ICES CM 2009/G:02 Not to be cited without prior reference to the author

Developments in the shallow coastal ecosystem of the Väike Strait (Baltic Sea): a comparative study with different modelling scenarios

Ülo Suursaar, Robert Aps, Tiit Kullas, Georg Martin, and Mikk Otsmann

Abstract

The Väike Strait in the West Estonian Archipelago has been closed by a 3 km-long road dam since 1896, forming two separate bays. Conditions in the northern part are mainly governed by the more saline Baltic Proper; in the S-part relatively nutrient-rich Gulf of Riga waters prevail. In recent decades, ecological conditions have been deteriorating in the area and acceleration of accumulation and eutrophication has been observed. Reproduction areas of whitefish and herring have been disturbed. According to widespread opinion, as no water exchange occurs through the strait, the existence of the dam is the reason for those adverse changes. Therefore, a question of constructing some bridged openings has risen several times in the past. The aims of the study are to analyse the ongoing changes in the area, and to decide whether re-opening of the strait may help to restore the past situation. The study is based on monitoring and special measurements data in the area. The possible influence of openings were analysed using high-resolution 2D hydrodynamic modelling. Changes in renewal times of water masses were calculated and matter fluxes were studied. The results suggested that most of the past changes have probably occurred due to natural and of larger scale causes, such as isostatic land uplift, regional increase in water temperature, sedimentation and overall eutrophication of the sea. They cannot be undone merely by re-opening the strait. On the other hand, anticipated manifestations of climate change, such as sea-level rise, increase in temperature and storminess, will increasingly affect the ecosystem.

Keywords: water exchange, eutrophication, modelling, spatial planning, climate change, trends, Baltic Sea.

Contact author: Ülo Suursaar Estonian Marine Institute, University of Tartu Mäealuse 10a, Tallinn, 12618, Estonia Phone: +372 6718950, Fax: +372 6718900 [email protected]

1 1. Introduction

The Baltic Sea is a large transition area between limnic and marine conditions, where plants and animals are a mix of marine and limnic species together with some genuinely brackish water forms (HELCOM, 2003; BACC, 2008). Due to the existence of several semi-enclosed sub-basins, distinct ecological subsystems have developed, which maintain their separation either by features of the complex coastline or hydrological fronts. Apparently, the sensitivity to a set of natural and anthropogenic changes is quite different in these different sub-areas. A few dramatic examples of such co-existence of different subsystems can be found in the West Estonian Archipelago Sea (the Väinameri Sea).

The West Estonian coastal sea is a shallow and naturally diverse marine environment. It consists of semi-enclosed sub-basins, a zone of straits, numerous islands and shoals, where large reproduction areas of herring, eel, whitefish, garfish and pikeperch locate.

a)

b)

Figure 1. Map of the West Estonian Archipelago Sea (Väinameri) and the Gulf of Riga with borders of 2D hydrodynamic model domain. The specific (nested) study area of the Väike Strait is marked with “2” (a). Bathymetry of the Väike Strait and its division into sub-areas (“boxes” B1, etc., see also Table 1 for their measures) (b).

2 The aquatic areas of the Archipelago Sea together with a narrow strip of mainland serve as the biosphere reserve of UNESCO-s MBP Program, as well as for Natura 2000.

The relatively narrow and shallow (1-2 m) Väike Strait (“small strait” in Estonian) locates between the islands of Saaremaa and Muhumaa (Fig. 1b). It has been closed by a 3 km- long road dam. Hence, it is actually not a strait any more, but two relatively shallow bays. The construction work of the Väike Strait dam began in 1892 and the opening of the road link between the Saaremaa and Muhumaa islands took place on 27 July 1896. Now, lasting for nearly 10 years already, preparatory work for bridging also the Suur Strait and discussions on its possible environmental impacts are going on. Being the main strait of the Gulf of Riga – Väinameri system (Fig. 1), the Suur Strait ("big strait" in Estonian) has a width of 4 km and maximum depth of 21 m.

