Molluscan live and dead assemblages in an anthropogenically stressed shallow-shelf: Levantine margin of Israel

    Molluscan live and dead assemblages in an anthropogenically stressed shallow-shelf: Levantine margin of Israel Yael Leshno, Yael Edelman-Furstenberg, Henk Mienis, Chaim Benjamini PII: DOI: Reference:

S0031-0182(15)00265-5 doi: 10.1016/j.palaeo.2015.05.008 PALAEO 7276

To appear in:

Palaeogeography, Palaeoclimatology, Palaeoecology

Received date: Revised date: Accepted date:

22 September 2014 28 April 2015 19 May 2015

Please cite this article as: Leshno, Yael, Edelman-Furstenberg, Yael, Mienis, Henk, Benjamini, Chaim, Molluscan live and dead assemblages in an anthropogenically stressed shallow-shelf: Levantine margin of Israel, Palaeogeography, Palaeoclimatology, Palaeoecology (2015), doi: 10.1016/j.palaeo.2015.05.008

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT MOLLUSCAN LIVE AND DEAD ASSEMBLAGES IN AN ANTHROPOGENICALLY

PT

STRESSED SHALLOW-SHELF: LEVANTINE MARGIN OF ISRAEL

Yael Leshnoa,b, [email protected]

RI

* Yael Edelman-Furstenberga, [email protected]

SC

Henk Mienisc , [email protected]

Geological Survey of Israel, 30 Malchei Yisrael St., 95501, Jerusalem, Israel

b

Department of Geological and Environmental Sciences, Ben Gurion University of the Negev,

MA

a

NU

Chaim Benjaminib, [email protected]

Beer Sheva 84105, Israel

National Mollusk Collection at the Hebrew University of Jerusalem, Jerusalem 91904, Israel

Abstract

AC CE P

* Corresponding author

TE

D

c

Mismatches between live and dead molluscan assemblages associated with recent, rapid, anthropogenic changes, can be used to track these changes. The eastern Mediterranean is naturally oligotrophic, but recent urbanization of the Israeli coastal plain has enriched the littoral environment by injection of large amounts of treated wastewater onto the shelf. The largest point source is the Dan region wastewater project (Shafdan). Taxonomy and species rank order of abundance of modern (sediment-top) death assemblages were compared to live-collected mollusk assemblages, from two clean control stations (PL29, PL64) and a polluted site (PL3), near the Shafdan sewage sludge outlet at 36 m water depth. Seasonal variability was captured by

ACCEPTED MANUSCRIPT dredge and box-core sampling in winter (January), spring (May), summer (July) and fall (November) of 2012.

PT

Over 11,000 individuals of bivalves and gastropods were collected and analyzed. Diversity

RI

indices and nMDS and cluster analysis showed significant differences between live and dead assemblages from all stations and seasons. However, the rank order of the abundant live and

SC

dead species remained similar at each sampling station, resulting in high live-dead agreement of

NU

Jaccard-Chao index and Spearman’s rho similarity coefficient. Live-dead agreement of molluscan assemblages was preserved in the Shafdan area by the naturally high abundance of

MA

species that are tolerant to pollution, and by the annual dispersion of the sludge by winter storms, which prevent development of long-term anoxia. Moreover, the high frequency of storms in the

TE

D

winter of 2012 prevented a substantial amount of sludge from accumulating on the seafloor, explaining the similarity between the live assemblages of the polluted and the control sampling

AC CE P

sites. Live-dead agreement is a test that is too conservative to track the impact of the Shafdan sludge on the macrobenthic fauna in the Israeli shelf. Other tests, especially the increase in the abundance of deposit-feeding pollution-tolerant species in the live assemblage, show that there is an ongoing impact of the sludge on the benthic community.

1. Introduction Human modification of marine environments has led to the deterioration of sea-floor ecosystems worldwide (Magni, 2003; Halpern et al., 2008, 2012). The introduction over time of excess nutrients and contaminants to the biologically rich and economically valuable coastal areas has degraded marine biodiversity and changed the distribution of the marine biota (Jackson et al., 2001; Lötze et al., 2006). Despite a growing awareness of the danger to macrobenthic

ACCEPTED MANUSCRIPT communities, in most cases monitoring of live benthic assemblages starts only after the onset of human activity; and information on the natural conditions prior to modification is incomplete

PT

(Kidwell, 2008, 2013). A record of the community prior to disturbance is preserved in time–

RI

averaged death assemblages of shelled faunas that serve as a baseline for the composition and structure of pre-anthropogenic live communities (Kidwell, 2007). Death assemblages are formed

SC

by the constant accumulation of remains of dead shells of macrofauna in the sediment top. As

NU

shells from consecutive generations are mixed together by bioturbation, physical reworking and other taphonomic processes, they construct a time-averaged record that flattens out short-term

MA

variability and represent the summation of subannual and year-to-year variations in the live community (Kidwell, 2007, 2009).

TE

D

Quantitative tests in unaltered modern environments have shown that the composition and relative abundance of the local living community may be well preserved in molluscan death

AC CE P

assemblages (Kidwell and Bosence, 1991; Aller, 1995; Kidwell, 2001, 2002). Live-dead agreement of molluscan assemblages, termed live-dead fidelity, has been applied in recent years to determine the degree of anthropogenic impact to marine environments (Kidwell, 2009; Zuschin and Stachowitsch, 2009; Weber and Zuschin, 2013; Korpanty and Kelley, 2014). Dead specimens of macrobenthic fauna sieved from the top sediment are used to reconstruct an ecologic baseline, which are then compared to the local living fauna (Kidwell, 2007, 2013). Live-dead fidelity is measured by changes in the abundance and the presence or absence of species between the live and dead assemblages and can be quantified by univariate (comparison of species richness, evenness, dominance, changes in the relative abundance of species or their taxonomic similarity) and multivariate (dendrograms and ordinations of species abundances) methods. Rapid and intense anthropogenic changes create a time lag between the composition

ACCEPTED MANUSCRIPT and structure of the live assemblage and the local dead counterpart. In this case, the time lag of response of death assemblages results in a decrease in live-dead fidelity, rendering it valuable in

PT

environmental assessment (Kidwell, 2007). However, it has been considered a ‘conservative

RI

tool’, as not all areas with known anthropogenic stress exhibit low fidelity (Kidwell, 2007), and is more likely to fail to detect anthropogenic eutrophication than to falsely indicate it. Notably,

SC

most cited studies were limited to intertidal or shallow subtidal environments to ca. 20 m.

NU

Kidwell (2007) performed a global meta-analysis of 73 live-dead datasets to test the use of live-dead fidelity of molluscan assemblages as a proxy for ecological change. Live-dead

MA

similarities across a gradient of increasing anthropogenic eutrophication were presented in the form of a cross-plot in which live-dead taxonomic similarity (Jaccard-Chao index, y axis) was

TE

D

plotted against live-dead agreement in species rank order of abundance (Spearman’s rho, x axis) (figure 1 in Kidwell, 2007). Datasets from open shelf, pristine and naturally nutrient-poor

AC CE P

environments fell in the upper right quadrant of the cross plot; this is the field of high live-dead agreement. Datasets from areas of mild to severe anthropogenic impact showed a wider spread into the lower left quadrant of the cross plot, associated with lower live-dead agreement. Kidwell (2007) concluded that a Jaccard-Chao < 0.6 and Spearman’s rho < 0.1 can operationally distinguish areas impacted by anthropogenic eutrophication using live-dead disagreement. The Levantine basin is the southeastern-most part of the Mediterranean sea. The damming of the Nile River in 1965 reduced the flow of nutrients to the eastern Mediterranean, which then became highly-oligotrophic (Inman and Jenkins, 1984). Subsequently, urbanization of the Israeli coast has led to considerable nutrient input in the form of treated wastewater via direct outfalls (EEA, 2001, Kress et al., 2004). The largest single contributor of nutrients is the Dan region wastewater project (Shafdan) that has been discharging sewage sludge onto the shelf

ACCEPTED MANUSCRIPT at 38 m water depth, , since 1987 (Kress et al., 2004; Hyams-Kaphzan et al., 2009). This recent point-source eutrophication of the Israeli Mediterranean shallow proximal shelf is particularly

PT

suited to test for live-dead fidelity as a proxy for ecosystem modifications, presenting a unique

RI

opportunity for an actualistic study of live-dead agreement over an eutrophication gradient. Previous studies on the effects of the discharge of sewage sludge in the in the Levantine basin

SC

focused on living foraminifera (Hyams-Kaphzan et al., 2009) and soft-bodied macrofauna

NU

(polychaetes) (Kress et al, 2004). The present study on shelly macrobenthos evaluates the level of fidelity between the coastal community composition and relative abundance of modern

MA

(sediment-top) death assemblages and live-collected molluscan assemblages, at one polluted and two control stations near the Shafdan. The hypothesis that was tested was that the live-dead

TE

D

fidelity of the polluted station would be low as a result of the anthropogenic activity in the area, while the control stations would yield high live-dead fidelity. An important by-product of this

AC CE P

study was characterization of the live molluscan fauna at 35-40 m, depths at which molluscan communities in general are poorly known.

