Effects of fluvial discharges on meiobenthic and macrobenthic variability in the Vistula River prodelta (Baltic Sea)

Journal of Marine Systems 157 (2016) 135–146

Contents lists available at ScienceDirect

Journal of Marine Systems journal homepage: www.elsevier.com/locate/jmarsys

Effects of fluvial discharges on meiobenthic and macrobenthic variability in the Vistula River prodelta (Baltic Sea) Maria Włodarska-Kowalczuk a,⁎, Mikołaj Mazurkiewicz a,b, Emilia Jankowska a, Lech Kotwicki a, Mateusz Damrat c, Marek Zajączkowski a a b c

Institute of Oceanology, Polish Academy of Sciences, Powstańców Warszawy 55, 81-712 Sopot, Poland Centre for Polar Studies, Leading National Research Centre, Będzińska 60, 41-200 Sosnowiec, Poland Institute of Geological Sciences, Jagiellonian University, Oleandry 2A, 30-063 Kraków, Poland

a r t i c l e

i n f o

Article history: Received 19 July 2015 Received in revised form 24 December 2015 Accepted 25 December 2015 Available online 31 December 2015 Keywords: Macrofauna Meiofauna Benthos Diversity Community structure River prodelta Disturbance Organic matter Seasonality Baltic Sea

a b s t r a c t The role of environmental variability produced by river discharges in shaping the spatial and seasonal patterns of meiobenthic and macrobenthic communities was studied in the Vistula River (Baltic Sea) prodelta. Seven stations located in the delta front, the plume influence area and the distal zone of the prodelta were visited over the four seasons of 2012. Meiofauna, macrofauna, water (temperature, salinity, and suspended matter) and sediments (grain size, POC, TN, δ15N and δ13C and photosynthetic pigments) were analysed. The seasonal variations in the river discharges (with maximum flows in spring) resulted in a strong temporal variability in the studied environmental characteristics. In the benthic biota, the signals of seasonal variability, if present, were much weaker than spatial zonation. The benthic communities inhabiting the delta front where the main bulk of fluvial materials was deposited were taxonomically impoverished. The richest fauna dwelled within the plume influence area where the physical disturbance ceased and primary marine production was enhanced by river transported nutrients. In the distal zone outside the river influence, the fauna was dominated by deeper dwelling species, and the numbers of individuals and taxa decreased. Factors related to the riverine discharges (i.e., salinity, mineral suspension, POC and δ13C in the water and sediments) were identified as having high correlation with variability in the meiofaunal and macrofaunal community descriptors. Evidently, the interplay of food (i.e., the quantity and quality of organic matter) and disturbance (i.e., the deposition of river transported minerals) constraints shaped the patterns of benthic variability in the prodelta of the second largest river entering the Baltic Sea. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Despite their relatively small areal extent, marine systems influenced by large rivers play a great role in maintaining global ocean carbon budgets and are the most productive regions in coastal ocean (Bianchi and Alison, 2009). Subaqueous parts of the river deltaic system (prodeltas) typically extend into coastal waters. An intensive accumulation of river transported mineral particles forms dynamic delta fronts. Turbid water plumes extending beyond this region define river boundaries and physically and biogeochemically impact marine waters (Bonifacio et al., 2014, Rhoads et al., 1985). The large amounts of organic carbon in the prodelta system are mixtures of allochthonous (terrestrially derived) and autochthonous (marine primary production fuelled by nutrients transported by the rivers) materials (Doi et al., 2005), which, despite reduced diversity, supports high standing stocks and productivity of pelagic and benthic consumers (Elliott and McLusky, 2002). The quantities and quality of the riverine and marine materials settling to ⁎ Corresponding author. E-mail address: [email protected] (M. Włodarska-Kowalczuk).

http://dx.doi.org/10.1016/j.jmarsys.2015.12.009 0924-7963/© 2016 Elsevier B.V. All rights reserved.

the sea bottom and other environmental and biological parameters and processes change dramatically across prodelta benthic habitats, which are commonly categorized as: 1) the delta front, i.e., areas in the vicinity of the river mouth with the highest loads of river transported materials; 2) the plume area, i.e., areas within the range of the riverine plume influence; and 3) the distal zone, i.e., areas with no or little river impact (Bonifacio et al., 2014, Rhoads et al., 1985). Over the past 2000 years, European estuaries have been under high anthropogenic pressure. The industrial revolution has increased organic waste discharge in the past century, necessitating construction projects to provide flood protection and land reclamation (McLusky, 1999). Changes in land use (i.e., deforestation, intensive agriculture and urbanization) have increased sediment yield and deposition in river catchments leading to detrimental consequences on the biodiversity and functioning of river-proximal coastal systems (Akoumianaki et al., 2006; GESAMP, 1993). However, construction projects, e.g., river dams, have been shown to improve the nutritious quality of organic matter in the river prodeltas by enhancing phytoplankton production and preventing the accumulation of land-derived vascular plant debris (Bianchi and Alison, 2009). Global changes have also influenced

136

M. Włodarska-Kowalczuk et al. / Journal of Marine Systems 157 (2016) 135–146

estuaries by modifying their biotic and abiotic parameters (Chaalali et al., 2013). Due to more frequent severe weather events such as floods or droughts, the intensity of the river inflows and their effects on coastal ecosystems have changed (Grilo et al., 2011). Higher precipitation during warm periods and increased runoffs during melting seasons may result in increased freshwater input and changes in estuarine water stratigraphy and stronger salinity fluctuations (Scavia et al., 2002). Higher water discharges can improve water quality by diluting nutrients and increasing oxygen concentrations (Struyf et al., 2004) but are accompanied by higher loads of terrestrial (organic and mineral) materials (Syvitski and Andrews, 1994). However, global warming may also result in increased summer warm periods, with decreased precipitation and river discharges and a marinization of an estuary (David et al., 2007). In this scenario, water salinity increases, species compositions have increased marine components and estuaries are more efficiently utilized as nurseries by marine pelagic fishes (Chaalali et al., 2013, Pasquaud et al., 2012). The recognition of the variability and structure of benthic biota and their interactions with a dynamic estuarine environment is crucial for the understanding of the whole river prodelta system. Benthos is an important element of the energy and matter flow in a river prodelta and contributes in the remineralization of organic matter deposited in estuarine sediment. Organic particles are buried into deeper sediment layers, surface sediments are biologically mixed, and some substances are released into water (Aller et al., 2008; Herman et al., 1999). To a large extent, the fate of terrigenic organic carbon transported by rivers and preserved in marine sediments is regulated by the presence, standing stocks and composition of a range of benthic biota (from microbes to large invertebrates). Only few studies have assessed the relationship between prodelta/estuarine benthic communities and environmental variables (Bonifacio et al., 2014, Chainho et al., 2006; Rhoads et al., 1985; Wijsman et al., 1999). These studies assessed and quantified the relationships among the spatial and temporal variability of biota and such abiotic factors as biochemical descriptors of organic matter in water columns and in sediments. However, the majority of river prodelta benthic studies have been restricted to macrofauna (invertebrates retained on a 0.5-mm sieve), little attention has been paid to smaller invertebrates (meiofauna). Few studies have included both benthic communities (Patrício et al., 2012; Włodarska-Kowalczuk et al., 2007; Montagna and Kalke, 1992). Vistula is the longest and the second largest river entering the Baltic Sea. During the last 120 years since the opening of an artificial river mouth (‘Przekop Wisły’), the river mouth has moved seaward by 1.5 km (Koszka-Maroń, 2009). A number of studies have investigated the Vistula. Studies have included identifying the pathway of sediment formation from the shallow area at the river mouth to the depositing area in the Vistula prodelta (Zajączkowski et al., 2010; Damrat et al., 2013), the composition, origin and spatial distribution of organic matter in the water column and sediments (Kuliński and Pempkowiak, 2008; Maksymowska et al., 2000), chloropigments in the water column and sediments (Szymczak-Żyła and Kowalewska, 2007), and the hydrochemical and biological impacts of Vistula on the pelagic zone (Ochocki et al., 1995; Pastuszak, 1995). However, the effects of these environmental gradients on benthic fauna have not been investigated. The aim of the present study is to assess the effects of riverine discharges on the spatial and temporal (seasonal) variability of meiofaunal and macrofaunal components of the Vistula prodelta benthic system and to identify the main environmental factors driving the variability in benthic standing stocks (density), diversity and composition. Many of the published studies have been based on materials collected during one (usually summer) season. However, seasonal fluctuations can impose severe changes to the reported patterns observed during the summer and cannot be neglected to understand the function of the estuarine systems (Chainho et al., 2006). We therefore base our study on materials collected over four seasons throughout a year period to determine whether seasonal variations in the riverine discharges significantly

