The BOX project aims to reduce the negative effects of anthropogenic nutrient inputs to the Baltic Sea, such as cyanobacterial blooms and oxygen depletion in the deep waters with resulting “dead bottoms, by mechanically deploy oxygen-rich halocline water to the deep water. This will be done by a set of wind-driven pumps. By increasing the area of oxygenated bottoms, the amount of phosphorus in the water column will be reduced as it now can be bound to the sediments. When less of the plant nutrient phosphorus (P) is available, less cynaobacterial blooms are possible.

Cyanobacterial bloom in the Baltic Sea
Cyanobacterial bloom in the Baltic Sea

The kick-off meeting took place in Gothenburg in February and mainly focused around the planning of the monitoring program. Measurements will take place in Byfjorden, on the west-coast of Sweden and will start in April 2009 and continue twice a month over the coming three years. It is important to start the surveys prior to the pumping as this data will be used as background information in order to asses the environmental changes that will take place once the oxygenation of the deep water begins. Parameters that will be measured include salinity, temperature, oxygen, pH and plant nutrients. Permanent loggers for salinity, temperature and oxygen will also be deployed.

Planning of Utö surveys. Professor Anders Stigebrandt (University of Gothenburg) and Professor Lars Rahm (Linköping University)
Planning of Utö surveys. Professor Anders Stigebrandt (University of Gothenburg) and Professor Lars Rahm (Linköping University)

On the east coast of Sweden a location north east of Utö is selected. However, existing measurements in this area are sparse and it cannot at this moment be immediately decided that this location is suitable, for which it is necessary that it suffers semi-permanent periods of deep-water stagnation. The first task of BOX is here to investigate whether this is the case and, if not, find another suitable location that also is sheltered enough to allow reasonable working conditions for both background data surveys and halocline pumping work.

Ecological programs

To ensure that no negative ecological effects are created during the bottom oxygenation, continuous bottom fauna studies will be conducted. Mussels (Mytilus edulis) will be used as indicators of toxin emission from the sediments as they accumulate substances in their tissue. Sediment surveys will also be undertaken in collaboration with Geological Survey of Sweden (SGU).

The project crates a unique environment to study the colonisation of previously dead bottoms. Both biological and chemical analyses are of outmost interest when the oxygen conditions and denitrificational processes drastically will change. Denitrification is the process that allows nitrate to be reduced to nitrogen which demands low-oxygen conditions.

Collaborations

The BOX project is a collaboration program between the departments of Earth Sciences and Marine Chemistry at the University of Gothenburg and Department of Water and Environmental Studies at Linköping University. The project will run over 3 years and will also initiate further associations with e.g. ecological departments. Contact BOX at BOX-listan[at]gvc.gu.se.

How to reduce the negative effects of anthropogenic nutrient emissions in the Baltic Sea.

Cyanobacterial blooms.
Cyanobacterial blooms.

Excess of the plant nutrient phosphorus in the Baltic Sea is characterised by extensive summer blooms of cyanobacteria.

During the 1990s the phosphorus concentrations in the Baltic Sea were reduced due to the prevailing weather conditions which increased the oxygen levels. The period with reduced nutrient availability was concurrent with reduced cyanobacterial blooms.

Anders Stigebrandt, professor of oceanography, suggests that we learn from nature’s own experiment when we take actions in order to deal with the eutrophication problems in the Baltic Sea. We can simulate the effects of the natural weather conditions and in the same manner oxygenate the Baltic Sea deep water. This is achieved by mechanical pumping of oxygen-rich water from halocline levels down to the deep, oxygen depleted levels. Funding from the Swedish Environmental protection Agency and the Swedish Research Council Formas will enable a pilot study that will asses and evaluate the method.

1. Eutrophication in the Baltic Sea

Key symptoms of eutrophication in the Baltic Sea includes extensive summer blooms of cyanobacteria
bottom-water hypoxia resulting in “dead” bottoms.

Extensive summer blooms of cyanobacteria

Plant uptake of nitrogen and phosphorus is in the ratio of 16:1 (atoms) or 7.2:1 (weight). Since year 2000 about half of the available phosphorus is left in the water column when the spring bloom has used up the available nitrogen. .This is a favourable set-up for the nitrogen-fixing cyanobacteria whose blooming period starts when the water temperature is high enough. The extent of the bloom is set by the amount of phosphorus available in the water, since there is no limit to the available amount of nitrogen when it is taken from the air.

Figure 1a: Phosphourus content in the Baltic Sea
Figure 1a: Phosphourus content in the Baltic Sea
Figure 1b: Dissolved inorganic nitrogen in the Baltic Sea
Figure 1b: Dissolved inorganic nitrogen in the Baltic Sea

The nutrient concentrations in the Baltic Proper vary greatly. The winter concentrations of phosphate (PO4), dissolved inorganic nitrogen (DIN), which mainly consists of nitrate, and the biological production have almost doubled since the 1960s, see Figure 1a and 1b.

The phosphate concentration is related to the oxygen concentration i the deep water (Figure 1c).