In recent decades, the ecological condition is said to be deteriorating in the Väike Strait area, fish stocks are declining, acceleration of accumulation and eutrophication processes have been observed (Porgassaar, 1993; Ecological…, 1994; Ojaveer, 1995; Kotta et al., 2008). According to widespread opinion, as no water exchange between the Gulf of Riga and the Väinameri Sea occurs through the Väike Strait, the existence of the dam should be the main reason for those adverse changes. Therefore, a question of re-opening of the strait by constructing some bridged openings has risen several times in the past: in 1993 (Ecological…, 1994), 1999 (Suursaar et al., 2000), and most recently in 2008. The aims of these studies were to analyse the past changes in the Väike Strait area, and to decide whether the planned holes have the desired effect on water exchange and on restoring the past situation. The paper summarizes our last results on that issue. The second aim of the study is, in a more wide perspective, to illustrate the ecological developments in such a marine area, where separated by an artificial dam, the different parts are governed by different forcing conditions. The conditions in the northern part are more connected to the regional (“marine”) changes in the Baltic Proper as a whole, while the developments in the southern part are more related to the conditions in the more eutrophied Gulf of Riga sub- basin and local pollution.

2. Material and methods 2.1. Monitoring data

The analysis of the past and present situation was based on the data of Estonian national marine monitoring, which is carried out since 1967 (since 1993 by the Estonian Marine Institute), as well as on special measurements in the area (e.g. in 1993, 1996, 2006-2008). The data included nutrient concentrations, oxygen, water temperature and salinity, as well as some hydrobiological variables in the 6 locations (Fig. 2). Some historical background data from the Marine Station of the Estonian Marine Institute at Pärnu were used as well (see e.g. Suursaar and Tenson, 1998; Suursaar et al., 2005; Kotta et al., 2008).

According to the Estonian Water Policy Directive (and the EU Water Framework Directive), the main indicators for seawater quality are phytoplankton biomass and Chl a, and the secondary parameters are nutrients, zooplankton and status of the bottom vegetation. In 2006 and 2008, correspondingly, altogether 33 and 28 samples were collected for phytoplankton, Chl a, and nutrients (Ntot, Ptot, NO 3+NO 2, PO 4, SiO 4) in the study area. The samples covered the ice-free period from May to October. The sampling depth was 1 m and the analysis methods followed the HELCOM standards in the fully accredited biological laboratory of the Estonian Marine Institute. Bottom vegetation and

3 zoobenthos was studied in 13 locations in 2008. In addition, data on fish catches were analysed within the ICES subdivisions 28 and 29, small quadrants No. 245 and 172 (see also Martin, 2008; Vetemaa et al., 2006).

VV 1

VV 2

VV 3

VV 4 VV 5 VV 6

Figure 2. The locations for hydrochemical and biological monitoring in 2006-2008.a.

2.2. Hydrodynamic modelling

The basic hydrodynamic properties, as well as the possible influence of holes were analysed by means of several relatively simple hydrodynamic models. They included, firstly, a 2D hydrodynamic model with the grid step of 1 km. It has 16 405 marine points in the Gulf of Riga and 2510 points in the Väinameri (Fig. 1a). The depth-averaged free- surface model is based on the so-called shallow sea equations. In our case it is forced by the wind data from the Vilsandi meteorological station and by open boundary sea level variations according to the tide gauges of Ristna or Sõrve. The model provides both the sea level variations and depth-averaged currents in the grid points (see e.g. Suursaar et al., 2001; Suursaar and Kullas, 2006, for more details and some previous applications).

Further on, especially for this study, a high-resolution Väike Strait model grid (2822 points) with a spatial resolution of 200 m was nested within the coarser grid. Water fluxes were allowed through the borders of the model. The final simulations of currents and water exchange were carried out within this small modelling domain (Fig. 1b). For facilitating the interpretation, water fluxes, water exchange and so-called residence times of water were calculated for different realistic and simplified forcing conditions, as well as for some dam-opening options within the six designated sub-areas (or “boxes”, Fig. 1b, Table 1). The dam is “closed” for the present situation and entirely “open” for the past or one marginal state. One point-size opening with the depth of 0.2 m represents the originally planned two 1 meter-deep “actual” holes with the width of 20 m (as they were initially planned for the dam).