alter the patterns of benthic response to the estuarine environmental processes. 2. Material and methods 2.1. Study area Vistula is the longest river flowing into the Baltic Sea, with a catchment area of 194,000 km2 (Pruszak et al., 2005). In the past, River Vistula has created a large delta (Żuławy), but since 1895, its outflow has restricted to one artificial channel (Przekop Wisły, Fig. 1). The average water discharge near the Vistula mouth is 1080 m3 s−1, varying seasonally from 250 to 8000 m3 s−1 (Cyberski et al., 2006). Maximum discharges are typically noted in spring, due to ice and snow melting in the entire catchment area (Pruszak et al., 2005). In 2012, the river flow varied from 424 m3 s−1 in September to 1810 m3 s−1 in March (Polish Institute of Meteorology and Water Management). In most days of September and October, water discharge did not exceed 500 m3 s−1. In March and April, daily water discharge exceeded 1200 m3 s−1. Due to long period of drought during summer, the average annual river flow (798 m3 s−1) was lower than those reported in the previous years (Polish Institute of Meteorology and Water Management). The annual sediment transport into the Gulf of Gdańsk ranged from 0.7 to 2.2 million tons of suspension (Emelyanov and Stryuk, 2002) or from 0.6 to 1.5 million m3 of sediment in total (Pruszak et al., 2005). During the year 2012, an average sediment load in riverine waters was 14.1 mg dm−3, varying from approximately 10 mg dm−3 during typical river discharges (i.e., January, August, and November) to 30 mg dm−3 in spring (i.e., May, Damrat, unpublished data). A study of the sediment pathway from the Vistula mouth to the outer prodelta demonstrated the significant role of sediment redeposition because the Vistula prodelta is highly influenced by waves (Damrat et al., 2013). Two-thirds of the deposited sediment was remobilized and transported to the Gdańsk Basin during winter storms (Damrat et al., 2013). Average sedimentation rates reach 6000 g m−2 yr−1 with the maximum occurring during spring (March to May) and the minimum between August and January. However, a study on sediment accumulation in decadal time scales (estimated based on 210Pb sediment dating) showed that only 2000 g m−2 yr−1 of stored material remained at the bottom of the Vistula prodelta (Damrat et al., 2013). 2.2. Sampling Material was collected from the board of the r/v “Oceania”. Samplings were conducted during the four seasons of 2012: winter (10–11th January), spring (17th May), summer (27th August) and autumn (28th November). Seven sampling stations representing four zones defined by increasing distance from the river mouth were located along two transects starting from station N1 located 2.0 km north of the Vistula river mouth, stations N2, N3, and N4 extending northward (4.3 km, 6.9 km, and 8.9 km) and stations E2, E3, and E4 extending eastward from the river mouth (4.3 km, 8.5 km, and 20.2 km, respectively, Fig. 1). Samplings and measurements conducted at each station and in each season included hydrological measurements, water and sediment samples for geochemical analyses and meiofauna and macrofauna samples. Water salinity and temperature and turbidity were measured in vertical profiles using a Sensordata 204 CTD. Surface and bottom waters (for suspended matter analyses) were collected using a Niskin bottle at each station. Once collected, the samples were vacuum-filtered onto pre-weighed filters (MN GF5 with 0.4-μm openings). Samples of surface sediment for geochemical and meiofauna analyses were collected from a box corer using a syringe (3.5-cm diameter). These sediment samples consisted of upper 1 cm for grain sizes, POC, TN, δ15N and δ13C and photosynthetic pigments, and upper 5 cm for meiofauna (3 replicates). Samples for macrofauna were collected using the van Veen grab (0.1 m2 catching area, 3 replicates). Sediment samples for geochemical

M. Włodarska-Kowalczuk et al. / Journal of Marine Systems 157 (2016) 135–146

137

Fig. 1. Location of sampling stations in the Gulf of Gdańsk.

analyses were frozen (for photosynthetic pigment analysis, in −80 °C; other samples, in − 20 °C). Samples for macrofauna were sieved on board though a 0.5-mm sieve. Macrofauna and meiofauna samples were fixed in 4% formaldehyde and seawater solution.

2.3. Laboratory analysis The organic and mineral matter in suspensions collected on filters were analysed after drying at 60 °C for 24 h and combustion at 450 °C for 24 h. Chlorophyll a (Chl a) and pheopigment concentrations in the sediment samples were measured using a fluorometric method. Pigments from freeze-dried sediments were extracted in 90% acetone for 24 h at 4 °C (Evans et al., 1987). A Perkin Elmer LS55 Fluorescence Spectrometer was used for measurements. Emissions at 671 nm and excitations at 431 nm were measured before and after sample acidification with 1 M HCl and used to calculate the chlorophyll a and pheopigment concentrations, according to the method described by Evans and O'Reilly (1982). POC and TN in suspended matter and sediments were analysed using an Elemental Analyser Flash EA 1112 Series (Thermo Electron Corp., Germany). The freeze-dried, ground sediment samples were weighed into silver capsules to the nearest 1 mg. Prior to measurements, the samples were acidified with HCl fumes for 24 h to remove carbonates (Hedges and Stern, 1984). Then, they were dried again at 60 °C for 24 h. The POC and TN measurements were calibrated with standard reference materials from Thermo Electron Corp. and environmental certified reference materials from HEKAtech GmbH (Germany). To determine the grain size distribution, a Malvern Mastersizer 2000 particle size analyser was used. Data were recalculated using the GradiStat 4.0. software. Meiobenthic samples were treated according to the method described in Pfannkuche and Thiel (1988). A standard decantation technique was used to extract animals from the sediments. The extracted animals were then stained with Rose Bengal and formaldehyde. All invertebrates that passed through a 500-μm sieve and were retained on a 32-μm sieve were counted and identified to the major taxa level. Macrofauna samples were counted and identified to the lowest possible taxonomic level. In the case of samples with very high numbers of Oligochaeta, all individuals from other taxa were extracted from the sample and analysed, and the remaining oligochaetes were split into 8 subsamples. Then, only one of the oligochaete subsamples was examined, and the results were later extrapolated to the entire sample.

2.4. Data analyses A dataset of 32 environmental variables (listed in Table 1A) was used to identify patterns of environmental variability and relationships with biological data. The environmental data were tested for normality in distribution and collinearity. Eight variables were log transformed to assess the normality of distribution. Six variables were excluded from the analyses because they were strongly correlated (Pearson's correlation r N 0.9) with the other variables. The final dataset used for the analyses comprised 25 variables (Table 1A). The variability of environmental data was visualized with the use of ordinations based on the normalized data. The Euclidean distance was used as a similarity measure. PCO (Principal Coordinates Analyses) plots of Euclidean distances among centroids computed for groups of samples representing season/ zone combinations were plotted. CAP (Canonical Analyses of Principal Coordinates) ordinations that best discriminated groups of samples were defined by zones and seasons. Species (taxonomic) richness, defined as the number of taxa in a sample (S), species diversity as measured by the Shannon–Wiener diversity loge based index (H), and evenness of distribution of individuals among taxa as expressed by the Pielou index (J), were calculated for all meiobenthic and macrobenthic samples. Differences in univariate characteristics (i.e., density, S, J, and H) among the four zones (1, 2, 3, and 4) and four seasons (winter, spring, summer, and autumn) were tested using the two-way PERMANOVA model based on a similarity matrix created from the Euclidean distances among the samples. Unbiased estimates of each of the components of variation (CVs) were calculated from mean squares to compare the amounts of variation attributable to different terms in the model. When a significant effect of a factor was indicated by a main test, pairwise post-hoc comparisons were performed. When both a significant effect of a factor and a significant interaction between two factors (zone and season) were detected, pairwise tests for differences between different levels of a factor were performed separately within each level of the other factor, as recommended by Anderson et al. (2008). Bray–Curtis similarities were calculated for fourth root-transformed data of meiobenthic higher taxa and macrobenthic species abundances in the samples. The patterns of meiobenthic and macrobenthic composition were illustrated with PCO ordinations plotted for centroids computed for groups of samples representing all season/zone combinations. Additionally, the CAP were used to visualize the variability along the

