Impact of water salinity on long-term dynamics and spatial distribution of benthic invertebrates in the Small Aral Sea

The Aral Sea region occupies the northernmost location in the continental subtropical climate zone (Zhitomirskaya 1964). The area receives a large amount of solar heat, equal to an average of 5860 MJ m⁻² per year. The average annual precipitation is 100–115 mm. It is characterized by a pronounced intra-annual air temperature difference, ranging from +43°C in summer to −36°C in winter.

The two largest rivers of Central Asia – the Amu Darya and the Syr Darya – flow into the Aral Sea. The length of the Amu Darya River is 1415 km. The area of its basin is 309 000 km²(Kosarev 1975). The Amu Darya River is formed by the confluence of the Panj and the Vakhsh rivers. It is fed by snow and ice. The river flows into the Large Aral Sea. In recent years, water from the Amu Darya River does not reach the sea for most of the year. The Syr Darya River flows into the southern part of the Small Aral Sea. It is predominantly fed by snow and, to a lesser extent, by glaciers and rainwater (Armstrong et al. 2018). The length of the Syr Darya River is 2212 km. It is formed by the confluence of the Naryn and the Kara Darya rivers in the eastern part of the Fergana Valley (Uzbekistan). The basin area is 219 000 km².

The total long-term average annual river runoff to the Aral Sea reached 56.0 km³ in 1911–1960 (UNESCO 2017). In 1961–1980, it decreased to 30.00 km³ and in 1981–1990, it reached a minimum of 3.45 km³. The average runoff of the Amu Darya River for 1992–2014 amounted to 9.04 km³, ranging from 0.40 km³ in 2001 to 17.6 km³ in 2005. The runoff volume of the Syr Darya River for the same period was on average 5.96 km³, ranging from 10.3 km³ during high-water years (2004–2005) up to 2.5 km³ in dry 2000 (Gaybullaev et al. 2012).

The Aral Sea is located in the Turan Lowland near the eastern edge of the Ustyurt Plateau in the territory of Uzbekistan and Kazakhstan. Its geological history is characterized by gradual isolation from the ocean, which resulted in desalination and changes in the chemical composition of water (Alekin & Lyakhin 1984). Until the second half of the 20th century, the Aral Sea was the largest brackish water body in Central Asia.

At the absolute level of 53.0 m, the Aral Sea water area reached 66 000 km², the maximum depth was 69.0 m and the average depth was 16.1 m (Bortnik & Chistyaevaya 1990; Micklin 2014). The coastline was 4430 km. The greatest length of the sea was 424 km, with a width of 292 km. The salinity of water varied insignificantly, with an average value of 10.2 PSU (Alekin & Lyakhin 1984). Chlorides, sulfates and sodium dominated in the chemical composition of water.

The complex of anthropogenic and natural factors caused a significant reduction in the water surface area and progressive salinization of the sea (Andreev 1999). By 1988–1989, the sea was divided into two isolated parts: the northern one, i.e. the Small Aral Sea, and the southern one, i.e. the Large Aral Sea (Aladin et al. 1998; Gaybullaev et al. 2012; Micklin 2014). Since 1988–1989, succession of the northern and southern parts of the sea has progressed independently of each other.

At the level of 53.0 m BS (Baltic System), the area of the Large Aral Sea reached 60 090 km² and the volume of water was 984.00 km³ (Bortnik & Chistyaevaya 1990). In the period from 1986 to 2004, the area of the Large Aral Sea decreased from 38 560 to 16 400 km², the volume of water was reduced from 380.63 to 93.46 km³ (UNESCO 2017). By 2007, at the absolute level of 29.50 m, the Large Aral Sea was divided into western and eastern parts (Aladin et al. 2008). The area of the Western Aral Sea was 4450 km² and the volume of water was 19.76 km³. The area of the Eastern Aral Sea was 7030 km² and the volume of water was 49.5 km³ (UNESCO 2017). By the end of the 1990s, the salinity of the Large Aral Sea exceeded 130–150 PSU (Aladin & Plotnikov 2008).