3. Results and discussion 3.1. Current ecological conditions in the Gulf of Riga and in the Väike Strait

The north-western half of the Väike Strait (or the NW-bay, to be more precise) has a length of about 11 km and an area of 38 km² (Table 1). The SE-bay has the inner and the outer part. The inner (8 x 3 km) basin is very shallow (average depth 0.6 m). The outer part (6 x

4 5 km) is deeper (about 3 m) and the volume is nearly 15 times larger than that of the inner part of the SE-bay.

The study area is practically tideless. However, the meteorologically forced sea level variations may reach up to 3 m (Suursaar and Sooäär, 2007). Being differently exposed, the sea level is different on both sides of the dam for most of the year. Also different are the hydrophysical and -chemical regimes of the study area, which are determined by the conditions of the Väinameri Sea and the Gulf of Riga, respectively.

The Gulf of Riga has an area of 17 913 km² and a average depth of 23 m, while the Väinameri has an area of 2243 km² and a depth of 4.7 m. Due to differences in water exchange, as well as in riverine and nutrient inflows, the trophic status of these sub-areas is not equal. While the salinity is just somewhat higher in the Väinameri (typically 6–6.7 against 5–6.2 in the northern part of the Gulf of Riga), the difference in nutrient concentrations, especially in Ntot, is far more remarkable (Fig. 3; Ojaveer, 1995; Martin et al., 2008; Martin, 2009). The differences in phytoplankton biomass and Chl a are remarkable well (Figs. 3,4).

According to the long-term monitoring data, in the SE-bay the relatively nutrient-rich Gulf of Riga waters prevail. In the Väinameri, including the NW part of the study area, the influence of more saline and nutrient poor water masses of the Baltic Proper is more pronounced. Typical nutrient concentrations in the Väinameri are 0.2–0.3 µM P-PO 4, 0.3– 0.5 µM Ptot, 0.1–0.6 µM N-NO 3 and 10–15 µM Ntot versus 0.3–0.5 µM P-PO 4, 0.4–0.7 µM Ptot, 0.5–1.3 µM N-NO 3 and 15–25 µM Ntot in the northern part of the Gulf of Riga in autumn in roughly similar conditions (Ojaveer, 1995; Porgassaar, 1993). In summer 1993 the average concentration of total phosphorus was 0.74 µM in the NW-bay versus 1 µM in the SE-bay. Respective values for the total nitrogen were 33 µM vs. 44 µM (Ecological…, 1994). In 2008, the values for Ptot were 1.1 times and for Ntot 1.8 times higher in the SE part compared with the NW-bay, the difference could easily be up to 2–3 folds on different sides of the dam (Figs. 3,4).

Nowadays, ecological status in the Estonian coastal waters is assessed within designated sub-areas (in relation to graduation, which is slightly different in each of the sub-area). The main input for the procedure is data on phytoplankton and Chl a, and secondary ones are nutrient concentrations (so called Water Quality Elements). According to that classification, the conditions in the S part are mostly “moderate” (3) or “poor” (2), and “good” (4) in the northern side of the dam (Martin, 2009).

Thus, the dam replaces the hydrochemical front, which in natural conditions pulsates somewhere between the Suur and Hari Strait (Suursaar and Kullas, 2006), depending on the prevailing winds and currents (Fig. 5). The “natural” part of the front is rather smooth, while the portion of the front, which is represented by the artificial dam, is sharp and practically in a fixed position.

According to our flow and matter exchange measurements in 1993–1996, as well as water exchange modelling (Suursaar et al., 2001), the annual sum of southward and northward fluxes through the Suur Strait is about 120–140 and 150–170 km 3 respectively. The positive freshwater balance of about 30 km 3 appears due to riverine inflow. The flows in the straits vary depending mainly on wind direction and compensating each other in order to maintain (together with sea level variations) water balance. The hypothetical role of the

5 Väike Strait prior to its closure in 1896 was small. The water fluxes in this shallow strait would have been about 10–20 times smaller than that of the Suur Strait. Thus, the Väike Strait would not influence the ecological conditions of the Gulf of Riga or the Väinameri. But the existence or absence of the dam obviously influences the Väike Strait itself.

Figure 3. Average concentrations of nutrients (N tot , P tot ), Chl a, and phytoplankton biomasses in the northern (N) and southern (S) parts of the Väike Strait (VV). VV´06 and VV’08 represent the averages for the whole Väike strait in 2006 and 2008, respectively. The plot represents the 25-75% percentiles, as well as median and extreme values. The green lines indicate the type-specific borders between good and moderate water quality (Martin, 2009).