138

M. Włodarska-Kowalczuk et al. / Journal of Marine Systems 157 (2016) 135–146

two axes that best discriminated groups of samples defined by zones and by seasons. Spearman rank correlation vectors of species abundances with 2 canonical axes were overlaid on the CAP plots. The twoway PERMANOVA models (with two fixed factors, i.e., season and zone), were applied to the similarity matrices. Both main tests and post-hoc pairwise comparisons were performed. Relationships between environmental variables and meiobenthic and macrobenthic densities, species richness and community composition were investigated using the Distance-based Linear Models (DISTLM) procedure in PERMANOVA (Anderson et al., 2008). The dataset of 24 environmental variables was used for the analyses (Table 1A). Depth was not included in the analyses because we did not consider it as a true environmental driver; depth was considered a proxy of other factors. The forward selection procedure was based on an adjusted R2 as a selection criterion and was used to determine the best combination of predictor variables. This procedure first chooses the variable with the best value of the selection criterion and then

selects the next variable that together with the first improves the selection criterion the best. The procedure stops when no further improvement in the selection criterion is possible (Anderson et al., 2008). 3. Results 3.1. Environmental settings The unconstrained ordinations of environmental data measured at the stations and in the seasons showed that the seasonal changes were much stronger than the spatial changes. However, both seasons and zones shaped the documented variability (PCO, Fig. 2A). The PCO plot showed that the spring samples were located far from those of the other seasons. A similar plot was shown for the seasonconstrained CAP ordination (Fig. 2B). The vectors drawn on the CAP ordination indicated that water temperature was best correlated and increased with the directional changes in the environmental gradients

Fig. 2. Ordination plots based on Euclidean distances computed for normalized environmental data. (A) PCO plot for centroids computed for groups of samples representing season/zone combinations. Symbols represent seasons, and the points representing the same zone are linked with a line. (B) CAP ordinations best discriminating groups of samples defined by zones (left) and seasons (right), symbols represent zones (left) and seasons (right). Vectors indicate the variables best correlated with the ordination coordinates. Vector lengths correspond with the correlation values. For the clarity of the graph only variables of correlation r N 0.6 are presented. Variable labels are explained in Table 1A.

M. Włodarska-Kowalczuk et al. / Journal of Marine Systems 157 (2016) 135–146

towards the summer samples. A group of indicators of suspended organic matter and sediments varied with the directional change between the spring and autumn/winter samples. Increasing values of mineral and organic suspended matter, POC concentration and δ15N in suspensions increased towards the spring samples, and the POC/TN ratio in sediments increased towards the winter/autumn samples (Fig. 2B). When the zones were considered, zone 1 was clearly outstanding in terms of environmental settings (as visualized in the PCO and zoneCAP, Fig. 2A and B). Zones 2, 3 and 4 gradually changed (overlapping to some extent) with increasing distance from the river mouth. The samples representing the four zones were located along the CAP1 axis. The overlaid vectors indicated that the shift in zones from 1 to 4 was accompanied by increasing depth and salinity (from 7.1 to 7.7) and decreasing δ15N in water (from 5.4‰ to 2.6‰) and decreasing δ15N (from 4.4‰ to 3.1‰) and δ13C in sediments (from − 28.7‰ to −26.3‰, Table 1A). The mean grain size, sorting, skewness, mud, Chl a and POC contents in the sediments were best correlated to the CAP2 axis (Fig. 2B). Finer grain sizes (mean grain size 192.8 μm) were documented in the first zone than in zones further away from the delta front (mean grain size from 388.7 μm to 269.6 μm, Table 1A). Additionally, in zone 1, the CPE was much higher (391.8 μg) than in the other zones (from 130.4 μg to 83.1 μg, Table 1A). Both POC (1.3%) and Chl a (84.9 μg g−1) were the highest in zone 1, decreased in zones 2 (0.4%, 13.5 μg g− 1) and 3 (0.3%, 10.7 μg g− 1) and reached relatively high values in zone 4 (1.1%, 23.5 μg g−1, Table 1A). 3.2. Meiofauna Twelve higher taxa of meiofauna were identified in samples. Meiofauna was dominated by Nematoda (92% of all collected individuals) and followed by Turbellaria (4%) and Gastrotricha (1%; Table 3A). PERMANOVA identified significant differences in all univariate characteristics of the meiofauna community (i.e., density, S, H and J) among the different zones. Seasonal effects were detected only for S (Table 1). Density was significantly lower in zone 4 (mean, 548.6 ± 144.1 for a 0.95 confidence interval ind. 10 cm− 2) than in the other zones (mean values above 1000 ind. cm−2, Fig. 3, Table 3A). The number of taxa per sample was the lowest in zone 1 (6 ± 1) and the highest in zone 3 (9 ± 11). The number of taxa was also significantly higher in spring than in winter or summer (Fig. 3, Table 3A). Both the Shannon– Wiener index and Pielou index were significantly lower in zone 1 than in the other three zones (Fig. 3). PERMANOVA identified significant effects (at P b 0.001) of season and zone on meiobenthic community composition (Table 1). Zone had a stronger effect than season as indicated by the higher CV. A high CV value for residuals also indicated the large variability among the replicate samples. Post-hoc pairwise tests indicated significant contrasts among all

139

pairs of groups of samples defined by zones and seasons (Table 2A). Zones were clearly separated on a centroid-based PCO (Fig. 4) and zoneoriented CAP ordinations (Fig. 5). A taxonomic composition of meiofauna showed shifts in the taxa composition across the four zones. The vectors plotted on the ordination plots indicated the taxa that were best correlated to the axis of variability that best discriminated the zones (Fig. 5). Nematoda was indicated as the best correlated with variability in the direction of the zone 1. Indeed, the highest abundance of Nematoda and the highest percentage made by Nematoda in the total meiofaunal abundance was documented in zone 1 (1741.1 ind. 10 cm−2, 97%). In the other zones, the nematode abundance varied from 490.9 to 1188.4 ind. 10 cm−2, and the percentage in the total abundance varied from 89 to 93%. Polychaeta and Turbellaria defined zone 2. Turbellaria occurred at an abundance (100.0 ind. 10 cm−2) twice as high as that in zone 3 (44.4 ind. 10 cm−2). In the remaining two zones Turbellaria abundance was even lower (13.1–15.8 ind. 10 cm−2). Copepoda nauplii and Harpacticoida were best discriminated in zone 3. Both taxa were almost 2 times higher in zone 3 (22.2 ind. 10 cm−2) than in zone 2 (12.6 ind. 10 cm−2) and 3 and 4 times higher than in the other two zones (8.0–4.5 ind. 10 cm−2). The other taxa, Ostracoda and Oligocheata, were correlated with zones 2 and 3. Their abundances were similar at those two zones (Oligochaeta 4.0 and 2.0 ind. 10 cm−2, Ostracoda 10.7 and 15.0 ind. 10 cm−2), which were much higher than at zones 1 and 4 (Oligochaeta 0.9 and 0.6 ind. 10 cm−2, Ostracoda 4.9 and 4.6 ind. 10 cm−2). Kinorhyncha were correlated with zones 3 and 4, where the mean abundance of this taxon was rather similar (4.0 ind. 10 cm−2 in zone 3 and 3.2 ind. 10 cm−2 in zone 4); they were not present or were negligible in zones 1 and 2. The season-oriented CAP ordination points representing summer and autumn samples were separated; points representing winter and spring overlapped (Fig. 5). The strongest vector of correlation was noted for Gastrotricha in the autumn season. The average abundance in the group was indeed much higher in autumn (40.2 ind. 10 cm−2) than in the other seasons (between 3.5 and 1.0 ind. 10 cm−2). Other taxa that occurred with increased numbers in a single season were Tardigrada (autumn), Harpacticoida (winter) and Bivalvia (summer). Nine variables were included in the best fitting model, accounting for 76% of the total variation in meiofauna density (DISTLM analyses). Five variables were significant in the model, i.e., the mineral matter concentration in the surface waters and the δ13C, POC, mud content and skewness in grain size distribution in the surface sediments (Table 2). For the number of taxa per sample, 12 predictor variables were included in the model (accounting for 87% of the variation). Four variables were significant, i.e., δ13C in deep waters, POC/TN and POC content, and skewness in the surface sediments. Of the 12 predictor variables included in the model used to explain the variations in the meiofauna community, five were significant: the salinity, the organic and mineral