At the level of 53.0 m BS, the area of the northern part of the sea was 5990 km² and the volume of water was 79.7 km³ (Bortnik & Chistyaevaya 1990). From 1986 to 2004, the area of the Small Aral Sea varied from 2090 to 3240 km², with the water volume of 12.03–27.03 km³ (UNESCO 2017). To restore the Small Aral Sea, a sand dam was constructed in the Berg Strait in 1992 to prevent the runoff of the Syr Darya River into the Large Aral Sea (Aladin & Plotnikov 1995). After its destruction in 1993, another temporary dam was built in the spring of 1997. It existed for about two years (Aladin et al. 2005). In 2005, a permanent dam was built, designed to fill the northern part of the sea up to 42.2 m. By 2007, the area of the Small Aral Sea increased to 3280 km² and the water volume increased from 22.52 to 26.33 km³ as a result of the accumulation of the Syr Darya River runoff. As the level of the Small Aral Sea increased to 42.0 m between 2006 and 2008, the average salinity of water dropped to 12.9 PSU (Krupa & Grishaeva 2011).

Samples of macrozoobenthos were collected in the summer of 1996–1998 and 2001–2008. A standard grid of 20 locations covered the entire water surface area of the Small Aral Sea, except for the partially dry Saryshyganak Bay (Fig. 1). A small Petersen bottom grab with a sampling area of 0.025 m² was used to sample the macrozoobenthos. At each station, two samples of macrozoobenthos were collected. Samples were fixed with formaldehyde solution (10%) or solution of ethyl alcohol (70%). A total of 320 macrozoobenthos samples were collected and processed. Physicochemical properties of water at the sampling locations were measured simultaneously with macrozoobenthos sampling. Water temperature was measured at each location using a Hanna HI 98129 portable meter. Water transparency was determined by a Secchi disk. The dominant type of soil was determined by the ratio of fractions in bottom grab samples. To determine the chemical composition and salinity, water samples were collected in plastic containers with a volume of 1 l.

Figure 1

Conventional methods of chemical analysis of water were used (Semenova 1977). Water samples were analyzed in three to four replications. The error of estimate for main ions in water was 0.5–5.0%, depending on the analysis.

Macrozoobenthos samples were processed according to currently accepted methods (Barulin 1984). The taxonomic composition of benthic invertebrates was determined using MBS-10 and MS-300 microscopes according to Shilova (1976), Pankratova (1977), Tsalolikhin (1995), Bogutskaya et al. (2013). The number of individuals of each species was calculated in each sample. To determine the biomass, specimens were weighed on torsion (WT-1000, from 0 to 1000 mg) and electronic balances (OHAUS Adventurer TM AR5120, up to 510 g). Further, the abundance and biomass of each species and of the total macrozoobenthos were calculated per 1 m² of the seabed of the water body.

Average values with a standard deviation were calculated for all variables in Excel. To analyze the biomass distribution of dominant species of benthic invertebrates in a salinity gradient, scatterplots were constructed. Long-term data were used for this purpose. Spearman correlation coefficients (R) between biological variables and salinity were calculated using the Statistica 12.0 software, with p < 0.05. With regard to the coordinates of the sampling locations, distribution maps of the analyzed variables were drawn using the same software. For this purpose, we used data from 1997, 2001, 2006, 2007 with different average salinity of water. Next, the outlines of the Small Aral Sea were drawn in Adobe Photoshop based on Google maps. For better visibility, statistical maps of the analyzed variables were combined with a contour map of the sea in the same software. All statistical maps show the gradient of the measured variables in color.

The Small Aral Sea has a highly dissected coastline that forms three large bays – Shevchenko in the west, Butakov in the north and Saryshyganak in the northeast (Fig. 1). During the study period, the average depth of the Small Aral Sea was 5.0 m, ranging from 2.6 to 11.5 m. The water transparency varied from 0.8 to 1.8 m. Water temperature ranged from 17.5 to 22.4°C. The western part of Shevchenko Bay as well as the central and northeastern parts of the water area were characterized by the greatest depths and transparency of water (Fig. 2). The minimum values of all variables were recorded in the impact zone of the Syr Darya River. The dominant types of soil were dark gray silt and sand.

Figure 2

The average salinity of water varied considerably over the years (Table 1). The maximum values of this variable were observed in the summer of 1996, 1997, 2001, while the minimum ones – in the spring of 2006–2007. In all the years, the minimum values of salinity were recorded in the impact zone of the Syr Darya River. Shevchenko and Butakov bays, located far from the impact of the river runoff and with a slow water exchange, were characterized by the highest content of total dissolved solids in water.

