Titel: Titel

18 downloads 0 Views 6MB Size Report
Chelidonichthys lucernus rode poon tub gurnard. Cyclopterus lumpus snotolf lumpsucker. Eutrigla gurnardus grauwe poon grey gurnard. Liparis liparis slakdolf.
INSTITUTE FOR AGRICULTURE AND FISHERIES RESEARCH ANKERSTRAAT 1 B-8400 OOSTENDE, BELGIUM TEL. 059 34 22 50 - FAX 059 33 06 29 [email protected] WWW .ILVO.VLAANDEREN.BE

ANIMAL SCIENCES UNIT – RESEARCH AREA FISHERIES

REPORT ILVO-ANIMAL SCIENCES-FISHERIES N° 2

Title: Title

Biological Environmental Research

Monitoring the effects of sand extraction on the benthos of the Belgian part of the North Sea

Annelies De Backer, Ine Moulaert, Hans Hillewaert, Sofie Vandendriessche, Gert Van Hoey, Jan Wittoeck and Kris Hostens

February 2010

INSTITUTE FOR AGRICULTURE AND FISHERIES RESEARCH ANKERSTRAAT 1 B-8400 OOSTENDE, BELGIUM TEL. 059 34 22 50 - FAX 059 33 06 29 [email protected] WWW .ILVO.VLAANDEREN.BE

ANIMAL SCIENCES UNIT – RESEARCH AREA FISHERIES

REPORT ILVO-ANIMAL SCIENCES-FISHERIES N° 2

Monitoring the effects of sand extraction on the benthos of the Belgian part of the North Sea

Annelies De Backer, Ine Moulaert, Hans Hillewaert, Sofie Vandendriessche, Gert

Van Hoey, Jan Wittoeck and Kris Hostens

Overzichtsrapport 2008 van het ILVO overeenkomstig artikel 2 & 4 van het samenwerkingsakkoord tussen de Federale Overheid en het Vlaamse Gewest van 21 december 2005 betreffende het onderzoek naar de invloed van de exploratie- en exploitatieactiviteiten op het BCP op de sedimentafzettingen en op het mariene milieu.

February 2010

ACKNOWLEDGMENTS We wish to thank everybody who has contributed to this project and especially the following persons: The crew of the research vessel Belgica and Zeeleeuw for their assistance during the sampling campaigns. Bart Goes for processing of the samples CONTACT Annelies De Backer Institute for Agriculture and Fisheries Research Animal Sciences Unit–Fisheries Research Ankerstraat 1 B-8400 Oostende, Belgium Tel. 059 56 98 77 – Fax 059 33 06 29 [email protected]

REFERENCE De Backer A, Moulaert I, Hillewaert H, Vandendriessche S, Van Hoey G, Wittoeck J and Hostens K (2010) Monitoring the effects of sand extraction on the benthos of the Belgian Part of the North Sea. ILVO-report, pp. 117

i

CONTENTS ACKNOWLEDGMENTS ........................................................................................................................ I CONTENTS ............................................................................................................................................ II LIST OF FIGURES ............................................................................................................................... IV LIST OF TABLES ................................................................................................................................ IX EXECUTIVE SUMMARY ..................................................................................................................... X 1.

2.

3.

INTRODUCTION ............................................................................................................................1 1.1.

Sand extraction on the Belgian Part of the North Sea ............................................................ 1

1.2.

Monitoring the benthos .......................................................................................................... 4

1.3.

General description of sampling strategy and methods used ................................................. 5

1.4.

The aim of this report ............................................................................................................. 6

EVALUATION OF THE CONSEQUENCES OF LONG TERM SAND EXTRACTION ON THE STRUCTURAL CHARACTERISTICS OF THE MACROBENTHOS COMMUNITIES ..............................................................................................................................7 2.1.

Introduction ............................................................................................................................ 7

2.2.

Compilation of historical data ................................................................................................ 8

2.3.

Results .................................................................................................................................. 11

2.4.

Discussion ............................................................................................................................ 20

2.5.