Table 1 Results of two-way PERMANOVA tests for differences in meiobenthic and macrobenthic univariate (density, number of taxa per sample (S), Shannon–Wiener index (H), and Pielou index (J)) and multivariate (comp — Bray–Curtis similarity) characteristics among zones (Zo) and Seasons (Se). PsF — PERMANOVA pseudoF, CV — component of variation. Significant effects: * P b 0.05, ** P b 0.001. Density Source

df

PsF

Meiofauna Zo Se ZoxSe Res

3 3 9 64

7.8** 1.5 1.8

Macrofauna Zo Se ZoxSe Res

3 3 9 65

51.9** 9.0** 3.4*

S CV

H PsF

J CV

Comp

PsF

CV

PsF

483.6 134.3 334.1 816.1

17.1** 2.9* 1.1

1.4 0.5 0.2 1.5

7.7** 1.6 1.1

0.11 0.03 0.02 0.18

5.2* 1.3 0.9

7087.1 2928.1 3076.3 4395.9

12.5** 16.1** 2.7*

1.0 1.1 0.7 1.2

58.7** 29.0** 25.8**

0.2 0.1 0.2 0.1

20.8** 9.1** 13.9**

CV

PsF

CV

3.9E−2 1.1E−2 −0.6E−2 8.2E−2

9.4** 4.4** 1.4

12.3 8.1 5.3 18.7

5.5E−2 3.6E−2 8.9E−2 5.5E−2

73.4** 7.1** 3.0**

26.2 7.9 8.8 13.6

140

M. Włodarska-Kowalczuk et al. / Journal of Marine Systems 157 (2016) 135–146

Fig. 3. Meiofauna density [ind. 10 cm−2], number of taxa per sample, Shannon–Wiener index of species diversity and Pielou index of evenness in groups of samples defined by a zone (1, 2, 3, 4) or a season (Wn — winter, Sp — spring, Sm — summer, At — autumn). Mean and 0.95 confidence interval values are presented.

matter contents in deep waters and the δ13C and the POC in the sediments. The variables included in the model accounted for 75% of the variation.

3.3. Macrofauna In the macrobenthic samples, 23 species were identified (Table 4A). Oligochaeta accounted for 31% of all collected individuals, followed by Hydrobia spp. (23%) and Macoma balthica (13%).

There were significant differences and interactions in all univariate characteristics of macrobenthic communities among the seasons and zones (PERMANOVA, Table 1). Density was the lowest in zone 4 (mean 4622.1 ± 939.5 m−2), higher in zone 3 (10,326.1 ± 1656.8 m−2) and highest in zones 2 and 1 (20,249.6 ± 3816.2 m−2 and 13,698.2 ± 3523.45 m−2, respectively). These contrasts remained consistent across the four seasons (Fig. 6, Table 4A). The seasonal patterns were less consistent; however, a decline in density in autumn could generally be observed (Fig. 6). All diversity indices (S, H, and J) had maximum average values in zone 3 and minimum average values in zone 1 (Fig. 6). The

Fig. 4. PCO plots based on Bray–Curtis similarities of species/taxa abundances in samples for meiofauna (left) and macrofauna (right). Ordinations for centroids computed for groups of samples representing season/zone combinations are presented. Data were fourth root transformed. Symbols represent seasons, and the points representing the same zone are linked with a line.

M. Włodarska-Kowalczuk et al. / Journal of Marine Systems 157 (2016) 135–146

141

Fig. 5. CAP ordinations best discriminating groups of samples defined by zones (left) and seasons (right), based on of Bray–Curtis similarities of meiobenthic taxa abundances between centroids computed for groups of samples representing season/zone combinations. Data were fourth root transformed. Symbols represent zones (left) and seasons (right). Vectors indicate the taxa best correlated with the ordination coordinates (r N 0.2). Vector length corresponds to the correlation values.

differences between the zones had varying levels of significance depending on the season (Table 4A). Density and diversity indices (S, H, and J) did not vary seasonally in zone 1 (Table 4A). In the other zones, the highest values (and in most cases, significantly different from the other seasons) of these characteristics were observed in the summer (Fig. 6, Table 4A). The clear separation of macrofauna among the four zones is wellillustrated in the centroid-based PCO (Fig. 4), and the zone-oriented CAP ordination (Fig. 7). The vectors in the CAP ordination discriminated the species that best defined the fauna in the four zones. Hediste diversicolor strongly correlated with zone 1, where it occurred with the highest abundance at 843.3 ind. m−2 (whereas in the other zones, the abundance varied between 186.4 and 26.7 ind. m−2). Oligochaeta were correlated with zones 1 and 2. The highest abundance of Oligochaeta was noted in zone 2 (8411.7 ind. m−2), followed by zone 1 (6522.5 ind. m− 2), zone 3 (1291.9 ind. m− 2) and zone 4 (241.0 ind. m−2). A vector for Hydrobia spp. indicated its abundance in zone 2 that was at least two times higher than in the other zones (6377.7 ind. m− 2 and from 135.7 to 2359.8 ind. m− 2, respectively).

Corophium volutator was correlated with zones 2 and 3 where the abundance was higher (1581.8 and 1928.8 ind. m−2) than in the two other zones (from 193.3 to 36.3 ind. m−2). In zone 3, Pygospio elegans was indicated as the defining species with the highest abundance (2023.4 ind. m− 2) when compared with the other zones (870.6– 22.8 ind. m− 2). Halicryptus spinulosus was correlated with zones 3 and 4, and occurred with abundances of 465.1 ind. m− 2 and 382.5 ind. m−2, respectively, and lower numbers in zones 1 and 2 at 5.0 ind. m−2 and 30.0 ind. m−2, respectively. Three species were strongly correlated with zone 4: Monoporeia affinis, Pontoporeia femorata and Saduria entomon. Their abundances were much higher (347.5, 322.5, and 30.3 ind. m− 2, respectively) than in zone 3 (5.0, 27.5, and 10.0 ind. m− 2) and zone 2, where only P. femorata was present (5.0 ind. m−2). None were present in zone 1. The season-oriented CAP ordination showed that the samples collected in autumn were clearly different from the samples collected in the other seasons. The winter, spring and summer samples did not overlap; however, these groups were close to each other (Fig. 7). The two bivalve species, Cerastoderma glaucum and Mytilus edulis, were correlated with the autumn season;

Table 2 Results of DISTLM procedure for fitting environmental variables to meiofauna univariate characteristics (density, number of taxa per sample — S) and community composition (comp, analysed with the Bray–Curtis similarities of fourth root transformed data). Sets of best fitted descriptor variables selected in sequential test are presented. Variable labels are explained in Table 1A. psF — PseudoF, cumR2 — cumulative R2, significant effects: * P b 0.05. Meiofauna Density Variable 13

δ C min_surf skew POC/TN_deep POC mud kurt Chl a temp

S

Comp

psF

R2

cumR2

Variable

psF

R2

cumR2

Variable

psF

R2

cumR2

6.1* 6.7* 4.4* 2.8 4.7* 4.5* 3.1 4.7 1.3

0.19 0.17 0.10 0.06 0.08 0.07 0.04 0.03 0.02

0.19 0.36 0.46 0.52 0.60 0.67 0.72 0.74 0.76

POC POC/TN δ13C_deep temp salin δ15N mud kurt skew δ13C min_deep min_surf

9.4* 6.0* 4.7* 3.4 2.0 2.8 3.8 3.5 6.6* 1.5 1.9 2.3

0.27 0.14 0.09 0.06 0.04 0.05 0.06 0.04 0.06 0.01 0.02 0.02

0.27 0.41 0.51 0.57 0.61 0.65 0.71 0.75 0.82 0.83 0.85 0.87

salin POC δ13C min_deep sort kurt POC/TN_surf min_surf δ13C_deep POC/TN_deep min_deep δ13C_surf

3.9* 4.0* 3.8* 4.6* 2.4 2.4 1.9 1.8 1.6 1.3 2.4* 1.3

0.13 0.12 0.10 0.11 0.05 0.05 0.04 0.04 0.03 0.02 0.04 0.02

0.13 0.25 0.35 0.46 0.51 0.56 0.60 0.64 0.67 0.69 0.73 0.75

142

M. Włodarska-Kowalczuk et al. / Journal of Marine Systems 157 (2016) 135–146

Fig. 6. Macrofauna density [ind. m−2], number of taxa per sample, Shannon–Wiener index of species diversity and Pielou index of evenness in groups of samples defined by a zone (1, 2, 3, 4) or a season (Wn — winter, Sp — spring, Sm — summer, At — autumn). Mean and 0.95 confidence interval values are presented.