Ten species of benthic invertebrates were identified over a period of 10 years. The polychaete Hediste diversicolor Muller as well as the mollusks Abra ovata Philippi, Cerastoderma glaucum (Bruguière) and Caspiohydrobia sp. were constant in the benthic community. Other species of benthic invertebrates [oligochaetes, chironomids Procladius ferrugineus Kieffer, Chironomus behningi Goetghebuer, mollusk Theodoxus pallasi Lindholm, shrimp Palaemon elegans Rathke, mysid Paramysis (Mesomysis) lacustris Czerniavsky] were rare and were recorded mainly in desalinated areas of the sea.

The long-term average annual abundance of macrozoobenthos was equal to 1962±624 specimens m⁻², with the biomass of 107.1 ± 20.6 g m⁻². During the study period, the maximum quantitative variables of benthic invertebrates were observed in the western region of the sea – Shevchenko Bay (Table 2). In general, the desalinated southern part of the water area, which is affected by the Syr Darya River flow, was characterized by the minimum biomass of macrozoobenthos. The persistently low biomass of benthic invertebrates in this region (except for 2001) was due to the absence of large mollusks in the community. The eastern and central regions of the sea were characterized by intermediate values of the macrozoobenthos biomass. From 1996 to 2008, the quantitative variables of macrozoobenthos decreased by an order of magnitude for all surveyed areas of the sea. The correlation between the water salinity and the biomass of the benthic community was positive and statistically significant at R=0.75 and p<0.05.

The mollusk Abra ovata was the dominant taxon in the macrozoobenthos – on average 55.4% of the abundance and 64.1% of the biomass of the community. In terms of abundance, the polychaete Hediste diversicolor (24.6%) sub-dominated and was followed by the mollusk C. glaucum (20.4%). Relatively high abundance and biomass values of the mollusk Caspiohydrobia sp. were recorded in particular years. During the study period, the abundance and biomass of the dominant species of macrozoobenthos changed simultaneously (R = 0.89–0.92, p < 0.05). Therefore, only the biomass values of benthic invertebrates were used for further analysis.

The highest biomass values of the dominant invertebrates were recorded in the period before 2003 (Fig. 3). In the following years, the biomass of all species decreased by an order of magnitude. Since 2004, the mollusk Caspiohydrobia sp. has almost disappeared from the benthic community.

Figure 3

The spatial distribution of benthic invertebrates was analyzed for 4 years (1997, 2001, 2006 and 2007) characterized by different average salinity (Table 1). In 1997 and 2001, high values of salinity were recorded almost throughout the entire area of the Small Aral Sea, except for the southernmost part near the mouth of the Syr Darya River. As the water level increased in 2006 and 2007, the area of high salinity zones decreased.

Statistical mapping (Fig. 4) showed that the accumulation of the mollusk Abra ovata was recorded mainly in Butakov Bay. In 2007, in the conditions of strong desalination of the entire sea area, the A. ovata concentration was also recorded in the western part of Shevchenko Bay. The highest biomass of the mollusk Caspiohydrobia sp. was recorded in those areas of the sea, which were far from the impact of fresh river water. In 2007, these mollusks were found only in the western part of the sea, while in 2006 they were not found at all. The polychaete Hediste diversicolor preferred areas of the sea with relatively low salinity. In 1997, high biomass of polychaetes was recorded not only in the desalinated zone but also in the central part of the sea and in Butakov Bay, where water salinity reached its maximum. Aggregations of the bivalve mollusk Cerastoderma glaucum were recorded in various parts of the sea, including the southeastern freshened water area.

Figure 4

The analysis of the correlation data showed that statistically significant positive correlations were most frequently observed between the water salinity and the biomass of mollusks A. ovata and C. glaucum (Table 3). Salinity had a significant positive impact on the polychaete H. diversicolor in 2006 and 2008, but in 2005 this relationship was negative. Positive and statistically significant moderate correlations between the biomass of the mollusk Caspiohydrobia sp. and water salinity were recorded in the period up to 2004. No correlation was recorded in the following years.

To further analyze the effect of water salinity on quantitative variables of the benthic invertebrates, we prepared scatterplots. Figure 5A demonstrates that the biomass of A. ovata increased almost linearly with the increase of salinity. In general, the distribution of biomass of Caspiohydrobia sp. and C. glaucum was characterized by a similar pattern of variability within the water salinity gradient (Figs 5C,D). Unlike other species, the distribution of H. diversicolor biomass within the water salinity gradient was even (Fig. 5B). High biomass of the polychaete was recorded both at low (about 1.0–1.5 PSU) and increased salinity (15.0–20.0 PSU).

Figure 5