Figure 4. Phytoplankton biomass in the northern stations (w1,w2,w3; see also Fig. 2) and in the southern ones (w4,w5,w6) in 2006 (blue bars) and in 2008 (red bars) (Martin, 2009).

6

Figure 5. Modelled snapshot of current velocities showing fast currents in the Suur Strait, but lack of fluxes through the Väike Strait (indicated with arrow). N wind 12 m/s.

3.2. Observed changes over the last century

Although Estonian marine monitoring began in the 1960s, the long-term data on the study area is very limited. Away from large pollution point sources, the region had low priority status for monitoring activities.

In the Gulf of Riga, from the mid 1970s to 1991 winter nitrate concentrations increased significantly, followed by a gradual decrease from 1992 onwards (Suursaar, 1994; BACC 2008). Winter phosphate concentrations increased continuously from 1981-1998. The periods of rising and falling nitrate concentrations coincide with corresponding changes in river runoff (BACC, 2008). This indicates that the river input (as a manifestation of climate variability and change) is one of the main driving factors of the long-term dynamics of the nitrogen pool in the Gulf of Riga and the corresponding biological response (Dippner and Ikauniece, 2001). During the last 40 years in the Baltic Sea, the wintertime phosphorus levels have increased eight-fold and the nitrogen levels four-fold (e.g. Elmgren, 2001). The accumulation of nutrients occurred mainly in the deep layers of the sea, as well as in the sub-basins (including the Gulf of Finland and Gulf of Riga). Thus both anthropogenic, natural and “semi-natural” (e.g. climate regime shifts) causes should be considered.

The calculated heat content of the Baltic Sea according to Schrum et al. (2003) and the local sea surface temperatures in the northern part of the Gulf of Riga (Fig. 6) show rather synchronous variations with maximums around 1975, 2990 and 2002, and the last minimum in 1993-95. The overall trends for water temperatures are increasing and decreasing for salinity (e.g. BACC, 2008).

Although Estonia still experiences a weak (0-2.5 mm/yr) postglacial uplift, the sea level in most of the Estonian tide gauges is increasing, especially in annual maximum sea level values (Figs. 7,8), which reflect increasing westerly wind component and vitalizing storminess above the Baltic Sea (e.g. Suursaar and Kullas, 2006).

7 The influence of the documented forcing shifts and trends in regional (Baltic) scale, such as increase in water temperature, nutrients, storminess, and mixing processes, and decrease in salinity (e.g. BACC, 2008) is overlapped with local conditions. In the study area, they include local pollution and eutrophication, as well as recovery from past anthropogenic pressure (Kotta et al., 2008; Suursaar et al., 2009).

Some evidence of local pollution was found near Orissaare already in 1970s–1990s (Ecological…, 1994). In 1998 a new wastewater treatment plant has been built in Orissaare and there should be no remarkable local pollution anymore. According to investigations of phytoplankton, shore and benthic vegetation, the eutrophication is more evident in the SE- bay (Martin, 2008). Macrophytes and mud appear in the shallow parts of the sea and in the vicinity of the dam. The spawning grounds of fish, especially of whitefish, have been disturbed and fishery catches from this area have decreased (Ecological…, 1994). These developments are in general of local character and pelagic zone is not suffering more than the Gulf of Riga as a whole.

18 y = 0.0016x + 10.78 16 May-June

14

12 y = 0.0175x - 27.71 10 April-May 8 Temperature(deg.C) 6

4 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Year Figure 6. Decadal variations in water temperature in the upper layer of the northern part of the Gulf Riga according to the annual data samples grouped by different months.

60

40

Pärnu + 40 cm 20 Ristna

Sealevel (cm) 0

-20 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 Year

Figure 7. Decadal variations in annual mean relative sea levels (until 2008) together with 11-year moving averages and linear trendlines at Pärnu and Ristna tide gauges. The series are not corrected with land uplift rates, which are different in different gauges, including maximum (2.5 mm/yr) uplift at Ristna.