their average abundances were much higher in autumn than in the three other seasons (for C. glaucum: 82.3 in autumn, 17.3 in winter, 10.0 in spring, and 10. ind. m−2 in summer). PERMANOVA identified significant effects of zone, season and interaction between these two factors in the macrobenthic composition (Table 1). The highest CV value for zone indicated that this factor explained the largest fraction of the macrobenthic variability. The posthoc pairwise tests identified significant contrasts among the zones in all seasons. There was no seasonal variability in the macrobenthic composition in zone 1, while either all or most pairs of seasons differed significantly in this respect in the other zones (Table 2A). Fifteen variables were included in the model developed for the macrobenthic density by DISTLM analyses. Together, they accounted for 79% of density variation (Table 3). Only the first two variables were significant (i.e., salinity and POC/TN in the surface waters) and accounted for 32% of the variation. For species richness, 16 variables were determined as the best predictors of variation. They accounted for 96% of the variation. Four variables were significant (the temperature and POC content in water and the δ15N and chlorophyll a concentrations in sediments). With the concentration of mineral matter in deep waters, the four variables also accounted for 79% of the variations in species number per sample. Regarding the species composition of the macrobenthic communities, 19 variables accounted for 94% of the variation. The δ15N and mineral matter contents in water and the δ13C and POC in the sediments were identified as significant in the model (Table 3). 4. Discussion Fluvial discharges shape the patterns of benthic variability in the Vistula River prodelta. A list of factors significantly correlated to the biological variability includes variables that are related to the freshwater input (salinity) and to the quantity and characteristics of the materials

transported to sea in the riverine discharges (i.e., mineral suspension concentrations in surface and bottom waters and the descriptors of organic matter in sediments and suspended matter). The riverine impacts dictate a clear zonation and seasonal changes in the environmental settings of the neighbouring coastal systems, and the fauna respond in clearly defined spatial patterns, while it remains relatively robust to environmental seasonality. 4.1. Zonation of environmental settings and benthic communities The station located in the vicinity of the river mouth (zone 1) was clearly separated from the other sites in terms of environmental conditions and benthic community descriptors. Zone 1 represents the delta front, i.e., an area that is directly influenced by river discharge and has high turbidity and high sedimentation rates (Wijsman et al., 1999). The main bulk of river transported material is discharged at the river delta front and forms a steep slope covered by unstable, easily eroded sediments (Zajączkowski and Włodarska-Kowalczuk, 2007) with relatively fine grain sizes. The Vistula River delta front sediments are also characterized by elevated concentrations of organic carbon and photosynthetic pigments (i.e., chlorophyll a and pheopigments). The accumulated organic matter was apparently of terrestrial origin and transported with riverine discharges as indicated by depleted δ13C values (mean − 28.7‰). δ15N levels increased closer to the river mouth (4.4‰), reflecting the strong anthropogenic impact of Vistula watersheds and large loads of sewage transported by the river. The fauna in the delta front (zone 1) of the Vistula River was depauperate in terms of macrofaunal species and meiofaunal higher taxa richness and in other diversity indices. Macrofauna and meiofauna were strongly dominated by Oligochaeta and Nematoda, respectively. Macrofaunal density was also lower at the river mouth (despite the high concentrations of organic carbon and chlorophyll in the sediments) than a few kilometres away in zone 2. Several previous studies have shown

M. Włodarska-Kowalczuk et al. / Journal of Marine Systems 157 (2016) 135–146

143

Fig. 7. CAP ordinations best discriminating groups of samples defined by zones (upper) and seasons (down), based on of Bray–Curtis similarities of macrobenthic species abundances between centroids computed for groups of samples representing season/zone combinations. Data were fourth root transformed. Symbols represent zones (upper) and seasons (down). Vectors indicate the species best correlated with the ordination coordinates. Vector length corresponds to the correlation values.

that sedimentary environments at delta fronts tend to have reduced density, biomass and diversity of benthic macroinfaunal communities. Furthermore, these environments are dominated by pioneering, small, surface-dwelling species (Rhoads et al., 1985; Aller and Stupakoff, 1996; Meire et al., 1991; Ysebaert et al., 2003; Włodarska-Kowalczuk et al., 2007). The delta fronts are directly influenced by river discharges, turbid water, and high sedimentation rates (Wijsman et al., 1999; Zajączkowski and Włodarska-Kowalczuk, 2007). Typically low fauna diversity is due to high rates of deposition of river transported sediments, periodic erosional depositional events, fluid muds, and unstable seabeds off the river mouths (Rhoads et al., 1985; Aller and Stupakoff, 1996). High sedimentation can be destructive to benthic fauna because it can bury larvae and adult animals, impede animals from maintaining optimum positions in the sediments and clog the feeding and respiratory organs of macrobenthic animals, especially filter feeders (Moore, 1977; Ahrens and Morrisey, 2005). High rates of sedimentation (N 4 cm year−1)

close to the river mouths may even totally exclude infaunal benthos (Rhoads et al., 1985). These episodic deposition and erosion events that benthic populations face can be characterized as routine local extinctions. As sedimentation occurs, frequently on the unstable and steep slopes of river deltas, surface sediments and the associated fauna are physically removed, dragged along the slope, and redeposited elsewhere (Zajączkowski and Włodarska-Kowalczuk, 2007). Examinations of physical structures in sediment cores collected in estuarine delta fronts showed that even if biogenic structures were evident in X-ray radiographs, they were often truncated by erosive horizons or buried in depositional layers, indicative of failed colonization (Aller et al., 2008). The numbers of meiofaunal individuals in the proximal sediments of the Vistula River mouth did not differ from those recorded in the plume area. Aller et al. (2008) even stated that close to river mouths, bacteria and meiofauna can dominate benthic biomasses and activities because

144

M. Włodarska-Kowalczuk et al. / Journal of Marine Systems 157 (2016) 135–146

Table 3 Results of DISTLM procedure for fitting environmental variables to macrofauna univariate characteristics (density, number of taxa per sample — S) and community composition (comp, analysed with the Bray–Curtis similarities of fourth root transformed data). Eight most important variables selected in sequential tests performed using the forward selection procedure are presented. Variable labels are explained in Table 1A. psF — PseudoF, cumR2 — cumulative R2, significant effects: * P b 0.05, ** P b 0.001. Macrofauna Density Variable salin POC/TN_surf sort POC mean min_surf δ13C δ15N δ13C_deep temp POC_deep min_surf org_deep min_deep kurt

S psF 6.6* 4.4* 1.6 3.4 2.6 2.0 1.4 1.6 1.5 1.8 1.3 1.8 1.1 1.2 1.2

R2 0.20 0.12 0.04 0.08 0.06 0.04 0.03 0.03 0.03 0.03 0.03 0.03 0.02 0.02 0.02

cumR2 0.20 0.32 0.36 0.44 0.50 0.54 0.57 0.61 0.64 0.67 0.70 0.73 0.75 0.77 0.79

Variable temp δ15N min_deep POC_surf chlor δ15N_deep mean POC POC/TN_deep δ13C_deep min_surf POC/TN skew kurt POC/TN_surf mud

Comp psF 10.4* 10.8* 3.5 14.5** 6.3* 3.0 2.4 3.1 3.2 2.8 2.8 2.6 1.5 2.3 1.8 1.7

the macrofauna are eliminated by the frequently disturbed physical conditions. Warwick et al. (1990) also hypothesized that meiofauna were less sensitive to sediment instabilities and physical disturbances than larger invertebrates. However, in the proximal habitat of the Vistula River mouth, the meiofaunal community, even if not depleted in numbers, was clearly impoverished in terms of diversity (richness and evenness). This could be perceived even in analyses performed at relatively low taxonomic resolutions (higher taxa). Additionally, the higher variability in density values at a station located on the delta front indicates the unstable and dynamic environmental conditions in this zone, corroborating the effects of increased variation in biological features reported from other deltaic, disturbed environments (Akoumianaki et al., 2006). There are few comparative studies of meiobenthic and macrobenthic responses to physical disturbances, and their conclusions are inconsistent. The relative insensitiveness of meiofauna was reported by Warwick et al. (1990) who compared the effects of sediment disturbances of large ships on two ecological groups, and by Austen et al. (1989) who studied the effects of mechanical disturbance by humans digging for shellfish. However, Włodarska-Kowalczuk et al. (2007) documented parallel density decreasing patterns and clear changes in the compositions of meiofauna and macrofauna from a stable central basin to the slope of a glacial river in an Arctic fjord. Meiofauna and macrofauna also responded similarly to physical disturbances produced by megafaunal bioturbation (Austen et al., 2003), dredging disposal (Somerfield et al., 1995) and mechanical disturbances by deep-sea sediments (Ingole et al., 1999). The domination of nematodes, especially larger species, was also often reported for disturbed meiofaunal communities. Nematoda is a genera containing a wide range of morphological and functional forms and is often regarded as resistant to disturbances of various natures and origins (Wieser et al., 1974). Experimental studies showed that nematodes are able to actively migrate through interstitial spaces in deeper sediment layers and thereby withstand high sedimentation and burial by deposited mineral materials (Schratzberger et al., 2000). The relatively high concentrations of organic matter in delta front sediments (as recorded in the present study) do not translate into high standing stocks of benthic animals. This is due to the predominating effects of physical disturbance and to the specific characteristics of the organic matter available in this zone. Organic carbon in

