The effect of the abundance of venus clams (Gafrarium tumidum Röding, 1798) on sediment accretion in the sandy intertidal of Pari Island

Bivalves may play an ecological role as ecosystem engineers by changing sediment physical structure. As a filter/suspension feeder, they can produce biodeposits, and their burrowing activities can stir sediment. In this study, we investigated whether the abundance of Venus clam (Gafrarium tumidum Röding, 1798) influences sediment accretion in the sandy intertidal area of Pari Island, Indonesia. We created 18 plots consisting of nine defaunated and nine non-defaunated treatment plots. On each plot, 111-559 clams were deployed, and four pieces of one-inch PVC were planted randomly to see sediment elevation. Sediment accretion was measured by sediment elevation. Observations were carried out for seven months. At the end of the observations, clam abundances were 0-64 ind/m2 in the defaunated plots and 0-47 ind/m2 in the non-defaunated plots. The average sediment accretion in the defaunated plots was 0.98-3.3 cm, while in the non-defaunated plots was 1.95-4.43 cm. This study showed that the clam had no significant effect (p-value > 0.05) on sediment accretion in both plots as its abundance was insufficient to alter sediment. The sediment accretion occurred not caused by clam but by others such as sipuncula and physical oceanography factors in the research location.


Introduction
Bivalves may act as ecosystem engineers in the marine environment [1][2][3][4].Ecosystem engineers are organisms that create, change or maintain environmental conditions affecting other organisms' resources [4].They change the physical condition of the environment and then cause a domino effect on other organisms and also themselves.Benthic biota such as bivalves alter physical condition which is the sediment profile through allogenic and autogenic engineering.
The widely known autogenic engineer is reef-building bivalves such as mussels (Mytilidae) and oysters (Ostreidae).Their physical structure can primarily alter the sedimentary habitat [5].In addition, they create fecal and pseudofaecal biodeposits that build up in the reef and its surroundings as filter feeders [6][7][8].Non-reef-building bivalves, such as cockles and clams, usually carry out allogenic engineering.They alter the sedimentary habitat through their activities.Bivalves' behaviour, such as digging, plowing sediments, and shell movement, cause the instability of the sediment surface [9].At low densities, bivalves act as bioturbators, but at high densities, they can enhance sediment stability [10].Sediment stability occurs because the higher the density of bivalves produces more biodeposits (mucus and pseudofeces) that bind sediment [11].It is important to know the ability of bivalves to change their environment because on large scale it will impact other organisms significantly.
In Pari Island, many species of bivalves can be found inhabiting sandy intertidal areas [12][13][14], and the most abundant is venus clams Gafrarium tumidum (Roding, 1798).It was estimated that the stock of venus clams in 2017 in Pari Island was 18,902 individuals [14].G. tumidum is endobenthic organisms, submerging vertically, although occasionally, with its shell posteriors appear on the substrates' surface [15].G. tumidum also does not create reefs.
G.tumidum may have the highest abundance on the island of Pari, but until now, we have yet to learn the effect of G. tumidum on the environment, especially sediment.Research on G. tumidum in Pari Island is limited to its stock [14] and distribution [12,13].In Indonesia, it is still limited to bioecology [15,16], the relationship between habitat quality and its distribution patterns [12,17], and its mineral [18] and microplastics contents [19,20].So that this study aims to determine the ability of the G. tumidum to alter the sediment physical properties in Pari Island.We assumed that the abundance of G. tumidum would positively affect sediment accretion.

Method
This study was carried out in Pari Island, Seribu Islands of Indonesia (Figure 1).This island is a coral island (coral clays) and is approximately 35 km northwest of Jakarta, the capital city of Indonesia.Furthermore, this island is a small tourism area (about 4 square kilometers) with domestic activity.We conducted the study in the southern part of Pari Island, in an intertidal seagrass ecosystem with sandy sediment.We conducted the research from April until November 2021.

Figure 1. Map of the research location
Eighteen plots measuring 3x3m 2 were created and divided into nine defaunated plots (Figure 2) and nine non-defaunated plots (Figure 3).Defaunation treatment aimed to eliminate other biotas' influence on the observed variables.To eliminate the biota from the sediment, defaunation was carried out by covering the surface of the sediment with a plastic tarpaulin for 28 days.After that, the tarpaulin was removed and left for two weeks to remove the toxic gasses produced during the defaunation process.Success defaunation treatments as indicated by the dark color on the surface of the sediment and the smell of sulfide gas from the decomposition process.After that the plots were ready to use and the G. tumidum were added.G. tumidum were added in May 2021 in each plot with different abundances.Six plots were not added G. tumidum, six plots were added 111 clams (13 ind/m 2 ) for each plot, and the other six plots were given 559 clams (63 ind/m 2 ) for each plot.In November, the abundance of clams was observed using a 50x50cm frame three times in each plot.
We measured sediment accretion represented by cumulative sediment surface elevation in each plot [21].Four PVCs were plugged into each plot randomly in July (Figure 4) and then marked the height of the sediment (Figure 5).In November, the PVC poles were removed.Sediment heights at the beginning and the end were compared to see the accumulated sediment.The influence of abundance on sediment accretion in defaunated and non-defaunated plots was analyzed using simple regression analysis and the f test.The influence of defaunation on sediment accretion was analyzed by the dependent t-test.