8 250 300 . . 200 250 Pärnu 150 200

100 150

Pärnu sea level (cm) level sea Pärnu 50 100 Ristna levelseaRistna (cm) 0 50 1940 1950 1960 1970 1980 1990 2000 2010 Year

Figure 8. Decadal changes of annual maximum sea levels at Ristna and Pärnu.

In recent years a sub-surface sandy bank has risen in the central part of the inner bay. In the case of low sea level it may emerge as a shoal, leaving free water either near the coasts of Saaremaa or Muhumaa.

The main reason for that could be land uplift, sedimentation and siltation due to local eutrophication. The original dam was made of local building materials and evidently infiltration through the dam took place. After the reconstructions in 1990–1996 the dam was made higher and much more solid. As a result, during the next ten years the eutrophication has been allegedly accelerating. The appearance of the shoal in the middle of the SE-bay has also been associated mainly with the new and "thick" dam. Actually, these developments could, firstly, be in relation with the intensifying of the eutrophication processes around the whole marine coastal region of Estonia (Ojaveer, 1995; Martin, 2008). Though during the last 5–10 years the overall pollution load has decreased in Estonia, the monitoring data on nutrients still shows large natural variability, which does not allow any straightforward conclusions. Also, reaction of the marine environmental processes takes place with a time-lag and in most cases it is difficult to point out a certain reason.

Secondly, the Archipelago sea area has experienced post-glacial isostatic land uplift with a rate of about 2–2.5 mm/yr (Vallner et al., 1998). Due to this natural cause alone, the Väike Strait became shallower by about 20–30 cm over the last 110 years. As the average depth of the strait is below 1 m, this could have had a substantial effect on the area. An additional effect of local sedimentation and siltation could be even more pronounced in certain places (Lutt, 1985). However, the last estimations put the climate change-induced global sea level rise rate at about 1.7 mm/yr (Church and White, 2006; Suursaar and Sooäär, 2007), which now already roughly compensates for the land uplift in the Väike Strait area. Most likely the sea level rise accelerates and will exceed the uplift during the 21th century (Jylhä et al., 2004).

3.3. Possible effects of reopening of the strait: different scenarios

The flow simulations showed that the currents in the study area occur as a combination of wind-driven circulation and sea level variation-induced back and forth movements of water. The latter role is between 12 and 44% (Table 1). Despite the absence of tides, the variable wind creates sea level fluctuations (Fig. 9) of about 10–20 cm per day, which captures about 20% of the average water volume of the shallow B3 and B4 (Fig. 10).

9

The maximum current velocity (as simulated with 20 m/s sustained wind speed) may reach 80 cm/s only in the B4, being 25–40 cm/s in the rest of the areas. Still, the depth-averaged flows (Figs. 10,11) are larger in deeper boxes, especially in B6. Surprisingly, the maximum velocities increased in the “without dam” (WD) scenario two-fold in B3, B4 and B5, but just about 20% in B1, B2 and B6. Apparently, larger increase in circulation is prohibited by the shallowness of B3 and B4, which form a “bottle neck”. Even in the full-open Väike Strait, the major part of the circulation would still take place in the mouth sections (B1, B6) and the transit turnover of water would not increase that much, as one can expect.

Figure 9. The modelled horizontal distribution of the wind directions that produce the highest sea level of a location.

Figure 10. Modelled depth averaged water fluxes and local sea level isolines in case of 7 m/s NE (upper figure) and SE winds (lower figure).

10 As a control run (CR, Table 1, Figs. 10a, 11), we calculated the average water exchange (representing all the realistic winds in an annual cycle) of the sub-areas. Individual flows can both enter and leave through the box interfaces. The water exchange (m 3/s) was defined as a sum of flows entering a box through one (i.e. the left-hand side) border and leaving through another (right-hand side, Fig. 11) border. The residence time of water was calculated as an average period when all the water in a box would be replaced by the water from the neighbouring areas.

a) b)

Figure 11. An example of calculations of water exchange within the sub-areas for one particular wind event. The sceanrio with closed dam (a, the control simulation) and with holes (b). The summary of the whole set of simulations is presented in Table 1.

Table 1. Characteristics of the sub-areas (“boxes”, see also Fig. 2) and the results of water exchange calculations as control simulation (CS), with two openings in the dam (2O) and without the dam (WD).