Figure 1c: Extent of anoxic areas in the Baltic Sea
Figure 1c: Extent of anoxic areas in the Baltic Sea

When oxygen is available, the metals in the sea-floor sediments can bind phosphate. When oxygen is depleted, the metal-bound phosphate will again be released into the water (see below).

The amount of phosphorus in the Baltic Sea is equivalent to the amount of phosphorus released into the Baltic Sea from external sources (land, air, out-lets, adjacent seas etc.) during a period of 12 years. The amount of DIN in the Baltic Sea is equivalent only to about 1 year of external supply. This means that the nitrogen sink (removal of nitrogen), mainly due to denitrification (a process where nitrate is transferred into molecular nitrogen, N2), is much more efficient compared to the phosphorus sink, which is due to burial in the sea-floor sediments and to export to adjacent seas. The extreme efficiency of the nitrogen sink leads to summer-time nitrogen shortage in the Baltic Proper, despite such a large supply.

Bottom-water hypoxia resulting in “dead” bottoms

Higher forms of animal life demands oxygen for their life-supporting processes. If the concentration of oxygen is less than 2 ml per litre of water, most animals have difficulties with the oxygen supply. At even lower concentrations, which are often the case in the water just above the sea floor, the conditions no longer support suitable habitats for higher life forms and the bottoms are said to be “dead”.

Biogeochemical processes that transmit organic material and their nutrient contents to inorganic forms differ whether oxygen is present at the processes or not. The recycling of nutrients is hence greatly dependant on the existence, or lack, of oxygen.

In other words: the key symptoms of eutrophication – the extensive summertime cyanobacterial blooms and the “dead” bottoms – should be possible to less prominent by an increased supply of oxygen to the deep water. Natures own proof for this hypothesis is that both symptoms where diminished during the period in the 1990’s when the weather conditions caused increased oxygenation of the Baltic Sea deep water. At the same period the amount of available phosphorus was reduced by 50%.

2. Baltic Sea circulation

In order to calculate the effect of a deep-water oxygenation in the Baltic Proper one must know the order of magnitude of the natural oxygenation, which is the result of the natural, vertical water transports in the Baltic Sea.

The salinity in the Baltic Sea is low due to the large freshwater supplies, the narrow and shallow entrance area and the accumulation of low-salinity surface water from the Baltic Sea in the Kattegatt. Inflow of sea water from the Skagerrak, through Little and Great Belts and Öresund, is diminished. In Figure 2 it is seen how the salinity decreases with latitude in the Baltic Sea (light colours indicate low salinities).
Horizontal and vertical salinities in the Baltic Sea

Figure 2: Horizontal and vertical salinities in the Baltic Sea
Figure 2: Horizontal and vertical salinities in the Baltic Sea

The Baltic Proper has a strong vertical stratification, a halocline, where the salinity quickly increases with depth. The halocline normally begins at a depth of 60 meters (see Figure 2).

The position of the halocline in the Baltic Sea is determined by the mixing in the surface layer and by the replenishment of new deep water from the Kattegatt. In the winter the halocline is deepened by erosion, which means that the mixing incorporates water from the halocline into the surface layer. The replenishment of new deep-water lifts the halocline. In periods with large fresh-water supply the density is first reduced in the halocline and the water underneath it (see Figure 3). This is an effect of concurrent increased accumulation of low-salinity Kattegatt water which enhances the blocking of salt-water inflow from the Skagerrak.

 Figure 3: Salinity (upper panel) and oxygen levels (lower panel) in the Baltic Sea over the 20th century.
Figure 3: Salinity (upper panel) and oxygen levels (lower panel) in the Baltic Sea over the 20th century.

When the sea surface is higher in Kattegatt than in the south-west of the Baltic Sea, water flows into the Baltic Sea. To start with, the salinity of the inflowing water is only a little higher than in the surface layer of the Baltic Sea, but the salinity increases with the magnitude of the inflow. The variability of both magnitude and salinity of the inflowing water is extensive. The largest inflows carry more salt than the smaller ones and therefore only these can replace the deep water in the deepest parts of the Baltic Sea.

There are often several years between these replacements. This, in turn, leads to oxygen depletion in the deep. In Figure 3 (lower panel) it is shown that periods with no oxygen (greyed areas) in the deep parts have become more frequent during the second part of the 20th century.

Figure 4: Important processes for the ciruculation of water.
Figure 4: Important processes for the ciruculation of water.

The most important physical processes determining the Baltic Sea vertical circulation are shown in Figure 4. The North-Kattegatt front is a front between low-salinity surface water of the Kattegatt and high-salinity Skagerrak water. It is easily depictured how a deep halocline in the Kattegatt impedes the salty Skagerrak water to flow into the Baltic Sea. The water that does flow into the Baltic Sea move as a dense bottom current, which entrain the surrounding water and thereby reduces the salinity of the current. Between areas of dense bottom currents are areas where the new deep water is stored: dense bottom pools. The pools are subdued to turbulent mixing, which reduces the salinity and to filling which increases the salinity.

3. Natural variations of Baltic Sea phosphorus concentrations

The phosphorus content in Baltic Sea displays large variability in time, which can not be explained by variations in supply (Figure 5).

