 2 orders of magnitude lower than the value in the
samples · . Total analytical uncertainty for 129 I/ 127 I value is
normally less than 10% and the detection limit for 127 I,
calculated as 3SD of blanks, was 0.27 nM.
Results
The data sets of 127 I and 129 I together with temperature, oxygen
concentration and salinity for the two sampling campaigns
904 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 45, NO. 3, 2011
of August 2006 and April 2007 are presented in Supporting
Information Tables S-1 and S-2 and Figures S-2 and S-3.
Surface and deep water (above and below halocline, respectively)
distributions of the isotopes and their ratio
indicate several differences as shown by Figures 2 and 3. In
both August 2006 and April 2007, surface water 129 I concentration
(Figures 2a and 3a) have their highest values in the
Skagerrak basin with a maximum reaching up to 1683 × 10 8
atoms/L (or 106× 10 -8 for 129 I/ 127 I atomic ratio). The trend
of concentration decreases toward the Öresund (location
12), and remains rather constant, around 25-36 × 10 8 atoms/
L, in the Baltic Proper. Relatively more variable distribution
of 129 I between the two sampling times is revealed by the
concentration in the deep water (Figures 2b and 3b). The
plume of maxima shown in the Skagerrak during August 2006
is partly retained along the Swedish coasts during April 2007.
Relatively high concentrations are also retained throughout
the Arkona basin during April, but the more homogeneous
distribution in the southern and central parts of the Baltic
Proper during April is disturbed by an increase along the
Bornholm deep.
To summarize the 129 I depth distribution, we observe a
2-19 times decrease in concentration from the surface down
to bottom layer in the transition zone between the Kattegat
and Skagerrak (locations 17-19, Figure 4). While in the
Kattegat (locations 12-16), 129 I concentration gradually
increases with depth above halocline, around 35 m, whereas
the opposite trend is observed below 35 m. In the Arkona
basin (locations 1 and 2, Figures 2a, b and 3a,b), a rather
sensitive region for different seasons, high 129 I concentrations
are observed in August 2006 in the surface layer, extending
to around 30 m, and then the concentration decreases with
increasing depth. But an opposite trend was found in April
2007 where 129 I concentration increased from the surface
water (33 × 10 8 ( 3 atoms/L) to the bottom water (163 × 10 8
( 27 atoms/L). In the central Baltic Proper (locations 5-10),
there is an obvious 129 I stratified layer where the concentration
increases with depth and attains a maximal value 150 × 10 8
atoms/L in the bottom water in August 2006.