Characteristic Boxes: in total 0.23 km 3, area 113 km 2 B1 B2 B3 B4 B5 B6 Box water volume (km 3) 0.048 0.010 0.005 0.004 0.006 0.15 Box surface area (km 2) 23.5 7.5 8.3 11.4 9.3 52.9 Average depth (m) 2.1 1.3 0.6 0.4 0.7 2.9 Water exchange CS (m 3/s) 365 133 65 60 104 951 Role of sea level variations (%) 14 13 29 44 20 12 Water exchange 2O (m 3/s) 365 132 69 67 106 950 Water exchange WD (m 3/s) 360 142 118 120 131 932 Residence time CS (d) 1.53 0.84 0.88 0.85 0.68 1.85 Residence time 2O (d) 1.53 0.86 0.82 0.75 0.67 1.85 Residence time WD (d) 1.56 0.79 0.48 0.42 0.54 1.89 Relative res.time CS (d/km 3) 32 87 179 195 112 12 Relative res.time 2O (d/km 3) 32 88 167 172 109 12 Relative res.time WD (d/km 3) 32 82 98 97 88 12 Change CS -> 2O (%) -0.1 -2.0 6.5 11.5 2.4 -0.1 Change CS -> WD (%) -1.7 5.6 45.2 50.2 20.8 -2.1

11

Figure 12. A sketch illustrating the mixing of different water masses as a result of water exchange through the holes in the dam; the influence is very small and it is limited with the areas next to the dam.

It appeared that this period is between 0.7 and 1.9 days only (Table 1). Quite a busy water turnover proceeds even in the fully closed strait and by no means the area can be regarded as a stagnant one. The main explanation is the shallowness. The relative residence time (per 1 km 3) is up to 200 days, however. The same calculations were performed also with two openings (2O) as well as for the full open (WD) strait (Table 1) and between the scenarios the relative changes in the water exchange were found. That change included an increase by 6–12% only in B3 and B4, and virtually no changes in the other sub-areas of the Väike Strait.

Obviously, the initially planned 2 x 20 m openings are too small to have any considerable effect for the matter exchange between the Gulf of Riga and the Väinameri, or even for the Väike Strait as a whole. Even if the velocities in the holes may reach 150–200 cm/s, the fluxes cannot level out the sea level differences (Figs. 9,10), which appear due to wind- generated pile-up and which tend quickly to turn around due to natural wind variability.

Bearing in mind the mainly rapidly changing back and forth flows, the increase in water exchange by 6–12% (Table 1) means that as a result of openings the water masses in B3 and B4 will be mixed with each other according to that proportion. I.e., if the ingredient concentration were respectively 1 and 2, they will become more like 1.1 and 1.9. The change in other boxes will be negligible. Thus, the water quality will slightly improve in the SE part of the strait and slightly deteriorate in the NW side of the dam, but such changes will be visible only within a few kilometres from the dam. However, the water quality classes in both sides of the dam will remain unchanged, even if the assessment coefficients change somewhat (Martin, 2009).

Another important question is whether the increased flow velocities will increase resuspension by acting on the seabed? The present-time velocities (up to 30 cm/s, 80 cm/s in B4) can occasionally re-suspend the bottom sediments as the critical bottom shear of 0.5 dyn/cm 2 for fine sand is exceeded in velocities over 20 cm/s (Suursaar et al., 2001; Huttula, 1994). Still, the increase in flow speed counts for about 10% in 1–2 km from the holes and much less in more remote areas. According to our previous wave modelling studies (Suursaar and Kullas, 2009), much higher resuspension occurs as a result of orbital

12 velocities of waves, although the wave growth is limited by small depth and restricted fetches in the Väike Strait. Therefore the intensity of resuspension depends more on natural changes in wind speed and storminess, and less on the holes in the dam. We admit, however, that due to occasional large velocities in the openings, there may be some noticeable effect in the vicinity (say, within 50–100 m) of the openings, but the influence diminishes roughly proportionally with the second power of the distance.

3.4. Possible future developments due to climate change and local factors

The trophic status and ecological conditions will still mainly depend on the overall state of the Gulf of Riga, in one side, and that of the Väinameri, in the other side. Hence, the global climate change and sea level rise, as well as regional warming of the Baltic Sea will be the key factors for future developments. In addition, reducing regional and local pollution and hence, trophic status of the Gulf of Riga would help especially in the southern side of the dam.