R2 0.28 0.22 0.06 0.17 0.06 0.03 0.02 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01

cumR2 0.28 0.50 0.56 0.73 0.79 0.82 0.84 0.86 0.88 0.90 0.91 0.92 0.93 0.94 0.95 0.96

Variable 13

δ C POC min_surf δ15N_surf kurt POC/TN_surf salin min_deep mean org_surf δ15N δ13C_deep org_deep δ15N_deep POC_surf skew mud POC/TN_deep temp

psF

R2

cumR2

8.5** 12.2** 6.9** 2.4* 1.8 2.1 1.9 2.5* 2.3 2.0 2.2 2.2 1.8 2.1 1.5 1.6 1.3 1.1 1.4

0.25 0.25 0.11 0.04 0.03 0.03 0.03 0.03 0.03 0.02 0.02 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.01

0.25 0.50 0.61 0.65 0.67 0.70 0.73 0.76 0.79 0.81 0.83 0.85 0.87 0.89 0.90 0.91 0.92 0.93 0.94

these deposits are mostly terrestrially derived, refractory and resistant to microbial mineralization. Consequently, nutrient regeneration from the seabed is very modest (Rhoads et al., 1985). A strong dominance of oligochaete worms is often reported in upper estuaries (Attrill et al., 2009) and was noted in zones 1 and 2 in the Vistula River estuary; the locations of these worms can be linked to their unique ability to assimilate terrestrial carbon. In a study of three estuaries in Devon (UK), Attrill et al. (2009) showed clear evidence that oligochaetes acquired carbon from terrestrial sources (50–54% derived from terrestrial ground plans and tree leaves) in contrast with other estuarine taxa (only 2 to 17%), as indicated by mixing model analyses of stable isotope signatures. Attrill et al. (2009) indicated that this may be due to their evolutionary route; the worms invaded the estuaries from terrestrial/ freshwater systems (where there was an abundance of terrestrial plant material). In the Vistula prodelta sediments, the worms could take advantage of the huge supply of depleted carbon largely unavailable to the other taxa. Zones 2 and 3 were located within the range of the river water plume. The water turbidity, mineral sedimentation and associated physical disturbance of the seabed were negligible at zones 2 and 3 when compared with the zone at the delta front. The marine productivity was not hampered by the high turbidity and the organic matter in sediments, was largely composed of matter produced by marine plankton, and had a more labile character as indicated by the enriched values of δ13C (−27.0‰ and −26.3‰). The decreased physical disturbance and increased inputs of water-column productivity to the seabed resulted in the development of communities with higher standing stocks and a diversity of species representing various feeding and life-habits (Aller and Stupakoff, 1996). Estuarine macrobenthic densities and biomasses can be 10 to 100 times higher in deeper seaward regions than in shallow seabeds close to river outflows (Aller et al., 2008). Increased macrobenthic density and diversity were also observed in this plume area (zones 2 and 3) in the Vistula River prodelta. The diversity increased in terms of species richness and evenness of distribution of individuals among species. The increased density was limited only to stations in zone 2 and was produced by the large numbers of oligochaetes that coexisted with other estuarine taxa. Oligochaete worms were as numerous in zone 2 as in the delta front sediments (i.e., zone 1); however, further away from the river mouth (i.e., in zone 3) significantly decreased in

M. Włodarska-Kowalczuk et al. / Journal of Marine Systems 157 (2016) 135–146

numbers. This peak in macrobenthic density in zone 2 may indicate its transitional character. The highest abundance of macrofauna was expected to occur at a short distance from the area of maximum disturbance. For example, a study by Pearson and Rosenberg (1978) examined an organic enrichment gradient model. In this model, a transitional zone (i.e., areas immediately beyond the area of maximum disturbance) offered conditions that allowed opportunistic and disturbance-tolerant species to become extremely abundant and was sufficiently harsh to exclude many other better competing but less resistant species. The maximum abundance of benthic fauna was also reported in areas located immediately at the edge of strongest sedimentations of glaciofluvial materials transported to the fjords by tidal glaciers (Włodarska-Kowalczuk et al., 2005). Hydrobia spp. and Oligochaeta occurred at high numbers in zone 2, while Corophium sp. and P. elegans similarly occurred in zone 3. These species are regarded as typical and often dominant estuarine species due to their opportunistic, omnivorous and mobile characteristics (Chainho et al., 2006; Ysebaert et al., 2003, 1998). The zone 4 stations were located outside the plume range (the distal zone) and were not influenced by the river discharge. Typically, in coastal waters outside the plume influence, the numbers of benthic invertebrates drop as a result of decreased nutrient concentrations, reduced primary productivity and lower food availability for seabed dwellers (Wijsman et al., 1999). In the present study, macrobenthic densities in the distal zone also decreased, despite the good nutritious conditions as indicated by high concentrations of POC and chlorophyll a in the sediments (at levels even higher than those in the plume area). The stations of zone 4 were located at the edge of the hypoxic bottom waters that covered the outer, deeper parts of the Gulf of Gdańsk. The high macrobenthic diversity and the considerable numbers of P. femorata and M. affinis (a species that is highly sensitive to oxygen deficiency, Johansson, 1997) indicate the presence of well-oxygenated conditions at the studied stations. However, the temporary occurrence of anoxia or at least the shrinking of well-oxygenated upper layers of sediments could have occurred and may explain the decreased densities of benthic populations. Janas et al. (2004) reported that even if the bottom waters remained well-oxygenated, anoxia and hydrogen sulphides can develop in the deeper sediment layers and occasionally extend to the surface sediments in the outer parts of the Gulf of Gdańsk. Stable low temperatures in this zone are reflected by the presence of H. spinulosus and S. entomon, which are species that typically occur in the deeper waters of the Baltic Sea (Żmudziński, 1990). Lower temperatures and episodic oxygen deficits account for the higher organic matter content in the sediments and the lower densities of detrivore marofauna in this zone. 4.2. Seasonal variability in environmental settings and benthic communities The seasonal variability in environmental settings was stronger than the spatial zonation in the Vistula prodelta. Spring, the season of the maximal river discharges, was clearly separated from the other seasons. The environmental factors that best discriminated this season on the CAP ordination included various descriptors of organic and inorganic materials transported by the river to the sea. Elevated amounts of organic matter were coupled with enriched values of δ13C and decreased POC/TN, indicating that the freshly produced organic matter was compatible with the phenology of the phytoplankton blooms in the southern Baltic Sea (Witek et al., 1997). The benthic fauna community structure was relatively resilient to strong seasonal variability in environmental conditions. In the meiofauna, only species richness and composition significantly differed among seasons. In the macrofauna, seasonality was a significant factor responsible for differences in density, species richness, diversity and composition; however, the seasonal effects were weaker than those due to spatial zonation. The seasonal patterns in the macrobenthic communities were not consistent across the Vistula prodelta. This was likely