2. Study area

The Levantine coast of Israel trends in a N-NE - S-SW line over 180 km. The bathymetry of the southern sector is relatively simple, with contours aligned sub-parallel to the coastline. The width of the shelf is 25-30 km in the south, narrowing to ca. 10 km in the north (Inman and Jenkins, 1984; Rosentraub and Brenner, 2007). The upper 120 m of the Levantine water column is stratified during most of the year but becomes homogenous during winter, when surface cooling causes deepening of the mixed layer (Herut et al., 2000; Rosentraub and Brenner, 2007). High accumulation rates of sediments characterized the continental shelf of Israel (Schilman et

ACCEPTED MANUSCRIPT al., 2001) but were strongly reduced following the construction of the High Aswan Dam in 1965 (Stanley, 1988). The majority of transported sediments today are siliciclastics originating from

PT

the erosion of the Nile River delta, with a lesser contribution from local sources (Nir, 1984;

RI

Zviely et al., 2007).

The naturally oligotrophic Levantine basin is characterized by primary production

SC

estimated at ca.45 gC/m2y (Berman et al. 1984; Kress and Herut 2001; Kress et al. 2004) and

NU

Chlorophyll a values that range between 0.009 and 0.4 µg/l (Yacobi et al., 1995). The extreme nutrient depletion is caused by westward transport of deep water and nutrients as part of the anti-

MA

estuarine circulation in the Mediterranean (Coll et al., 2010). The Shafdan project treats the domestic and industrial waste water of the population of 2

TE

D

million residents in the Tel-Aviv Metropolitan area, producing 292,000 dry tons of activated sewage sludge annually (biosolids after secondary treatment), 55% of which is redirected to

AC CE P

agriculture and landfill. Since 1987, some 16,000 m3/day of excess sludge produced at the Shafdan has been discharged through a single seabed pipe line, emptying 5 km offshore 15 km to the south of Tel Aviv (Kress et al., 2004) (Fig. 1). The activated sewage sludge is composed mostly of organic biomass and nutrients, ca. 1% non-degraded particulate matter, and may contain bacteria (including pathogens), synthetic organic compounds, and some heavy metals (Kress et al., 2004). The area was not polluted prior to the outfall activation (Galil and Lewinsohn, 1981). Permanent polluted and control monitoring stations have been established in the area (Kress et al., 2004; Hyams-Kaphzan et al., 2009), and the present sampling program took place at three of these stations: one polluted station, PL3, located 200 m NE of the outfall at 36 m water depth; control station PL29 at 5.5 km NE of the outfall at 34 m water depth, and control station PL64, located 7 km N-NE of the outfall at 35 m water depth (Fig. 1). The sludge

ACCEPTED MANUSCRIPT was continuously injected from the Shafdan, yet there is no long term accumulation of sludge from year to year (Kress et al., 2004; Hyams-Kaphzan et al., 2009). The thickness of the sludge

PT

layer on the seafloor at the polluted station PL3 depends on the seasonal storm and current

RI

regimes (Kress et al., 2004), or biotic consumption, that tend to rework and reduce the concentration of the organic particles in the sediment (Kress et al., 2004; Hyams-Kaphzan et al.,

NU

SC

2009).

3. Methods

MA

3.1. Field methods

Sampling of the control (PL3) and polluted stations (PL29, PL64) was undertaken in the

TE

D

2012 winter (Jan), spring (May-Jun), summer (Jul) and fall (Nov) cruises of R/V Etziona, with the aim of capturing one year of seasonal variability of live species. Polluted station PL3 and

AC CE P

control station PL29 were sampled on all cruises. To ensure that the control station PL29 was indeed well outside the zone of contamination, control station PL64 located 1.5 km further to the northeast was added in the spring and fall samplings of 2012. Triplicate samples of sediment containing dead mollusks were collected from the top 1.5 cm of a 25X25 cm2 GOMEX boxcorer or a 28X36 cm2 Van Veen grab at each of the sampling stations on all cruises. All boxcorer or grab samples were additionally examined for the presence of live mollusks. Duplicate samples for live mollusks were collected by dragging a dredge (45X60X8 cm3) over a 30 m seafloor transect in each station. Sampling volume was designed to account for differences in number of specimens between living and time-averaged death assemblages. Since sediment samples contain on average ca. 10 times more dead than live specimens, death assemblages usually have a higher number of species than the local living assemblage (Kidwell, 2013). The

ACCEPTED MANUSCRIPT large quantities of sediment containing live mollusks ensured that an adequate number was collected for statistical inference.

PT

During each sampling cruise, environmental variables measured in the water column at

RI

each station were O2 concentration, and temperature and salinity measured by CTD. Organic enrichment was indicated by %TOC (dry wt.) from short sediment cores. On board, sediment

NU

SC

samples were wet sieved over 2 mm and the residue preserved in sea water.

3.2. Laboratory methods

MA

On shore, samples were again washed in the lab over 2 mm sieves and preserved in 96% ethanol within days of sampling. Samples were dried at 50˚C prior to picking and sorting.

TE

D

Identification was to the species level with notation as to whether individuals were live or dead at the time of sampling. Broken mollusk shells that retained at least half of the bivalve hinge line or

AC CE P

gastropods with apertures were counted as individuals. Specimens counted as live individuals had adhering soft tissue or were articulated, or had attached operculum for gastropods. Taxonomic identification of molluscan fauna was based on Barash and Danin (1992), Poppe and Goto (1993) and the World Register of Marine Species (WoRMS). The presence of introduced alien (e.g., Lessepsian) species was noted according to the CIESM atlas of exotic species (Zenetos et al., 2003; Galil, 2007). In addition, functional groups were assigned to each taxa (Todd, 2001 http://porites.geology.uiowa.edu/database/mollusc/mollusclifestyles.htm; MarLIN, 2006. BIOTIC- Biological Traits Information Catalogue [WWW Document]. URL http://www.marlin.ac.uk.; WoRMs).

ACCEPTED MANUSCRIPT 3.3. Data analysis Univariate and multivariate statistical analyses were performed on the live and dead

PT

assemblages within and between control and polluted sampling stations, and seasons.

RI

Univariate diversity indices were calculated on the raw abundance data of species, values were averaged for triplicate samples of death assemblages and duplicates of live assemblages.

SC

Species richness was calculated by Margalef’s richness index (d) since it takes into account both

NU

the number of species and the total number of individuals, and is relatively unaffected by sample size (Clarke and Warwick, 1994; Magurran, 2004). Shannon–Wiener index (H’) was used as it is

MA

a common measure of species diversity (Salas et al., 2006). The Shannon-Wiener index combines the relative abundance of species and the number of taxa found. Evenness was

TE

D

calculated using Hulbert’s (1971) probability of interspecific encounters (PIE) index, which is independent of sample size (Olszewski and Kidwell, 2007). Taxonomical diversity of the

AC CE P

different sample sizes of the live and dead assemblages was standardized using rarefaction curves (PAST software package; Hammer et al., 2001). Multivariate analyses were performed on pooled species abundances of duplicate liveand triplicate death- assemblages. Low-volume dredge samples with an insufficient number of live individuals (under 20; Kidwell, 2002) were excluded from the analysis. For statistical analysis, PRIMER v.6 of the Plymouth Marine Laboratory (Clarke and Warwick, 1994) was employed. Due to the different sample sizes, abundances were standardized by the total amount of individuals, and a log(X+1) transformation was used to reduce the influence of dominant species (Clarke and Warwick, 1994; Clarke et al., 2008). Bray-Curtis similarity coefficient was calculated for the analysis of taxonomic differences between the live and dead assemblage. Similarity percentage analysis (SIMPER) was performed to identify the species that contribute to

ACCEPTED MANUSCRIPT the differences between the live and dead assemblages. The R statistic of analysis of similarity (ANOSIM) was used to test for statistically significant differences between and within the

PT

sampling stations and seasons of the live and dead assemblages. Cluster analysis with the

RI

SIMPROF test was superposed on the nMDS figure.

Taxonomic similarity between the live and the dead assemblages at each sampling station

SC

was measured by the Jaccard-Chao index that ranges from 0 (no shared species) to 1 (all species

NU

in the two lists are shared) (Chao et al., 2005), calculated using EstimateS 9.0.0 software. Changes in the relative abundance of species between the live and the dead assemblages were

MA

evaluated by Spearman’s rho correlation coefficient of rank order agreement, ranging from -1 (species in one list are ranked in opposite order to the other according to their relative

TE

D

abundance) to +1 (taxa are ranked in identical order in the two lists). Spearman correlation was computed at www.wessa.net/rankcorr.wasp based on raw abundances of species. Jaccard-Chao

AC CE P

taxonomic similarity index and Spearman’s rho rank order coefficient were calculated for livedead assemblages for each station, pooling all the data for all seasons. Different combinations of live dead comparisons (pooled, individual samples, all stations, single station, etc.) were conducted and all showed the same trend. However, in order to enable a comparison to the procedures of Kidwell (2007), data from the season with the largest list of live species of the 2012 sampling was then compared to its corresponding death assemblage. In addition, the season with the largest live assemblage of 2012 was compared to its corresponding death assemblage collected 10 years earlier from the same polluted (PL3) and control station (only PL29). This enabled an additional point of comparison to samples from summer 2003 to winter 2004, representing an assemblage that maintained some of its characteristics before the injection of the sludge (“pre-impacted” fauna), but was also somewhat affected by the sludge. Death

ACCEPTED MANUSCRIPT assemblages sampled in 2003-4 were sieved over 1 mm, as opposed to 2 mm of the 2012 samples, which increased the number of larvae and young juveniles in the assemblages. The

PT

species between 1-2 mm in the 2003-4 dataset were mostly rare species that accounted for under

RI

1% of the total number of individuals, and thus were excluded from the analysis.