The local changes as a result of openings would be negligible. According to present wind and sea level statistics, the NW flows would be slightly prevail, yielding about 0.1 km 3/yr resulted NW flow (Suursaar et al., 2000). The corresponding matter fluxes are not remarkable, as compared e.g. to the resulted annual northward flow of about 30 km 3 through the Suur Strait. Statistically, as the annually cumulated flux of water will be directed from the Gulf of Riga to the Väinameri, correspondingly about 1000 t of resuspended sediments will be carried through the openings to the NW section of the strait, as well as some 0.3 t of Ptot and 10 t of Ntot. However, the resulting flow and matter flux direction varies with year. According to the long-term changes in regional wind regime, the annual average wind directions turn from SW to WSW, yielding less NW and more SE flows.

As it was also pointed out before, due to anticipated acceleration of global sea level rise (Jylhä et al., 2004; Suursaar and Kullas, 2006; Church and White, 2006), the Väike Strait will not turn any shallower any more, but likely the average depth of that very shallow area will increase. This will ease some (seeming) problems of silting up and congestion of the strait, and more that than the openings would do.

4. Conclusions

1) Conditions in the southern part of the Väike Strait are mainly governed by the relatively nutrient-rich Gulf of Riga waters. In the N part, more saline and less trophic waters of Baltic Proper prevail. The 3 km road closing the strait replaces the hydrochemical front. Thus, aspects of Baltic Proper, Gulf of Riga and local conditions overlap in the study area.

2) Water renewal as a result of wind-driven currents and water level fluctuations is relatively intense in the Väike strait area, but mainly due to shallowness. The planned openings have a minor positive effect on water renewal or nutrient exchange. The water exchange, depending on the scenario, will increase only by about 6–12%, and yet in the zones nearby the dam. Even if the size of the strait looks just about a half of that of the Suur Strait, it is much shallower, the flows are hampered by large friction and the water exchange in the openings would be about 300 times smaller than in the Suur Strait.

13 3) Most of the past changes in the Väike Strait have probably occurred either due to natural or of larger scale causes, such as isostatic land uplift, natural sedimentation and overall eutrophication of the sea. As it is common with natural system, you can’t turn back the time – e.g., merely by constructing openings, or even by entirely re-opening the strait.

4) In the nearest future, global warming, the anticipated increase in storminess and accelerating sea level rise will become decisive factors also for the Väike Strait. In the southern side of the dam, the ability with coping regional pollution problems for reducing trophic level of the Gulf of Riga may improve the ecological conditions a lot. The conditions in the northern part are more governed by the conditions in the Baltic Proper as a whole, and cannot be managed on the local level only.

Acknowledgements

The study was supported by the ESF grant 7609, the Väike Strait project (2008–2009) and the ERDF BaltSeaPlan Project.

References

Church, J.A. and White, N.J. 2006. A 20th century acceleration in global sea-level rise. Geophys. Res. Lett., 33, L01602. Dippner, J.W. and Ikauniece, A. 2001. Long-term zoobenthos variability in the Gulf of Riga in relation to climate variability. J. Mar. Sys ., 30, pp. 155–164. Elmgren, R. 2001. Understanding human impact on the Baltic ecosystem, changing views in recent decades. Ambio , 30, pp. 222–231. Ecological studies in the Aquatic Environment of Väike Väin Strait in West Estonia . 1994. Estonian-Finnish cowork during summer 1993. Report. Helsinki-Kuressaare. HELCOM, 2003. The Baltic marine environment 1999–2002. Balt. Sea Environ. Proc ., 87, Helsinki Commission: Helsinki. Huttula, T. 1994. Suspended sediment transport in Lake Säkylän Pyhäjärvi. Aqua Fennica, 24, pp. 171–185. Jylhä, K., Tuomenvirta, H. and Rusteenoja, K. 2004. Climate change projections for Finland during the 21st century. Boreal Environment Research, 9, pp. 127–152. Kotta, J., Lauringson, V., Martin, G., Simm, M., Kotta, I., Herkül, K. and Ojaveer, H. 2008. Gulf of Riga and Pärnu Bay. In: Schiever, U. (ed.), Ecology of Baltic coastal waters . Berlin, Springer, pp. 217–243. Lutt, J. 1985. Bottom sediments of the Väinameri . Valgus: Tallinn, 239 pp. [in Russian]. Martin, G. (ed.) 2008. Monitoring of the coastal sea in 2007 . Report. Estonian Marine Institute, Ministry of Environment: Tallinn, 113 pp. http://eelis.ic.envir.ee:88/seireveeb/aruanded/8070_ylevaateseire.doc. Martin, G. (ed.) 2009. Possibilities for improving the ecological state of the Väike Strait area . Report. Manuscript in the Estonian Marine Institute. Tallinn. (in Estonian). Martin, G., Jaanus, A., Põllumäe, A. and Kotta J. 2008. Monitoring of the coastal sea. Estonian environmental monitoring 2004 –2006, ed. K. Väljataga, Ministry of Environment: Tallinn, pp. 72–76. Ojaveer, E. (ed.) 1995. Ecosystem of the Gulf of Riga between 1920 and 1990 . Estonian Academy Publishers: Tallinn, 277 pp.