Results and discussion
The condition in defaunated and non-defaunated plots were different.In the non-defaunated plots, seagrass species Thalassia hemprichii, Enhalus acoroides, and benthic biotas were found in these plots.In the defaunated plot, abundance and richness of benthic biotas were lesser than in non-defaunated plot.Seagrass almost was not found unless it protruded from the edge of plots Sediment surface elevation indicated the ability of the clams to induce sediment accretion.In five months, accretion in defaunated plots ranged 0.98-3.3cm with an average of 2.96 ± 0.84 cm, while for non-defaunated plots, it ranged 1.95-4.43cm with an average of 2.22±0.82cm.Sediment accretion in defaunated and non-defaunated plots did not differ significantly (p-value >0.05) (Figure 6).The abundance of clams ranged from 0 to 64 ind/m 2 in the defaunated plot and 0-47 ind/m 2 in nondefaunated plots (Figure 7).Both were not significantly different (p>0.05).In both defaunated and nondefaunated plots, the correlation and influence of clam abundance on sediment accretion were weak and insignificant (p>0.05)(Figure 8 dan 9).Meanwhile, the natural abundance of G.tumidum found around the plots was 28 ind/m 2 .It was higher than in 2008, which was 1-25 ind/8m 2 [12].Bivalve groups that apparently have the ability as ecosystem engineers are epibenthic reef-building bivalves such as estuarine oysters Crassostrea virginica, blue mussels Mytilus edulis, freshwater zebra mussles Dreissena polymorpha and quagga mussels Dreissena bugensis [9,10].Their reef gives a habitat heterogenity, and also can alter ecosystem.On the other hand, non-reef-building bivalves change environmental conditions through their activities and only become significant at high abundances.

5
G. tumidum clam is an endobenthic bivalve that buries its bodies on the sediment surface [15], filterfeeding bivalves [22] and does not form reefs.Its effect on the environment possibly will appear when it has a high abundance such as other endobenthic bivalves Cerastoderma edule.C.edule which is also a non-reef-building bivalve, can affect sediment by burrowing, filter-feeding, and shell valve adductions [23,24].The higher the abundance of C. edule, the more significant the bioturbator effect, and leads to increase erosion in the sediment.However, when the abundance reached 312 ind/m 2 , the bioturbator effect changed to sediment stabilizer because it produced higher mucus secretion and pseudofaeces production [23,24].Another study showed that an abundance of 1000 ind/m 2 C.edule increased sediment stability in sandy sediment, and increased sediment elevation [10].Asiatic hard clam (Meretrix meretrix), with a density of 122-167 ind/m 2 had a higher accretion rate than 0-11 ind/m 2 density.Sediment elevation at higher density was +30mm and at lower density was only +15mm [21].
G. tumidum as a filter/suspension feeder produces biodeposits, although most biodeposits are likely resuspended by tidal or wave-induced currents [25].Biodeposits can elevate the sediment bed on the condition the density of the clam is sufficient.The study of the correlation between the abundance of the manila clam (Ruditapes philippinarum) with sediment accretion showed that a very high density in farming areas had a higher biodeposition rate impact than in natural areas.At that very high-density area, the sedimentation that occurred by biodeposition was higher than by physical oceanography [25].In this study, the abundance of G. tumidum had an insignificant correlation with sediment accretion due to the possibility that its abundance was insufficient as both a bioturbator and a sediment stabilizer.Sediment accretion was not related to its abundance but to other factors.The factors causing sediment accretion first was the presence of another bioturbator.During field observations, sand mounds appeared on defaunated and non-defaunated plots (Figure 10).Several mounds in the middle or beside the PVC poles caused a higher sediment elevation than in the unmounded PVC poles (Figure 11).The sand mound has a burrow at the top.To determine the inhabitant of these mounds, we took the biota using the hypoxia method.Apparently, it was inhabited by sipuncula.It was unusual for sipuncula to make a mound, usually shrimps like caridean shrimp [26] or callianassid shrimp [27] that made mounds in sandy seagrass areas.Sipuncula is a deposit feeder that makes burrows and redistributes matter from the sediment surface to within burrows [28,29].So, it needs to be further investigated how the sipuncula makes these mounds.Another factor was physical oceanography.Based on research on the characteristics of the islands in the Pari island group [30], the location of research plot was in the southern part of Pari Island and the inner part of the Pari island group.This area has very little wave influence, so the dominant process is sedimentation [30].

Conclusion
The abundance of G. tumidum in the range of 0-64 ind/m 2 did not affect sediment accretion in Pari Island.Sediment accretion was caused by the presence of another bioturbator that made sand mounds in the plot areas and physical oceanographic that caused sedimentation.

Figure 4 . 5 .
Figure 4. PVC poles in each plot Figure 5. Sediment elevation mark in PVC poles

Figure 8 .
Figure 8. Correlation between sediment accretion and clam abundance in defaunated plots.

Figure 9 .
Figure 9. Correlation between sediment accretion and clam abundance in non-defaunated plots.