SC

4. Results

NU

4.1. The molluscan fauna

The molluscan taxa identified from 86 samples of live and dead assemblages from the

MA

polluted and control stations in the 2012 excursions are listed in Table 1. A total of 11,280 individuals were identified, belonging to 58 bivalve and 43 gastropod species. Among them,

TE

D

3764 (33%) individuals were live, belonging to 16 bivalve and three gastropod species (Table 2). Ten alien (Lessepsian) species of bivalves and eight of gastropods were identified (marked * in

AC CE P

Table 1). Both live and dead assemblages were strongly dominated by bivalves (Table 2), especially by Corbula gibba. Gastropods were scarce in the live and dead assemblages, with a maximum of 9% of total individuals in the dead-, and a maximum of only 1% in the liveassemblage. Live gastropods were found only in the live assemblages of the control stations in spring (June). The very few found (only 23 individuals, Table 2) were all young and retained some of the shell’s original color. All assemblages were dominated by bivalves (90-95%) of which 99% were infaunal in both the live and dead assemblages (Appendix 1). Nearly all gastropods were carnivores (mostly empty shells of Nassarius pygmaeus). Fig. 2 shows that most of the live bivalves at the polluted station were deposit feeders (55-88%), while live bivalves at the control stations were both deposit (44-66%) and filter feeders (34-44%).

ACCEPTED MANUSCRIPT

4.2. Environmental characteristics of the stations

PT

A plume of sludge was visible at the water surface of polluted site PL3 as far as a few

RI

hundred meters from the point source. The sludge was nearly undetectable in most of the sediment cores of 2012, but sediments sampled from PL3 had a dark color and a strong organic

SC

odor. Three of the sampling cruises took place up to two weeks following events of strong

NU

storms that dispersed the sludge. The maximum thickness of the sludge layer observed in 2012 was only 3 cm in summer (July), and nearly absent the rest of the year.

MA

Sampling sites were located relatively close to one another and at similar depths. Therefore, the temperature and salinity profiles were similar between the polluted and control

TE

D

stations, and showed seasonal stratification or mixing of the water column. TOC (dry wt. %) in the sediment cores were higher by an order of magnitude at the polluted station vs. the control

AC CE P

stations (Table 3).

4.3. Diversity indices

The Margalef index of all the live assemblages (Appendix 2) averaged 1.0±0.4 and remained similar between the sampling stations and seasons (Fig. 3). Likewise, the ShannonWiener index of all the live assemblages averaged 1.2±0.2 across stations and seasons (Fig. 4). Diversity indices of the death assemblage also remained relatively constant across the different stations and seasons, with an average Margalef index of 3.0±0.4 (Fig. 3) and an average Shannon-Wiener of 1.8±0.2 (Fig. 4). Thus, both indices were significantly lower in the live than in the dead assemblage. The PIE evenness index (Fig. 5) did not show a significant difference between live and dead assemblages.

ACCEPTED MANUSCRIPT Rarefied species richness based on pooled data for the live and dead assemblages was significantly higher in the dead assemblage, indicating that diversity differences between the

PT

assemblages were not influenced by sample size (Fig. 6). The dead rarefaction curve reached a

RI

plateau, meaning the majority of the abundant species in the assemblage were sampled, while the

SC

live rarefaction curve is less pronounced.

NU

4.4. Taxonomic composition and rank order of abundance

SIMPER test showed a high similarity percentage between the live and dead assemblages

MA

(59%). The identity of abundant species (species contributing to >90% of total number of individuals) was similar in all assemblages, and the majority of live species were also found dead

TE

D

(76%). Despite similarity of the abundant species in both the live and dead assemblages, rare species found dead were not found live.

AC CE P

Similarity within the live assemblages was high (78%). The most abundant species in the live assemblage were Corbula gibba (33%), Nuculana pella (31%), Nucula nitidosa (18%) and Abra longicallus (11%). The relative abundance of a few species varied seasonally with the accumulation and dispersion of the sewage sludge layer at the polluted station, PL3. However, the rank order of abundance was preserved between the polluted and control sampling stations throughout the year. The death assemblages showed a high agreement in taxonomic composition within and between sampling stations and seasons. The death assemblages were dominated by Corbula gibba (21%), Abra longicallus (11%), Pitar rudis (11%), Nucula nitidosa (11%), Nuculana pella (10%), Anadara polii (8%) and one gastropod species, Nassarius pygmaeus (10%). The relative abundance of N. pella was much higher in the live assemblage compared to the dead (31% and 10%, respectively). The bivalves A. polii, P. rudis, and Acanthocardia

ACCEPTED MANUSCRIPT paucicostata of the dead assemblage drastically decrease in the live assemblage, while the relative abundance of C. gibba, N. pella and N. nitidosa increases significantly.

PT

High values of Jaccard-Chao index indicate the high taxonomic similarity of species

RI

between the live and the dead assemblages at each of the sampling stations, ranging from 0.76 to 0.89 (Table 4). Spearman's rho rank order correlation coefficient showed positive correlations

SC

between all live and dead comparisons at each of the sampling stations. All of the rank-order

NU

correlations were statistically significant (p < 0.05). Spearman’s rho ranged from 0.60 to 0.75 at the polluted station, and from 0.12 to 0.30 at the control station for all live-dead comparisons.

MA

Similarly, significant rank order correlation was found between the live assemblages of 2012 to their corresponding dead assemblage from 2003-4 for the polluted (PL3) and control (PL29)

TE

D

stations.

Figure 7 is a cross-plot of live-dead comparisons of taxonomic similarity and rank order

AC CE P

of abundance from each of the sampling stations in the present study. Following Kidwell (2007), for each of the polluted and control stations in this study, the season with the largest number of live individuals was compared to its corresponding dead assemblage (Table 4). Counterintuitively , live-dead comparisons of the Shafdan polluted and control stations both plot in the upper right quadrant, indicating high live-dead similarity, and do not enter the quadrants associated with anthropogenic eutrophication. Moreover, dead assemblages collected 10 years earlier from the same polluted and control stations also plot in the same quadrant as the live-dead assemblages of 2012.

ACCEPTED MANUSCRIPT 4.5. Multivariate analysis of live-dead assemblages Cluster and nMDS analyses based on Bray-Curtis similarity coefficient showed that live

PT

and dead assemblages formed two clusters of samples at a similarity level of 59% (Fig. 8). One

RI

cluster contained all the death assemblages from all stations and seasons; SIMPROF test indicated there was no significant sub-division within the cluster. The second cluster contained

SC

all of the samples of the live assemblages, which was further divided into two sub-groups. Death

NU

assemblages from all stations and seasons plotted very close to each other, while the live assemblages were heterogeneous and were scattered over the plot, especially the spring live

MA

assemblage of PL3 that plotted in the upper high left corner. Heterogeneity of live assemblages was correlated to season; fall-winter and spring-summer.

TE

D

Analysis of similarity (ANOSIM) also indicated statistically significant differences between clusters of the live and dead assemblages (Table 5). There was no significant difference

AC CE P

between seasons and sampling stations within the death assemblage nor between sampling stations within the live assemblage. However, seasonal significance was found between the live samples.

Dead assemblages from the same stations sampled in 2003-4, showed high similarity in species relative abundance and composition to the dead assemblages of the present study, separating into two clusters at 82% similarity. Comparison of the dead assemblages of 2003-4 to the live assemblages of 2012, showed a similar cluster pattern to that of the 2012 live-dead samples (Fig. 9).

ACCEPTED MANUSCRIPT 5. Discussion 5.1. Environmental conditions in 2012 and 2003-4

PT

The polluted station and the control stations were similar in terms of water depth,

RI

temperature and salinity. The seasonal cycle of accumulation and dispersion of sludge in 2012 differed from that of less stormy years, as the frequent storms that year did not allow sludge to

SC

accumulate. Although the Shafdan sewage discharge was visible as a plume at the water surface

NU

around the injection site, the maximum accumulation of sludge at the polluted station in 2012 was only 3 cm and was rapidly dispersed by storms. A sludge layer of 8.5 cm of sludge was

MA

measured in 2003, and the sea floor was rendered anoxic (Hyams-Kaphzan et al., 2009). Anoxic conditions did not develop in 2012, despite high TOC measured at the polluted site compared to

TE

D

the control stations (Table 3). In fact, maximum TOC throughout 2012 at the polluted station

AC CE P

(1.1-3.5%) was much lower relative to past records (0.8-16%) (Hyams-Kaphzan et al., 2009). .