145

due to the chronic physical disturbance produced by mineral sedimentation. Even if the intensities of these disturbances fluctuated, they were balanced by the seasonal effects in the delta front sediments. Similarly, in the Gulf of Papua, the macrofauna seasonal variability was only expressed in the outer parts of the estuary (Aller et al., 2008). In some estuaries, benthic invertebrates responded to regular seasonal variations in the riverine discharge by declining in numbers of species and individuals, biomasses and maximum body sizes in periods of rising river flows. In periods of decreased river discharge, the invertebrates responded by becoming more abundant and diverse (Aller and Stupakoff, 1996; Akoumianaki et al., 2006; Bonifacio et al., 2014; Chainho et al., 2007). Seasonal fluctuations in abundance and composition in coastal and estuarine waters can also be dictated by recruitment pulses and by the occurrence of extreme episodic events (e.g., storms or floods, Chainho et al., 2006). In the Vistula prodeltaic benthic communities, the seasonality was rather related to species phenology and organic matter exports from the pelagial. In every zone (except for zone 1), the numbers of macrobenthic species per sample began to increase in the spring and reached maximum values in the summer due to species reproduction cycles (Obrebski, 1979; Wenne and Wiktor, 1982). The seasonal patterns in the Vistula River prodelta were similar to other, non-impacted areas and habitats in the Baltic Sea. Wenne and Wiktor (1982) stated that in the coastal zone of the Gulf of Gdańsk after spring breeding, a rapid increase in the number of crustaceans, gastropods and molluscs occurred in the late summer or autumn. In the present study, two bivalve species, C. glaucum and M. edulis, reached their highest densities in autumn. The subsequent density decrease after autumn is likely due to natural mortality (Wenne and Wiktor, 1982). Still, the extended period of reduced water flows in the late summer and autumn in the Vistula prodelta can enhance the establishment of the benthic fauna; this was similar to the observations made in other estuaries with seasonal variations in fluvial discharges (Bonifacio et al., 2014). 5. Conclusions The interplay among the levels of physical disturbance (produced by the river transported minerals deposited at the delta front) and the quality and quantity of organic matter available in sedimentary habitats (regulated by the river via the transport of terrestrial organic matter and control of marine productivity) shapes the patterns of the benthic communities. The present study documents how these constraints, in terms of resources and perturbations, shape the patterns of variability of the benthic communities attributes (i.e., density, diversity and taxonomic composition) in river prodelta habitats. To fully understand and predict the effects of climate fluctuations or anthropogenic pressures on coastal ecosystems, future studies will explore how the environmentally driven variability in community structure affects their functions (e.g., its productivity, organic matter processing and biological sediment mixing). Acknowledgements The study was supported by the Institute of Oceanology, Polish Academy of Sciences, and the National Science Centre, grant no. 2011/ 01/B/ST10/06529. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jmarsys.2015.12.009. References Ahrens, M.J., Morrisey, D.J., 2005. Biological effects of unburnt coal in the marine environment. Oceanogr. Mar. Biol. Annu. Rev. 43, 69–122.

146

M. Włodarska-Kowalczuk et al. / Journal of Marine Systems 157 (2016) 135–146

Akoumianaki, I., Papaspyrou, S., Nicolaidou, A., 2006. Dynamics of macrofaunal body size in a deltaic environment. Mar. Ecol. Prog. Ser. 321, 55–66. Aller, J.Y., Stupakoff, I., 1996. The distribution and seasonal characteristics of benthic communities on the Amazon shelf as indicators of physical processes. Cont. Shelf Res. 16, 717–751. Aller, J.Y., Alongi, D.M., Aller, R.C., 2008. Biological indicators of sedimentary dynamics in the central Gulf of Papua: seasonal and decadal perspective. J. Geophys. Res. 113, 1–16. Anderson, M.J., Gorley, R.N., Clarke, K.R., 2008. PERMANOVA+ for PRIMER: Guide to Software and Statistical Methods. Primer-E Ltd, Plymouth (214 pp). Attrill, M.J., Rundle, S.D., Fraser, A., Power, M., 2009. Oligochaetes as a possible entry route for terrigenous organic carbon into estuarine benthic food webs. Mar. Ecol. Prog. Ser. 384, 147–157. Austen, M.C., Warwick, R.M., Rosado, M.C., 1989. Meiobenthic and macrobenthic community structure along a putative pollution gradient in southern Portugal. Mar. Pollut. Bull. 20, 398–405. Austen, M.C., Parry, D.M., Widdicombe, S., Somerfield, P.J., Kendall, M.A., 2003. Macrofaunal mediation of effects of megafaunal bioturbation on nematode community structure. Vie Et Milieu-Life Environ. 53, 201–209. Bianchi, T.S., Alison, M.A., 2009. Large-river delta-front estuaries as natural ‘recorders’ of global environmental change. PNAS 106, 8085–8092. Bonifacio, P., Bourgeois, S., Labrune, C., Amouroux, J.M., Escoubeyrou, K., Byscail, R., Romero-Ramirez, A., Lantoine, F., Deflandre, B., Gremare, A., 2014. Spatiotemporal changes in surface sediemnt characteristics and benthic macrofauna composition off the Rhone River in realtion to its hydrological regime. Estuar. Coast. Shelf Sci. 151, 196–209. Chaalali, A., Beaugrand, G., Boët, P., Sautour, B., 2013. Climate-caused abrupt shifts in a European macrotidal estuary. Estuar. Coasts 36, 1193–1205. Chainho, P., Costa, J.L., Chaves, M.L., Lane, M.F., Dauer, D.M., Costa, M.J., 2006. Seasonal and spatial patterns of distribution of subtidal benthic invertebrate communities in the Mondego River, Portugal — a poikilohaline estuary. Hydrobiologia 555, 59–74. Chainho, P., Costa, J.L., Chaves, M.L., Dauer, D.M., Costa, M.J., 2007. Influence of seasonal variability in benthic invertebrate community structure on the use of biotic indices to assess the ecological status of a Portuguese estuary. Mar. Pollut. Bull. 54, 1586–1597. Cyberski, J., Grześ, M., Gurty-Korycka, M., Nachlik, E., Kundziewicz, W.W., 2006. History of floods on the river Vistula. Hydrol. Sci. J. 51 (5), 799–817. Damrat, M., Zaborska, A., Zajączkowski, M., 2013. Sedimentation from suspension and sediment accumulation rate in the Vistula River prodelta, Gulf of Gdańsk (Baltic Sea). Oceanologia 55 (4), 937–950. David, V., Sautour, B., Chardy, P., 2007. Successful colonization of the calanoid copepod Acartia tonsa in the oligo-mesohaline area of the Gironde estuary (SW France) — natural or anthropogenic forcing? Estuar. Coast. Shelf Sci. 71, 429–442. Doi, H., Matsumasa, M., Toya, T., Satoh, N., Mizota, C., Maki, Y., Kikucji, E., 2005. Spatial shifts in food sources for macrozoobenthos in an estuarine ecosystem: carbon and nitrogen stable isotope analyses. Estuar. Coast. Shelf Sci. 64, 316–322. Elliott, M., McLusky, D.S., 2002. The need for definitions in understanding estuaries. Estuar. Coast. Shelf Sci. 55 (6), 815–827. Emelyanov, E.M., Stryuk, V.L., 2002. Distribution of suspended matter. In: Emelyanov, E.M. (Ed.), Geology of the Gdańsk Basin. Baltic Sea. Monograph. Yantarnyi skaz, Kaliningrad, pp. 78–81. Evans, C.A., O'Reilly, J.E., 1982. A manual for the measurement of chlorophyll A, net phytoplankton, and nanoplankton: provisional copy for use on vessels participating in FIBEX. BIOMASS Scientific Series 9 (40 pp). Evans, C.A., O'Reilly, J.E., Thomas, J.P., 1987. A handbook for the measurement of chlorophyll-a and primary productivity. BIOMASS Scientific Series 8 (114 pp). GESAMP, 1993. Anthropogenic influences on sediment discharge to the coastal zone and environmental consequences. GESAMP Reports and Studies 52, 1–72. Grilo, T.F., Cardoso, P.G., Dolbeth, M., Bordalo, M.D., Pardal, M.A., 2011. Effects of extreme climate events on the macrobenthic communities' structure and functioning of a temperate estuary. Mar. Pollut. Bull. 62 (2), 303–311. Hedges, J., Stern, J.H., 1984. Carbon and nitrogen determination of carbonate-containing solids. Limnol. Oceanogr. 29, 657–663. Herman, P.M.J., Middelburg, J.J., van de Koppel, J., Heip, C.H.R., 1999. Ecology of estuarine macrobenthos. Adv. Ecol. Res. 29, 195–240. Ingole, B.S., Ansari, Z.A., Matondkar, S.G.P., Rodrogues, N., 1999. Immediate response of meio and macrobenthos to disturbance caused by a benthic disturber. Proceedings of the Third Ocean Mining Symposium. Goa, India, pp. 191–197. Janas, U., Wocial, J., Szaniawska, A., 2004. Seasonal and annual changes in the macrozoobenthic populations of the Gulf of Gdańsk with respect to hypoxia and hydrogen sulphide. Oceanologia 46, 85–102. Johansson, B., 1997. Tolerance of the deposit-feeding Baltic amphipods to oxygen deficiency. Mar. Ecol. Prog. Ser. 151, 135–141. Koszka-Maroń, D., 2009. Facies model of the contemporary delta lobe of the Vistula River. Oceanol. Hydrobiol. Stud. 38 (Suppl. 1), 57–68. Kuliński, K., Pempkowiak, J., 2008. Dissolved organic carbon in the southern Baltic Sea: quantification of factors affecting its distribution. Estuar. Coast. Shelf Sci. 78, 38–44. Maksymowska, D., Richard, P., Piekarek-Jankowska, H., Riera, P., 2000. Chemical and isotopic composition of the organic matter sources in the Gulf of Gdansk (Southern Baltic Sea). Estuar. Coast. Shelf Sci. 51, 585–598. McLusky, D.S., 1999. Estuarine benthic ecology: a European perspective. Austral Ecol. 24, 302–311. Meire, P.M., Seys, J.J., Ysebaert, T.J., Coosen, J., 1991. A comparison of the macrobenthic distribution and community structure between two estuaries in SW Netherlands. In:

Elliott, M., Ducrotoy, J.-P. (Eds.), Estuaries and Coasts: Spatial and Temporal Intercomparisons. ECSA Symposium. Olsen & Olsen, pp. 221–230. Montagna, P.A., Kalke, R.D., 1992. The effect of freshwater inflow on meiofaunal and macrofaunal populations in the Guadalupe and Nueces Estuaries, Texas. Estuaries 15, 307. Moore, P.G., 1977. Inorganic particulate suspensions in the sea and their effects on marine animals. Oceanogr. Mar. Biol. Annu. Rev. 15, 225–363. Obrebski, S., 1979. Larval colonizing strategies in marine benthic invertebrates. Mar. Ecol. Prog. Ser. 1, 293–300. Ochocki, S., Nakonieczny, J., Chmielowski, H., Zalewski, M., 1995. The hydrochemical and biological impact of the river Vistula on the pelagic system of the Gulf of Gdańsk. Primary production and chlorophyll a*. Oceanologia 37 (2), 207–226. Pasquaud, S., Béguer, M., Larsen, M.H., Chaalali, A., Cabral, H., Lobry, J., 2012. Increase of marine juvenile fish abundances in the middle Gironde estuary related to warmer and more saline waters, due to global changes. Estuar. Coast. Shelf Sci. 104–105, 46–53. Pastuszak, M., 1995. The hydrochemical and biological impact of the river Vistula on the pelagic system of the Gulf of Gdańsk Part 1. Variability in nutrient concentrations *. Oceanologia 37 (2), 181–205. Patrício, J., Adão, H., Neto, J.M., Alves, A.S., Traunspurger, W., Marques, J.C., 2012. Do nematode and macrofauna assemblages provide similar ecological assessment information? Ecol. Indic. 14, 124–137. Pearson, T.H., Rosenberg, R., 1978. Macrobenthic succession in relation to organic enrichment and pollution of the marine environment. Oceanogr. Mar. Biol. Annu. Rev. 16, 229–311. Pfannkuche, O., Thiel, H., 1988. Sample processing. In: Higgins, R.P., Thiel, H. (Eds.), Introduction to the Study of Meiofauna. Smithsonian Institution Press, Washington D.C., pp. 134–145. Pruszak, Z., van Ninh, P., Szmytkiewicz, M., Hung, N.M., Ostrowski, R., 2005. Hydrology and morphology of two river mouth regions (temperate Vistula Delta and subtropical Red River Delta). Oceanologia 47 (3), 365–385. Rhoads, D.C., Boesch, D.F., Zhican, T., Fengshan, X., Liqiang, H., Nilsen, K.J., 1985. Macrobenthos and sedimentary facies on the Changijang delta platform and adjacent continental shelf, east China Sea. Cont. Shelf Res. 4, 189–213. Scavia, D., Field, J.C., Boesch, D.F., Buddemeier, R.W., Burkett, V., Cayan, D.R., Fogarty, M., Harwell, M.A., Howarth, R.W., Mason, C., Reed, D., Royer, T.C., Sallanger, A.H., Titus, J.G., 2002. Climate change impacts on U.S. coastal and marine ecosystems. Estuaries 25 (2), 149–164. Schratzberger, M., Rees, H.L., Boyd, S.E., 2000. Effects of simulated deposition of dredged material on structure of nematode assemblages — the role of burial. Mar. Biol. 136, 519–530. Somerfield, P.J., Rees, H.L., Warwick, R.M., 1995. Interrelationships in community structure between shallow-water marine meiofauna and macrofauna in relation to dredgings disposal. Mar. Ecol. Prog. Ser. 127, 103–112. Struyf, E., Van Damme, S., Meire, P., 2004. Possible effects of climate change on estuarine nutrient fluxes: a case study in the highly nutrified Schelde estuary (Belgium, the Netherlands). Estuar. Coast. Shelf Sci. 60 (4), 649–661. Syvitski, J.P.M., Andrews, J.T., 1994. Climate change: numerical modeling of sedimentation and coastal processes, eastern Canadian Arctic. Arctic, Antarctic, and Alpine Research 26, 199–212. Szymczak-Żyła, M., Kowalewska, G., 2007. Chloropigments a in the Gulf of Gdańsk (Baltic Sea) as markers of the state of this environment. Mar. Pollut. Bull. 55, 512–528. Warwick, R.M., Platt, H.M., Clarke, K.R., Agard, J., Gobin, J., 1990. Analyses of macrobenthic and meiobenthic community structure in relation to pollution and disturbance in Hamilton harbour, Bermuda. J. Exp. Mar. Biol. Ecol. 138, 119–142. Wenne, R., Wiktor, K., 1982. Fauna denna przybrzeżnych wód Zatoki Gdańskiej. (Bottom fauna of coastal waters of the Gdańsk Bay). Stud. Mater. Oceanol. 39 (6), 137–171. Wieser, W., Ott, J., Schiemer, F., Gnaiger, E., 1974. An ecophysiological study of some meiofauna species inhabiting a sandy beach at Bermuda. Mar. Biol. 26, 235–248. Wijsman, J.W.M., Herman, P.M.J., Gomolu, M.-T., 1999. Spatial distribution in sediment characteristics and benthic activity on the northwestern Black sea shelf. Mar. Ecol. Prog. Ser. 181, 25–39. Witek, Z., Ochocki, S., Maciejowska, M., Pastuszak, M., Nakonieczny, J., Podgórska, B., Kownacka, J.M., Mackiewicz, T., Wrzesińska-Kwiecień, M., 1997. Phytoplankton primary production and its utilization by the pelagic community in the coastal zone of the Gulf of Gdansk (southern Baltic). Mar. Ecol. Prog. Ser. 148, 169–186. Włodarska-Kowalczuk, M., Pearson, T.H., Kendall, M.A., 2005. Benthic response to chronic natural physical disturbance by glacial sedimentation in an Arctic fjord. Mar. Ecol. Prog. Ser. 303, 31–41. Włodarska-Kowalczuk, M., Szymelfenig, M., Zajączkowski, M., 2007. Dynamic sedimentary environments of an Arctic glacier-fed river estuary (Adventfjorden, Svalbard). II. Meio- and macrobenthic fauna. Estuar. Coast. Shelf Sci. 74, 274–284. Ysebaert, T., Meire, P., Coosen, J., Essink, K., 1998. Zonation of intertidal macrobenthos in the estuaries of Schelde and Ems. Aquat. Ecol. 32 (1), 53–71. Ysebaert, T., Herman, P.M.J., Meire, P., Craeymeersch, J., Verbeek, H., Heip, C.H.R., 2003. Large-scale spatial patterns in estuaries: estuarine macrobenthic communities in the Schelde estuary, NW Europe. Estuar. Coast. Shelf Sci. 57, 335–355. Zajączkowski, M., Włodarska-Kowalczuk, M., 2007. Dynamic sedimentary environments of an Arctic glacier-fed river estuary (Adventfjorden, Svalbard). 1. Flux, deposition and dynamics of the sediments. Estuar. Coast. Shelf Sci. 74, 285–296. Zajączkowski, M., Darecki, M., Szczuciński, W., 2010. Report on the development of the Vistula river plume in the coastal waters of the Gulf of Gdańsk during the May 2010 flood. Oceanologia 52, 311–317. Żmudziński, L., 1990. Świat zwierzęcy Bałtyku. Atlas makrofauny. (The fauna of the Baltic Sea. Atlas of macrofauna). WSiP, Warszawa (196 pp).