14 Otsmann, M., Suursaar, Ü. and Kullas, T. 2001. The oscillatory nature of the flows in the system of straits and small semienclosed basins of the Baltic Sea. Continental Shelf Research , 21, pp. 1577–1603. Porgassaar, V. 1993. Content and distribution of phosphorus and nitrogen in the coastal waters of West Estonia. Proc. Estonian Acad. Sci. Ecol ., 3, pp. 166–180. Schrum, C., Hübner, U., Jacob, D. and Podzun, R. 2003. A coupled atmosphere/ice/ocean model for the North Sea and the Baltic Sea. Clim. Dyn ., 21 pp. 131-151. Suursaar, Ü. 1994. Estonian marine monitoring 1968-1991: Results and evaluation. Finnish Marine Research , 262, pp. 123–134. Suursaar, Ü., Aps, R., Kotta, I. and Roots, O. 2009. North-East Estonian coastal sea: recovery from the past anthropogenic pressure and new stressors on the background of natural variability. In: WIT Transactions on Ecology and Environment , Vol. 122, pp. 331–342. Suursaar, Ü., Astok, V. and Kotta, J. 2005. Monitoring and assessment of the Estonian coastal sea in 1945–2003. Water protection of the Gulf of Finland and Estonian waterbodies (1945–2003) , ed. H.-A. Velner, TUT Press: Tallinn, pp. 129–144. Suursaar, Ü. and Kullas, T. 2006. Influence of wind climate changes on the mean sea level and current regime in the coastal waters of west Estonia, Baltic Sea. Oceanologia , 48, pp. 361–383. Suursaar, Ü. and Kullas, T. 2009. Decadal variations in wave heights off Cape Kelba, Saaremaa Island, and their relationships with changes in wind climate. Oceanologia , 51, pp. 39–61. Suursaar, Ü., Kullas, T. and Otsmann, M. 2001. The influence of currents and waves on ecological conditions of the Väinameri. Proc. Estonian Acad. Sci. Biol. Ecol ., 50, pp. 231–247. Suursaar, Ü., Otsmann, M. and Kullas, T. 2000. Exchange processes in the Väike Strait (the Baltic Sea): present, past, future. Proc. Estonian Acad. Sci. Biol. Ecol., 49, pp. 235–252. Suursaar, Ü. and Sooäär J. 2007. Decadal variations in mean and extreme sea level values along the Estonian coast of the Baltic Sea. Tellus A , 59, pp. 249–260. Suursaar, Ü. and Tenson, J. 1998. Hydrochemical regime and productivity of the Pärnu Bay in 1968-1996. In: EMI Report Series , 9, pp. 91–117. Vallner, L., Sildvee, H. and Torim, A. 1988. Recent crustal movements in Estonia. J. Geodyn ., 9, pp. 215–223. Vetemaa, M., Eschbaum, R., Verliin, A., Albert, A., Eero, M., Lillemägi, R., Pihlak, M. and Saat, T. 2006. Annual and seasonal dynamics of fish in the brackish-water Matsalu Bay, Estonia. Ecology of Freshwater Fish , 15, pp. 211–220.

15