5.2. The molluscan composition in the research area The data shows that the composition and relative abundance of species of the live assemblage remained consistent between the polluted and control stations (Table 5; Fig. 3, 4 and 5) indicating the natural durability of local species to extreme conditions. A few species did become more abundant seasonally, apparently tracking the accumulation and dispersion of sludge (Table 2). The death assemblage averaged out all seasonal, reproductive or short-term effects, such that no significant differences were detected between sampling stations or seasons (Table 4 and 5; Fig. 3, 4 and 5). Only minor differences were detected between the 2003-4 and 2012 death assemblages, due to time-averaging at all the sites and the time-lag of response of death

ACCEPTED MANUSCRIPT assemblages to environmental stress at the polluted site. For the present, therefore, the death assemblages record long-term relative abundance and composition regardless of the injection of

PT

sludge.

RI

The most abundant species were similar in the live and dead assemblages (SIMPER test). The similarity in the composition and rank order of live-dead assemblages (Table 4) may be

SC

attributed to a persistent community structure, where the abundant species and those ranked in

NU

the middle of the assemblage hardly changed. The Jaccard-Chao and Spearman’s rho are especially sensitive to drastic changes in the rank and identity of species, where environmental

MA

impact causes numerically abundant species to become rare (or absent), and vice versa (Kidwell, 2009). However, in the Shafdan the relative abundance of the dominant species did not change

D

drastically and their relative rank was kept high in both the live and dead assemblages, resulting

TE

in high Jaccard-Chao and Spearman’s rho values. The persistence of the species composition in

AC CE P

the stressed site may be explained in two ways: tolerance to the stress conditions, or rapid, effective repopulation after storm events with improved oxygen conditions, enabling utilization of the excess organic matter as a food resource (e.g., Levin and Gage, 1998). In particular, the known tolerant species Corbula gibba and Nuculana pella were highly dominant both in the live and the dead assemblages. C. gibba was the most common mollusk in nearly all samples of both live and dead assemblages. It is reportedly tolerant of low oxygen conditions on the sea floor (Diaz and Rosenberg, 1995) and polluted environments (Borja et al., 2000) where it can form over 80% of the benthic biomass (Hrs-Brenko, 1981, 2006; Holmes and Miller, 2006). N. pella is an opportunistic, deep burrowing deposit feeder that inhabits natural organically-enriched sediments , and tolerates organic input of anthropogenic origin (WoRMS; Simboura and Zenetos, 2002).

ACCEPTED MANUSCRIPT

5.3. Fidelity with respect to uni- and multivariate analysis

PT

Fidelity with respect to species richness was high; 76% of species found live were also

RI

found in the dead assemblage. Diversity indices were significantly higher in the dead than in the live assemblage (Fig. 3, 4) and rarefaction curves also showed higher rarefied species richness in

SC

the dead assemblage (Fig. 6). However, the PIE evenness index (Fig. 5) did not show a

NU

significant difference between live and dead assemblages, as described previously by Olszewski and Kidwell (2007). Most live-dead studies have been based on a single sampling of the live

MA

assemblage, yielding underestimation of live diversity (Kidwell, 2007). This study combined sampling over four seasons, collection of live individuals from a large volume of sediment, and

TE

D

preferential use of diversity indices that are unaffected by sample size. This sampling effort was aimed to reduce the bias between the live and dead assemblages caused by time averaging of the

AC CE P

dead assemblage. . These differences point to changes caused by the Shafdan pollution and not just by time averaging of the dead assemblages. Multivariate analyses showed a significant difference between the live and dead assemblages (Fig. 8). Live and dead assemblages grouped into statistically significant separate clusters at a similarity level of 59% (Table 5). Live and dead assemblages formed distinct, separate groups also in the nMDS. The live samples were widely scattered, while the dead plotted close together (Fig. 8). Time-averaging alone cannot explain this observed separation of centroids of live and dead assemblages in the nMDS. Under natural conditions, time-averaging reduces the dispersion of the dead samples around their centroid (Tomašových and Kidwell, 2011). Thus, the way the samples of the dead assemblage are well separated from the centroid of

ACCEPTED MANUSCRIPT the live assemblage (Fig. 8) can be explained, for example, by environmental stress

PT

(Tomašových and Kidwell, 2011).

RI

5.4. Live-dead fidelity across stress gradients

The diversity indices and multivariate analyses indicated significant differences between

SC

live-dead assemblages (Appendix 3). On the other hand, using Kidwell’s (2007) meta-analysis

NU

paradigm shows a high live-dead agreement of molluscan assemblages in both the control and polluted sampling stations (Fig. 7). Such high fidelity may be preserved where anthropogenic

MA

stresses producing ecological changes remain within the natural tolerance of the pre-impacted fauna (Kidwell, 2007, 2013). In the area influenced by the Shafdan, high live-dead agreement is

TE

D

preserved because sewage-sludge is dispersed annually, in winter by storms and currents, and in spring by consumption by Capitellidae polychaetes (Kress et al., 2004) that use the sludge as a

AC CE P

food resource (Hyams-Kaphzan et al., 2009). Due to this annual cycle, the already existing and locally abundant species that can tolerate organic load and low oxygen conditions (e.g. Corbula gibba) either survive or re-populate the area, maintaining the high agreement between live-dead molluscan assemblages.

It is also possible that the high live-dead fidelity may be the result of temporal autocorrelation, in which large numbers of recently dead individuals may dominate the death assemblage (Tomašových and Kidwell, 2009, 2011). In addition, Kidwell’s (2007) analysis is a ‘conservative test’, meaning it is more likely to fail to detect anthropogenic eutrophication than to falsely indicate it. Thus, fidelity between the live and dead assemblages in the Shafdan area can be high despite the known anthropogenic eutrophication in the area.

ACCEPTED MANUSCRIPT Kidwell (2007) considered all the datasets from the Mediterranean shelves to be moderately impacted through increasing population and coastal development over at least 2000

PT

years of anthropogenic activity. The Gulf of Trieste, located in the northern Adriatic, is one of

RI

the most degraded marine ecosystems worldwide, also with a long history of human impact (Lotze et al., 2006). Both low and high fidelity of the live-dead molluscan assemblages of the

SC

Gulf of Trieste were attributed to time-averaging (Weber and Zuschin, 2013). The composition

NU

of species in the live and dead assemblages of the Gulf of Trieste was similar, but uni- and multivariate methods showed significant difference between live and dead assemblages. Despite

MA

evidence for live-dead dissimilarity, the low fidelity was still attributed to within habitat timeaveraging, in that the long history of human changes in the area induced a new equilibrated

TE

D

benthic fauna (Weber and Zuschin, 2013). The Shafdan system on the other hand, did not experience long-term changes in relative abundance and/or composition of species. The high

AC CE P

live-dead fidelity may be attributed to the relatively short period since initiation of sludge injection, which has not yet effected a significant change in the benthic system.

5.5. Functional groups of bivalves in the live and dead assemblages The low diversity of the live assemblage and the species composition indicate that taxa that tolerate the effects of anthropogenic eutrophication in the Shafdan area are favored. Livedead comparison of the different feeding guilds show a decline in abundance of filter feeders and increase of deposit feeding species in the live assemblage in both the control and polluted stations (Fig. 2). Deposit feeders usually thrive in organically-enriched sediments and under oxygen stress (Pearson and Rosenberg, 1978), causing a shift in relative abundance of feeding strategies in areas of anthropogenic eutrophication (Kidwell, 2007, 2009). Shells of filter feeding

ACCEPTED MANUSCRIPT bivalves of the dead assemblage, such as Anadara polii, Pitar rudis, Acanthocardia paucicostata, and Azorinus chamasolen, were replaced in the live assemblage by the tolerant

PT

deposit feeders Nuculana pella and Nucula nitidosa in all stations, indicating impact of the

RI

introduced sewage sludge. Therefore, the relative increase of deposit feeders among the live species, along with low diversity of the live assemblage, and the large number of species limited

NU

SC

to the death assemblage, all point to influence of the injection of sewage sludge on the live fauna.

6. Conclusions

MA

This study is the first actualistic test of live-dead fidelity as a proxy for health conditions in the Levantine basin. The time-averaged death assemblages were found to be homogenous

TE

D

across seasons and sampling stations and also compared to samples from ten years earlier. The live assemblages were also similar in both the polluted and control stations.

AC CE P

Diversity indices and multivariate analyses showed a statistically significant difference between the live and dead assemblage. Despite evidence for low live-dead fidelity that resulted from the pollution (which probably exceeded the effect of time-averaging), the composition of the abundant species remained relatively constant between live-dead assemblages, and the correlation coefficient of rank order of species was found to be high. The high live-dead agreement preserved in the eastern Mediterranean is likely the result of the annual cycle of sludge removal by storms, which aerates the seafloor, together with the ability of several local living species to withstand anthropogenic eutrophication. In particular, the high frequency and relatively early storms during 2012 prevented sludge from accumulating at the polluted site that year. The discharge of sludge may, in addition, not have been operating long enough for the assemblages to reach equilibrium.