 Figure 5: Annual Added phosphorus (black line) and dissolved inorganic phosphorus (blue line).
Figure 5: Annual Added phosphorus (black line) and dissolved inorganic phosphorus (blue line).

Therefore, internal processes leading to removal or (sink) or input (source) must be present and causing the seen variability. The multi-year variability (Figure 5) correlates well with oxygen conditions it the deep water (Figure 1).

Some of the surface-produced biomass sinks into the deep water and oxygen is consumed when the organic matter is decomposed. Enhanced biological production leads to increased oxygen consumption in the deep water, which enhances the eutrophication in the Baltic proper. This scenario is referred to as the “vicious cycle”.

The vicious cycle
The vicious cycle

The ”natural experiment” during the 1990s

The phosphorus content in the Baltic Sea decreased substantially during the period 1990-1997. This was due to a rather unusual lowering of the halocline. The top of the halocline is usually found at about 60 m (Figure 6a).

Figure 6a: Normal state of the Baltic Sea
Figure 6a: Normal state of the Baltic Sea

During the period 1992-1997 the halocline top was eroded down to a depth of more than 90 m (Figure 6b).

 Figure 6b: The halocline of the Baltic Sea was eroded to a greater depth in the early 1990s.
Figure 6b: The halocline of the Baltic Sea was eroded to a greater depth in the early 1990s.
  The chain of events that lead to the natural experiment of the early 1990s.
The chain of events that lead to the natural experiment of the early 1990s.

An important effect of the deepening of the halocline was that the volume of the water beneath it, the deep water, decreased substantially. Another important effect was that the oxygen conditions improved considerably at depths between 80 – 120 m. This in turn increased the binding of phosphorus to iron oxides in the sediments (about 3 tonnes of phosphorus per km2). The bottom area in the interval 80-125 m is about 45 000 km2. The sediments here can in a short time bind about 135 000 tonnes of phosphorus, which can explain the concurrent observed reduction of phosphorus content.

Below ”Nature’s experiment” in the Baltic Sea is summarized:

The period 1992-1997
The halocline is eroded. The high oxygen concentrations at 80 to 125 meters depth led to low phosphate concentrations at these depths due to the binding of phosphate to the sediments when oxygenated. In combination with some erosion of the halocline at the later part of the period this resulted in decreased transports of phosphorus to the surface layer and lower phosphorus concentrations in this layer.

After 1997
The halocline has resumed its ”normal” depth and the oxygen concentration beneath the halocline is again lowered. The phosphate concentration just below the halocline is severely increased. The transport of phosphorus to the surface layer is thereby large, and the concentration in the surface layer consequently increases.

4. Mechanical oxygenation of the deep water

It was described above that the sediments in the Baltic Sea immediately can bind 3 tonnes of phosphorus per km2 when oxygenated. If the bottoms are kept oxygenated it is most likely that they can bind another 0.05-0.1 tonnes of phosphorus per km2 and year (long-term sink). If oxygenated bottoms again are subdued to hypoxia the reversed scenario is expected: the binding of phosphorus to the sediments will stop and the bottoms will release 3 tonnes of phosphorus per km2.

Oxygenation of the Baltic Sea deep water

Anders Stigebrandt and Bo Gustafsson are researchers at the Department of Earth Sciences, University of Gothenburg. They have worked out a proposal on how to mechanically increase the deep-water oxygen levels in the Baltic Sea. The proposal is here summarized:

The proposed solution
The proposed solution

The natural tendency of phosphorus binding to oxygenated sediments will be used to decrease the phosphate levels in the Baltic proper (as took place during the period 1992-1997). The oxygenation can be achieved by downward pumping of oxygen-rich water from about 50 meters depth (so called winter water) to about 125 meters depth. The pumps will be driven by wind mills (see figure above).

It is important to combine the pumping with continuous reduction of phosphate outputs from all countries around the Baltic Sea – it can not replace phosphate-reducing programmes.

If the sediments down to 125 meters depth are kept oxygenated the Baltic proper will quickly lower its phosphorus concentrations. The scenario will be similar to ”Nature’s experiment” 1992-1997, and the surface-water effect should be notable only a year or so after the start-up of the pumping.

Stigebrandt and Gustafsson estimate that a transport of about 100 kg of oxygen per second to the deep water is needed. The level in the winter water at 50 meters depth is at least 10 g O2 per m3. Thus, a transport of 10 000 m3 per second will be enough to keep the sediments oxygenated down to 125 meters (Stigebrandt and Gustafsson, 2007).

Figure 8 shows the present circulation between 50 and 125 meters in the Baltic proper (left panel) and the circulation after pumping (right panel). With pumping the halocline will be split into an upper and a lower part, and the Kattegat inflows will thereby reach a little deeper than at present (not shown in figure 8). Calculations show that with 65% efficiency of the pumps 100 MW is needed to run them. The estimated cost of 100 wind mill-driven pumps (1 MW) à 20 million SKR amounts to a total cost of 2 billion SKR.

Figure 8 (flows in km3 per year).

Figure 8: Flow chart before and after pumping starts. All units are in km3 per year.
Figure 8: Flow chart before and after pumping starts. All units are in km3 per year.