ACCEPTED MANUSCRIPT Although live-dead fidelity of molluscan assemblages was unable to show a definite impact of the injection of sewage sludge, the impact was indeed evident in the differences

PT

detected between feeding habits of the live and dead assemblages. The loss of sensitive species

RI

and dominance of tolerant deposit feeding species in the live assemblages demonstrates that the pollution impacts the molluscan community - and can be tracked. The robustly defined

SC

molluscan death assemblage can serve as a reliable baseline for future monitoring studies in the

NU

eastern Mediterranean.

MA

Acknowledgments

We acknowledge the assistance of the crew of the R/V Etziona for help in sampling. Ahuva

TE

D

Almogi-Labin is thanked for helpful and constructive comments. G. Dietl and D. Bottjer are thanked for their reviews, which improved the manuscript. This research was funded by the

AC CE P

Israel Ministry of National Infrastructures.

7. References

Aller, J.Y., 1995. Molluscan death assemblages on the Amazon Shelf: Implication for physical and biological controls on benthic populations. Palaeogeogr. Palaeoclimatol. Palaeoecol. 118, 181–212. doi:10.1016/0031-0182(95)00003-4 Barash, A., Danin, Z., 1992. Fauna Palaestina Mollusca I- Annotated List of Mediterranean Molluscs of Israel and Sinai. Publications of the Israel Academy of Sciences and Humanities Section of Sciences, Jerusalem. Beesley, P.L., Ross, G.J.B., Wells, A., 1998. Mollusca: The Southern Synthesis. CSRIO, Melbourne.

ACCEPTED MANUSCRIPT Berman, T., Townsend, D.W., EL Sayed, S.Z., Trees, C.C., Azov, Y., 1984. Optical transparency, chlorophyll and primary productivity in the eastern Mediterranean near the

PT

Israeli coast. Oceanol. acta 7, 367–372.

RI

Borja, A., Franco, J., Erez, V.P., 2000. A marine biotic index to establish the ecological quality

SC

of soft-bottom benthos within European estuarine and coastal environments. Mar. Pollut. Bull. 40, 1100–1114. doi:10.1016/S0025-326X(00)00061-8.

NU

Chao, A., Chazdon, R.L., Colwell, R.K., Shen, T.-J., 2005. A new statistical approach for

MA

assessing similarity of species composition with incidence and abundance data. Ecol. Lett. 8, 148–159. doi:10.1111/j.1461-0248.2004.00707.x.

D

Clarke, K.R., Somerfield, P.J., Gorley, R.N., 2008. Testing of null hypotheses in exploratory

TE

community analyses: Similarity profiles and biota-environment linkage. J. Exp. Mar. Bio.

AC CE P

Ecol. 366, 56–69. doi:10.1016/j.jembe.2008.07.009. Clarke, K.R., Warwick, R.M., 1994. Change in marine community: an approach to statistical analysis and interpretation. Plymouth Marine Laboratory, UK. Coll, M., Piroddi, C., Steenbeek, J., Kaschner, K., Ben Rais Lasram, F., Aguzzi, J., Ballesteros, E., Bianchi, C.N., Corbera, J., Dailianis, T., Danovaro, R., Estrada, M., Froglia, C., Galil, B.S., Gasol, J.M., Gertwagen, R., Gil, J., Guilhaumon, F., Kesner-Reyes, K., Kitsos, M.S., Koukouras, A., Lampadariou, N., Laxamana, E., López-Fé de la Cuadra, C.M., Lotze, H.K., Martin, D., Mouillot, D., Oro, D., Raicevich, S., Rius-Barile, J., Saiz-Salinas, J.I., San Vicente, C., Somot, S., Templado, J., Turon, X., Vafidis, D., Villanueva, R., Voultsiadou, E., 2010. The biodiversity of the Mediterranean Sea: Estimates, patterns, and threats. PLoS One 5, 1–36. doi:10.1371/journal.pone.0011842.

ACCEPTED MANUSCRIPT Costello, M.J., Bouchet, P., Boxshall, G., Arvantidis, C., Appeltans, W., 2008. European Register of Marine Species [WWW Document]. URL

PT

http://www.marbef.org/data/erms.php (accessed December 2012).

RI

Diaz, R.J., Rosenberg, R., 1995. Marine benthic hypoxia: A review of its ecological effects and

SC

the behavioral responses of benthic macrofauna. Oceanogr. Mar. Biol. 33, 245–303. EEA, 2001. Eutrophication in Europe’s coastal waters. Copenhagen. doi: topic report 7/2001.

NU

Galil, B., 2007. Seeing red: Alien species along the Mediterranean coast of Israel. Aquat.

MA

Invasions 2, 281–312. doi:10.3391/ai.2007.2.4.2.

Galil, B.S., Lewinsohn, C.H., 1981. Macrobenthic communities of the eastern Mediterranean

TE

D

continental shelf. Mar. Ecol. 2, 343–352. doi:10.1111/j.1439-0485.1981.tb00276.x. Halpern, B.S., Longo, C., Hardy, D., McLeod, K.L., Samhouri, J.F., Katona, S.K., Kleisner, K.,

AC CE P

Lester, S.E., O’Leary, J., Ranelletti, M., Rosenberg, A.A., Scarborough, C., Selig, E.R., Best, B.D., Brumbaugh, D.R., Chapin, F.S., Crowder, L.B., Daly, K.L., Doney, S.C., Elfes, C., Fogarty, M.J., Gaines, S.D., Jacobsen, K.I., Karrer, L.B., Leslie, H.M., Neeley, E., Pauly, D., Polasky, S., Ris, B., St Martin, K., Stone, G.S., Sumaila, U.R., Zeller, D., 2012. An index to assess the health and benefits of the global ocean. Nature 488, 615– 620. doi:10.1038/nature11397. Halpern, B.S., Walbridge, S., Selkoe, K. a, Kappel, C. V, Micheli, F., D’Agrosa, C., Bruno, J.F., Casey, K.S., Ebert, C., Fox, H.E., Fujita, R., Heinemann, D., Lenihan, H.S., Madin, E.M.P., Perry, M.T., Selig, E.R., Spalding, M., Steneck, R., Watson, R., 2008. A global map

of

human

impact

doi:10.1126/science.1149345.

on

marine

ecosystems.

Science

319,

948–952.

ACCEPTED MANUSCRIPT Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST: Paleontological Statistics Software Package for Education and Data Analysis. Palaeontologia Electronica 4, 9 (http://palaeo-

PT

electronica.org/2001_1/past/issue1_01.htm).

RI

Heller, J., 2011. Marine Molluscs of the Land of Israel. Alon Sefer, Israel.

SC

Herut, B., Almogi-Labin, A., Jannink, N., Gertman, I., 2000. The seasonal dynamics of nutrient and chlorophyll a concentrations on the SE Mediterranean shelf-slope. Oceanol. Acta 23,

NU

771–782. doi:10.1016/S0399-1784(00)01118-X.

United Kingdom 46, 401–493.

MA

Holme, N.A., 1966. The bottom fauna of the English Channel. Part II. J. Mar. Biol. Assoc.

D

Holmes, S.P., Miller, N., 2006. Aspects of the ecology and population genetics of the bivalve

TE

Corbula gibba. Mar. Ecol. Prog. Ser. 315, 129–140.

AC CE P

Hrs-Brenko, M., 1981. Population studies of Corbula gibba (Olivi) Bivalvia, Corbulidae, in the northern Adriatic Sea. J. Molluscan Stud. 47, 17–24. Hrs-Brenko, M., 2006. The basket shell, Corbula gibba Olivi, 1792 (bivalve mollusks) as a species resistant to environmental disturbances: A review. Acta Adriat. 47, 49–64. Hurlbert, S.H., 1971. The nonconcept of species diversity: A critique and alternative parameters. Ecology 52, 577-586. doi:10.2307/1934145. Hyams-Kaphzan, O., Almogi-Labin, A., Benjamini, C., Herut, B., 2009. Natural oligotrophy vs. pollution-induced eutrophy on the SE Mediterranean shallow shelf (Israel): Environmental parameters and benthic foraminifera. Mar. Pollut. Bull. 58, 1888–1902. doi:10.1016/j.marpolbul.2009.07.010.

ACCEPTED MANUSCRIPT Inman, D.L., Jenkins, S.A., 1984. The Nile littoral cell and man’s impact on the coastal zone of the southeastern Mediterranean. Scripps Inst. Oceanogr. Ref. Ser. 33, 1600–1617.

PT

doi:10.9753/icce.v19.

RI

Jackson, J.B., Kirby, M.X., Berger, W.H., Bjorndal, K.A., Botsford, L.W., Bourque, B.J.,

SC

Bradbury, R.H., Cooke, R., Erlandson, J., Estes, J.A., Hughes, T.P., Kidwell, S., Lange, C.B., Lenihan, H.S., Pandolfi, J.M., Peterson, C.H., Steneck, R.S., Tegner, M.J., Warner,

NU

R.R., 2001. Historical overfishing and the recent collapse of coastal ecosystems. Science

MA

293, 629–637. doi:10.1126/science.1059199.

Kidwell, S.M., 2001. Preservation of species abundance in marine death assemblages. Science

D

294, 1091–1094. doi:10.1126/science.1064539.

TE

Kidwell, S.M., 2002. Mesh-size effects on the ecological fidelity of death assemblages: a meta-

AC CE P

analysis of molluscan live–dead studies. Geobios 35, 107–119. doi:10.1016/S00166995(02)00052-9.

Kidwell, S.M., 2007. Discordance between living and death assemblages as evidence for anthropogenic

ecological

change.

PNAS

104,

17701–17706.

doi:10.1073/pnas.0707194104. Kidwell, S.M., 2008. Ecological fidelity of open marine molluscan death assemblages: Effects of post-mortem transportation, shelf health, and taphonomic inertia. Lethaia 41, 199–217. doi:10.1111/j.1502-3931.2007.00050.x. Kidwell, S.M., 2009. Evaluating human modification of shallow marine ecosystems: Mismatch in composition of molluscan living and time-averaged death assemblages. In: Dietl, G.P.,

ACCEPTED MANUSCRIPT Flessa, K.W. (Eds.), Conservation Paleobiology: Using the Past to Manage for the Future. The Paleontological Society Papers, Portland, pp. 113–139.

PT

Kidwell, S.M., 2013. Time-averaging and fidelity of modern death assemblages: Building a

RI

taphonomic foundation for conservation palaeobiology. Palaeontology 56, 487–522.

SC

doi:10.1111/pala.12042.

Kidwell, S.M., Bosence, D.W.J., 1991. Taphonomy and time-averaging of marine shelly fauna.

NU

In: Allison, P.A., Briggs, D.E.G. (Eds.), Taphonomy: Releasing the Data Locked in the

MA

Fossil Record. Plenum, New York, pp. 115–209.

Korpanty, C.A., Kelley, P.H., 2014. Molluscan live-dead agreement in anthropogenically

D

stressed seagrass habitats: siliciclastic versus carbonate environments. Palaeogeogr.

TE

Palaeoclimatol. Palaeoecol.‫‏‬410, 113-125.

AC CE P

Kress, N., Herut, B., 2001. Spatial and seasonal evolution of dissolved oxygen and nutrients in the Southern Levantine Basin (Eastern Mediterranean Sea): chemical characterization of the water masses and inferences on the N:P ratios. Deep Sea Res. Part I Oceanogr. Res. Pap. 48, 2347–2372. doi:10.1016/S0967-0637(01)00022-X. Kress, N., Herut, B., Galil, B.S., 2004. Sewage sludge impact on sediment quality and benthic assemblages off the Mediterranean coast of Israel-a long-term study. Mar. Environ. Res. 57, 213–33. doi:10.1016/S0141-1136(03)00081-3. Levin, L.A., Gage, J.D., 1998. Relationships between oxygen, organic matter and the diversity of bathyal macrofauna. Deep Sea Res. Part II 45, 129–163. doi:10.1016/S09670645(97)00085-4.

ACCEPTED MANUSCRIPT Lotze, H.K., Lenihan, H.S., Bourque, B.J., Bradbury, R.H., Cooke, R.G., Kay, M.C., Kidwell, S.M., Kirby, M.X., Peterson, C.H., Jackson, J.B.C., 2006. Depletion, degradation, and

PT

recovery potential of estuaries and coastal seas. Science 312, 1806–1809.

RI

doi:10.1126/science.1128035.

SC

Magni, P., 2003. Biological benthic tools as indicators of coastal marine ecosystems health. Chem. Ecol. 19, 363–372. doi:10.1080/02757540310001603718.

NU

Magurran, A.E., 2004. Measuring biological diversity. Blackwell, Oxford.

MA

MarLIN, 2006. BIOTIC- Biological Traits Information Catalogue [WWW Document]. URL http://www.marlin.ac.uk.

D

Morton, J.E., 1958. Molluscs. Hutchinson & Co. LTD, Great Britain.

TE

Nir, Y., 1984. Recent sediments of the Israel Mediterranean continental shelf and slope. PhD

AC CE P

Thesis, Inst. Mar. Geol, Goteborg University, Sweden (in English). Oliver, P.G., 1992. The Bivalve Seashells of the Red Sea: An Identification Guide. Verlag Christa Hemmen, Wiesbaden and the National Museum of Wales, Cardiff. Olszewski, T.D., Kidwell, S.M., 2007. The preservational fidelity of evenness in molluscan death assemblages. Paleobiology 33, 1–23. doi:10.1666/05059.1. Pearson, T.H., 1987. Benthic ecology of an accumulating sludge disposal site. In: Capuzzo, J.M., Kester, D.R. (Eds.), Oceanic Processes in Marine Pollution, Vol. I, Biological Processes and Wastes in the Ocean. R.E.Kriger Publishing, pp. 195–200. Pechnik, J.A., 2010. Biology of the Invertebrates, 6th ed. McGraw-Hill Higher Education. POEM group, 1992. General circulation of the Eastern Mediterranean. Earth-Science Rev. 32, 285–309. doi:http://dx.doi.org/10.1016/0012-8252(92)90002-B.

ACCEPTED MANUSCRIPT Poppe, G.T., Goto, Y., 1991. European Seashells. Verlag Christa Hemmen, Federal Republic of Germany.

PT

Purchon, R.D., 1968. The Biology of the Mollusca. Pergamon Press, Hungary.

RI

Rosentraub, Z., Brenner, S., 2007. Circulation over the southeastern continental shelf and slope

SC

of the Mediterranean Sea: Direct current measurements, winds, and numerical model simulations. J. Geophys. Res. 112, C11001. doi:10.1029/2006JC003775.

NU

Salas, F., Marcos, C., Neto, J.M., Patrício, J., Pérez-Ruzafa, A., Marques, J.C., 2006. User-

MA

friendly guide for using benthic ecological indicators in coastal and marine quality assessment. Ocean Coast. Manag. 49, 308–331. doi:10.1016/j.ocecoaman.2006.03.001.

D

Schilman, B., Almogi-Labin, A., Bar-Matthews, M., Labeyrie, L., Paterne, M., Luz, B., 2001.

TE

Long- and short-term carbon fluctuations in the Eastern Mediterranean during the late

AC CE P

Holocene. Geology 29, 1099–1102. doi:10.1130/0091-7613(2001)029. Simboura, N., Zenetos, A., 2002. Benthic indicators to use in Ecological Quality classification of Mediterranean soft bottom marine ecosystems, including a new Biotic Index. Mediterr. Mar. Sci. 3, 77–111.

Stanley, D.J., 1988. Low Sediment Accumulation & Erosion in the Nile Delta Shelf off Egypt. Mar. Geol. 84, 111–117. Todd, J. A, Jackson, J.B.C., Johnson, K.G., Fortunato, H.M., Heitz, A., Alvarez, M., Jung, P., 2002. The ecology of extinction: Molluscan feeding and faunal turnover in the Caribbean Neogene. Proc. Biol. Sci. 269, 571–7. doi:10.1098/rspb.2001.1923.

ACCEPTED MANUSCRIPT Todd, J.A., 2001. Neogene Marine Biota of Tropical America [WWW Document]. URL http://porites.geology.uiowa.edu/database/mollusc/mollusclifestyles.htm

PT

Tomašových, A., Kidwell, S.M., 2009. Preservation of spatial and environmental gradients by

RI

death assemblages. Paleobiology 35, 119–145. doi:10.1666/07081.1.

SC

Tomašových, A., Kidwell, S.M., 2011. Accounting for the effects of biological variability and temporal autocorrelation in assessing the preservation of species abundance.

NU

Paleobiology 37, 332–354. doi:10.1666/09506.1.

MA

Weber, K., Zuschin, M., 2013. Delta-associated molluscan life and death assemblages in the northern Adriatic Sea: Implications for paleoecology, regional diversity and conservation.

TE

D

Palaeogeogr. Palaeoclimatol. Palaeoecol. 370, 77–91. doi:10.1016/j.palaeo.2012.11.021. Wilson, E.O., 2003. The Encyclopedia of Life. Trends in Ecology and Evolution [WWW

AC CE P

Document]. URL http://www.sciencedirect.com/science/article/B6VJ1-47C8RDN3/2/befac60e32dd59e55ff8bfc75f9848c6. ISSN 0169-5347 (accessed December 2012). WoRMS,

2014.

World

Register

of

Marine

Species

[WWW

Document].

URL

http://www.marinespecies.org (accessed December 2012). Yacobi, Y.Z., Zohary, T., Kress, N., Hecht, A., Robarts, R.D., Waiser, M., Wood, A.M., Li, W.K.W., 1995. Chlorophyll distribution throughout the southeastern Mediterranean in relation to the physical structure of the water mass. J. Mar. Syst. 6, 179–190. doi:10.1016/0924-7963(94)00028-A. Yonge, C.M., Thompson, T.E., 1976. Living Marine Molluscs. William Collins Sons & Co Ltd, Great Britain.

ACCEPTED MANUSCRIPT Zenetos, A., Gofas, S., Russo, G., Templado, J., 2003. Molluscs. In: Brian, F. (Ed.), CIESM Atlas of Exotic Species in the Mediterranean. CIESM Publishers, Monaco, p. 376.

PT

Zuschin, M., Stachowitsch, M., 2009. Epifauna-dominated benthic shelf assemblages: Lessons

RI

from the modern Adriatic Sea. Palaios 24, 211–221. doi:10.2110/palo.2008.p08-062r.

SC

Zviely, D., Kit, E., Klein, M., 2007. Longshore sand transport estimates along the Mediterranean

AC CE P

TE

D

MA

doi:10.1016/j.margeo.2006.12.003.

NU

coast of Israel in the Holocene. Mar. Geol. 238, 61–73.

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 1: Location map of the Shafdan sewage sludge outlet and polluted and control sampling stations.

ACCEPTED MANUSCRIPT Dead assemblage Polluted

Live assemblage

Control

Polluted

Control

100 90

PT

70

n=409 n=1125 n=333 PL29 Fall

PL64 Fall

n=482 n=250 PL29 Spr

PL64 Spr

n=276 PL29 Win

PL3 Win

NU

PL29 Sum

n=112 n=428

PL64 Fall

PL3 Fall

Other

TE

D

Deposit feeder

PL3 Sum

n=563

PL64 Spr

MA

n=97

n=682

PL29 Fall

PL29 Sum

Filter feeder

SC

n=600

PL29 Spr

PL3 Spr

0

PL3 Spr

n=660 n=546

PL29 Win

10

n=163

n=794

PL3 Fall

20

n=837 n=649

30

PL3 Sum

40

n=844 n=740

50

RI

60

PL3 Win

%Individuals

80

AC CE P

Figure 2: Proportional abundance of filter and deposit feeding bivalves of the live and dead assemblages from the polluted (PL3) and control (PL29, PL64) stations for all seasons.

Figure 3: Margalef’s richness index of the live and dead assemblages from the polluted (PL3)

D

and control (PL29, PL64) stations for all seasons of 2012.

TE

AC CE P

0.0 PL64 D Fall

PL64 D Spr

PL29 D Fall

PL29 D Sum

0.5

PL29 D Spr

3.0

PT

3.5

PL29 D Win

2.0

RI

Live assemblage

PL3 D Fall

4.5

PL3 D Sum

1.0

SC

2.5

PL3 D Spr

1.5

PL3 D Win

PL64 L Fall

NU

PL64 L Spr

PL29 L Fall

MA

PL29 L Sum

PL29 L Spr

PL29 L Win

PL3 L Fall

PL3 L Sum

PL3 L Spr

PL3 L Win

Margalef's richness

ACCEPTED MANUSCRIPT

Dead assemblage

4.0

Figure 4: Shannon-Wiener index of the live and dead assemblages from the polluted (PL3) and

D

control (PL29, PL64) stations for all seasons of 2012.

TE

AC CE P

0.0 PL64 D Fall

PL64 D Spr

PL29 D Fall

PL29 D Sum

PL29 D Spr

1.5

PT

2.0

PL29 D Win

Live assemblage

PL3 D Fall

2.5

PL3 D Sum

RI

SC

1.0

PL3 D Spr

0.5

PL3 D Win

PL64 L Fall

NU

PL64 L Spr

PL29 L Fall

MA

PL29 L Sum

PL29 L Spr

PL29 L Win

PL3 L Fall

PL3 L Sum

PL3 L Spr

PL3 L Win

Shannon-Wiener

ACCEPTED MANUSCRIPT

Dead assemblage

Figure 5: PIE evenness index of the live and dead assemblages from the polluted (PL3) and

D

control (PL29, PL64) stations for all seasons of 2012.

TE

AC CE P

0.0 PL64 D Fall

PL64 D Spr

PL29 D Fall

PL29 D Sum

0.1

PL29 D Spr

0.6

PT

0.7

PL29 D Win

0.4

RI

Live assemblage

PL3 D Fall

0.9

PL3 D Sum

0.2

SC

0.5

PL3 D Spr

0.3

PL3 D Win

PL64 L Fall

NU

PL64 L Spr

PL29 L Fall

MA

PL29 L Sum

PL29 L Spr

PL29 L Win

PL3 L Fall

PL3 L Sum

PL3 L Spr

PL3 L Win

PIE evenness

ACCEPTED MANUSCRIPT

Dead assemblage

0.8

ACCEPTED MANUSCRIPT

MA

NU

SC

RI

PT

Ta xa (95 % co nfi de nc e)

D

Specimens

TE

Figure 6: Rarefied species richness (95% confidence intervals) plotted against the number of

AC CE P

pooled individuals of the live and dead assemblages. For any number of individuals selected randomly the dead has significantly more species than the live assemblage.

ACCEPTED MANUSCRIPT 1.0

PL29 control

PL3 polluted

PT

PL3 2003-4

RI

PL29 2003-4

NU

SC

0.5

MA

Taxonomic similiarty (Jaccard-Chao)

PL64 control

0.0 Rank abundance correlation (Spearman's rho)

TE

-1.0

D

0.0

PL29 control

PL3 2003-4

PL29 2003-4

AC CE P

PL3 polluted

1.0

PL64 control

Figure 7: Cross plot of taxonomic similarity (Jaccard-Chao) against correlation of rank order of abundance (Spearman’s rho) of live-dead samples from polluted (PL3) and control stations of 2012 (PL29 and PL64), and of live assemblages of 2012 compared to death assemblages sampled in 2003-4 (from polluted PL3 and control PL29. PL64 was not sampled in 2003-4). Note all samples plot in the upper right quadrant, associated with pristine settings.

AC CE P

TE

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 8: nMDS of the live and dead assemblages of 2012. Separate clusters of the live and dead samples were superimposed by full and dashed lines at increasing levels of similarity of 59% and 75%.

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

TE

Figure 9: Cluster analysis of live assemblages sampled in 2012 and death assemblages sampled

AC CE P

in 2003-4 from the same stations. Red lines mark there is no statistical significance found within groups of samples (SIMPROF test).

ACCEPTED MANUSCRIPT List of Figures Figure 1: Location map of the Shafdan sewage sludge outlet and polluted and control sampling

PT

stations.

RI

Figure 2: Proportional abundance of filter and deposit feeding bivalves of the live and dead

SC

assemblages from the polluted (PL3) and control (PL29, PL64) stations for all seasons. Figure 3: Margalef’s richness index of the live and dead assemblages from the polluted (PL3)

NU

and control (PL29, PL64) stations for all seasons of 2012.

MA

Figure 4: Shannon-Wiener index of the live and dead assemblages from the polluted (PL3) and

D

control (PL29, PL64) stations for all seasons of 2012.

TE

Figure 5: PIE evenness index of the live and dead assemblages from the polluted (PL3) and

AC CE P

control (PL29, PL64) stations for all seasons of 2012. Figure 6: Rarefied species richness (95% confidence intervals) plotted against the number of pooled individuals of the live and dead assemblages. For any number of individuals selected randomly the dead has significantly more species than the live assemblage. Figure 7: Cross plot of taxonomic similarity (Jaccard-Chao) against rank order of abundance (Spearman’s rho) of live-dead samples from polluted (PL3) and control stations of 2012 (PL29 and PL64), and of live assemblages of 2012 compared to death assemblages sampled in 2003-4 (from polluted PL3 and control PL29. PL64 was not sampled that year). Note all samples plot in the upper right quadrant, associated with pristine settings.

ACCEPTED MANUSCRIPT Figure 8: nMDS of the live and dead assemblages of 2012. Separate clusters of the live and dead samples were superimposed by full and dashed lines at increasing levels of similarity of 59% and

PT

75%.

RI

Figure 9: Cluster analysis of live assemblages sampled in 2012 and death assemblages sampled

SC

in 2003-4 from the same stations. Red lines mark there is no statistical significance found within

AC CE P

TE

D

MA

NU

groups of samples (SIMPROF test).

ACCEPTED MANUSCRIPT Table 1: Taxonomic list of the live (L) and dead (D) bivalves and gastropods in the control and

polluted sampling stations of the Israeli Mediterranean. Alien, migrant species to the

cf. Spisula? Chama gryphoides Clementia papyracea* Conomurex persicus* Conus mediterraneus Conus sp. Corbula gibba Cucurbitula cymbium* Cuspidaria cuspidata Cylichna cylindracea Diplodonta sp.* Donax venustus Ensis minor Epitonium clathrus Epitonium striatissimum

SC

RI

2012 total no. individuals L/D 537/632 0/80

MA

D

D D D D L and D D D D D D D D D D D D D D D L and D D D D L and D D D D D

2003-4 total no. individuals D 338 111

0/112

71

0/1 0/2 0/1 0/0 1/298 0/2 0/5 0/58 0/5 0/3 0/0 0/4 0/1

1 1 0 1 167 5 2 39 3 36 12 0 0

0/0 0/1 0/2 0/1 0/2 0/2 1643/4022 0/1 0/8 0/1 1/12 0/0 0/2 0/1 0/1

1 0 6 0 0 3 2453 2 3 1 14 2 0 0 0

NU

D

AC CE P

Abra longicallus Abra prismatica Acanthocardia paucicostata Aclis minor Acteon tornatilis Aequipecten opercularis Alvania sp. Anadara polii Anomia ephippium Aporrhais pespelecani Azorinus chamasolen Bela sp. Bittium latreillii Bittium reticulatum Bolinus brandaris Calliostoma zizyphinum

Found Live/Dead L and D D

TE

Species

PT

Mediterranean, are marked with *.

TE

RI

SC

0/9 0/0 2/1 0/3 0/2 0/1 0/2 0/1 0/1 0/1 0/1 0/1 0/0 0/1 0/1 0/1 0/1 0/0 0/10 0/0 0/1 11/0 0/4 32/1 2/5 0/2 0/0 1/1 0/1 0/8 2/0 0/4 0/1 0/1 0/436

NU MA

D

D D L and D D D D D D D D D D D D D D D D D D D L D L and D L and D D D L D D L D D D D

AC CE P

Eulima glabra Eulimella acicula? Fulvia fragilis* Fusinus sp.* Glycymeris nummaria Gouldia minima Gregariella petagnae Haitia acuta Haminoea orbignyana Haminoea sp. Hexaplex trunculus? Jujubinus exasperatus Kellia suborbicularis Limatula sp. Lithophaga lithophaga Lucinidae sp. Lyonsia norwegica Macoma cumana Mactra stultorum Mactra? Mangelia attenuata Megastomia conoidea* Mimachlamys varia Modiolus agglutinans Modiolus auriculatus* Modiolus sp. Morella donacina Murex forskoehlii* Muricidae sp. Musculus costulatus Musculus subpictus Mysella bidentata Nassarius gibbosulus Nassarius mutabilis Nassarius pygmaeus Naticarius stercusmuscarum Naticidae sp. Neverita josephinia Nucula nitidosa

PT

ACCEPTED MANUSCRIPT 15 1 0 0 8 0 0 0 0 0 0 0 5 0 0 0 0 6 0 3 1 8 4 1 0 5 7 1 0 7 0 34 0 297 1

D

0/17

8

D D L and D

0/4 0/41 297/527

0 13 237

ACCEPTED MANUSCRIPT 1109/495 0/0 0/8 0/12 0/1 0/1 0/1 0/11 1/7

L

1/0

1

0/2 0/1 0/1 0/6 0/1 97/501 0/0 0/1 0/1 0/2 0/2 0/40 0/0 1/0 0/1 11/0 0/2 1/0 15/63 0/4 0/1 0/1 0/0 0/1 0/2 0/2 0/1 0/1 0/5

0 0 5

TE

RI

SC

NU MA

D

D D D D D L and D D D D D D D D L D L D L L and D D D D D D D D D D D

PT

L and D D D D D D D D L and D

AC CE P

Nuculana pella Ondina vitrea Ostrea edulis Ostreola stentina Palliolum incomparabile Palliolum tigerinum Pandora pinna Pandora sp. Paphia textile* Papillicardium papillosum Pecten jacobaeus Phalium granulatum Phaxas pellucidus Philine aperta Philine catena Pitar rudis Pollia dorbignyi Psammobiidae sp. Psammotreta praerupta* Pseudominolia nedyma* Rhinoclavis kochi* Ringicula conformis Rissoa sp. Sphenia binghami* Spondylus spinosus* Syrnola fasciata* Tectonatica filosa Tellimya ferruginosa Tellinella pulchella Tellinidae sp. Thyasira sp. Timoclea roemeriana* Tonna sp.? Turbonilla sp.* "Turridae" sp. Typhinellus labiatus Veneridae sp. Ventomnestia girardi Venus verrucosa

284 6 4 0 0 0 0 8 1

2 333 1 0 0 0 0 301 1 0 0 0 0 0 16 0 0 0 1 0 3 0 2 0 0

ACCEPTED MANUSCRIPT D

0/0

1

AC CE P

TE

D

MA

NU

SC

RI

PT

Weinkauffia turgidula

ACCEPTED MANUSCRIPT

Table 2: Number of bivalve and gastropod species and individuals of the live and dead

Dead assemblage No. of No. of species individuals 45 6915 32 601 77 7516

SC

NU

AC CE P

TE

D

MA

Bivalvia Gastropoda Total

Live assemblage No. of No. of species individuals 16 3741 3 23 19 3764

RI

PT

assemblages of the polluted and clean station of the Israeli Mediterranean during 2012 sampling.

ACCEPTED MANUSCRIPT Table 3: Total organic carbon (%TOC) in surface sediments and bottom-water oxygen concentration of the polluted (PL3) and control (PL29, PL64) sampling stations during the four

PT

seasons of 2012.

PL29

PL64

PL3

2.00 1.14 2.93 3.53

0.33 0.37 0.46 0.41

n/a 0.29 n/a 0.35

225.25 195.55 178.61 186.12

D TE AC CE P

PL29

PL64

223.74 239.51 174.90 190.02

n/a 239.51 n/a 194.52

NU

SC

PL3

MA

2012 sampling Winter Spring Summer Fall

RI

Bottom water oxygen concentration (mmol/L)

%TOC [dry wt.]

ACCEPTED MANUSCRIPT Table 4: Live-dead comparisons of taxonomic similarity (Jaccard-Chao) and rank order agreement (Spearman’s rho) of the season with the largest number of live individuals and

PT

counterpart dead assemblage for 2012 and 2003-4 sampling, and for pooled 2012 live-dead

Live-dead comparisons largest live vs. dead 2012 pooled PL3 largest live vs. dead 2003-4 PL29 largest live vs. dead 2012 pooled PL29 largest live vs. dead 2003-4 PL64 largest live PL64 pooled PL64 Pooled stations All Live-Dead assemblages 2012

Jaccard-Chao 0.88 0.89 0.85 0.80 0.85 0.76 0.87 0.88 0.87

AC CE P

TE

D

MA

NU

SC

Station PL3

RI

assemblages per station. All rho values are statistically significant (p < 0.05). Spearman's rho 0.75 0.60 0.74 0.17 0.15 0.30 0.26 0.23 0.12

ACCEPTED MANUSCRIPT Table 5: Results of the analysis of similarity (ANOSIM) for live and dead assemblages from the three sampling stations for the four seasons of 2012. R(stat)

p-value

0.984

0.01

0.034

0.43

-0.083

0.50

0.001

0.67

0.167

0.30

0.083

0.40

0.259

0.30

0.167

0.60

Stations within samples of the death assemblage (control vs. polluted)

0.286

0.33

Season within samples of the live assemblage 2012 winter. Global R: Pairwise test (groups- winter, spring, summer, fall)

0.459

0.02

winter, spring

0.667

0.10

winter, summer

0.500

0.33

winter, fall

-0.583

1.00

spring, summer

0.167

0.40

spring, fall

0.889

0.10

summer, fall

0.917

0.10

Stations within samples of the live assemblage (control vs. polluted)

0.139

0.17

Sampling sites (control vs. polluted)

-0.196

0.85

PT

Groups compared (factors)

SC

RI

Live vs. Dead assemblages across all stations and seasons Seasons within samples of the death assemblage 2012 winter. Global R: Pairwise test (groups- winter, spring, summer, fall) winter, spring

NU

winter, summer winter, fall

MA

spring, summer spring, fall

AC CE P

TE

D

summer, fall

ACCEPTED MANUSCRIPT List of Tables Table 1: Taxonomic list of the live (L) and dead (D) bivalves and gastropods in the control and

PT

polluted sampling stations of the Israeli Mediterranean. Alien, migrant species to the

RI

Mediterranean, are marked with *.

Table 2: Number of bivalve and gastropod species and individuals of the live and dead

SC

assemblages of the polluted and clean station of the Israeli Mediterranean during 2012 sampling.

NU

Table 3: Total organic carbon (%TOC) in surface sediments and bottom-water oxygen concentration of the polluted (PL3) and control (PL29, PL64) sampling stations during the four

MA

seasons of 2012.

Table 4: Live-dead comparisons of taxonomic similarity (Jaccard-Chao) and rank order

TE

D

agreement (Spearman’s rho) of the season with the largest number of live individuals and counterpart dead assemblage for 2012 and 2003-4 sampling, and for pooled 2012 live-dead

AC CE P

assemblages per station. All rho values are statistically significant (p < 0.05). Table 5: Results of the analysis of similarity (ANOSIM) for live and dead assemblages from the three sampling stations for the four seasons of 2012.

ACCEPTED MANUSCRIPT Highlights Live-dead fidelity of mollusca was tested from a polluted and two control sites.



Statistical analyses show significant difference between live and dead assemblages



Species composition and rank order between live-dead assemblages remained constant.



Annual sludge removal and pollution tolerant species preserved live-dead fidelity



Eutrophication variations can be tracked by molluscan trophic levels.

AC CE P

TE

D

MA

NU

SC

RI

PT