Conclusion ........................................................................................................................... 22

THE NEED FOR BASELINE STUDIES ......................................................................................23 3.1.

4.

5.

Data analysis ........................................................................................................................ 23

BASELINE STUDIES MACROBENTHOS .................................................................................24 4.1.

General overview: situating the extraction zones within the BPNS .................................... 24

4.2.

Zone I (Thornton and Gootebank) ....................................................................................... 30

4.3.

Zone II (Kwintebank, Buitenratel and Oostdyck) ................................................................ 36

4.4.

Zone III ................................................................................................................................ 42

4.5.

Zone IV (hinderbanken) ....................................................................................................... 42

BASELINE STUDIES EPIBENTHOS ..........................................................................................50 5.1.

General overview: situating the extraction zones within the BPNS .................................... 50

5.2.

Zone I (Thornton and Gootebank) ....................................................................................... 54

5.3.

Zone II (Kwintebank, Buitenratel and Oostdyck) ................................................................ 59 ii

6.

7.

8.

9.

5.4.

Zone III ................................................................................................................................ 66

5.5.

Zone IV (Hinderbanken) ...................................................................................................... 66

BASELINE STUDIES DEMERSAL FISH ...................................................................................70 6.1.

General overview: situating the extraction zones within the BPNS .................................... 70

6.2.

Zone I (Thornton and Gootebank) ....................................................................................... 74

6.3.

Zone II (Kwintebank, Buitenratel and Oostdyck) ................................................................ 77

6.4.

Zone III ................................................................................................................................ 81

6.5.

Zone IV (hinderbanken) ....................................................................................................... 82

POST-EXTRACTION EVOLUTION OF A MACROBENTHIC COMMUNITY OF A HIGHLY EXTRACTED SANDBANK IN THE SOUTHERN BIGHT OF THE NORTH SEA.................................................................................................................................................85 7.1.

Introduction .......................................................................................................................... 85

7.2.

Methods ............................................................................................................................... 86

7.3.

Results .................................................................................................................................. 87

7.4.

Discussion ............................................................................................................................ 96

7.5.

Conclusions and recommendations ...................................................................................... 98

7.6.

Acknowledgements .............................................................................................................. 99

GENERAL CONCLUSIONS AND RECOMMENDATIONS ...................................................100 8.1.

Conclusion ......................................................................................................................... 100

8.2.

Recommendations. ............................................................................................................. 101

REFERENCES .............................................................................................................................103

ANNEX I: SPECIES LIST OF MACROBENTHOS ..........................................................................107 ANNEX II: SPECIES LIST OF EPIBENTHOS ..................................................................................112 ANNEX III: SPECIES LIST OF DEMERSAL FISHES .....................................................................115 ANNEX IV: IN PREP RESULTS PRESENTED AT THE 10TH VLIZ YOUNG SCIENTISTS DAY 27 NOVEMBER 2009 ........................................................................................................117

iii

LIST OF FIGURES Figure 1-1: Overview of the concession zones for the exploitation and exploration of sand extraction on the Belgian Continental Shelf (source: MUMM). ................................................................................. 2 Figure 1-2: Blackbox data of the extraction activities on the BPNS from 2002 to 2007 (source: FOD Economie – Fonds voor Zandwinning). ..................................................................................................... 3 Figure 1-3: Overview of the sampled locations for macrobenthos (left) and epibenthos and demersal fish (right)................................................................................................................................................... 6 Figure 2-1: Map of the Belgian Continental Shelf with location of the old and new sand extraction areas............................................................................................................................................................ 8 Figure 2-2: Detailed map of the sampling locations on the Kwintebank. White dots: ILVO locations; red dots: locations of the historical records from UGent/Marine Biology. (Background Multibeam images: Fund for Sand Extraction) ............................................................................................................. 9 Figure 2-3: Black box records indicating extraction intensity summarised over the period 1996-2005 (Fund for Sand Extraction). ........................................................................................................................ 9 Figure 2-4: Median grain size and percentage of silt recorded for station ZG01 for the period 19802007. ......................................................................................................................................................... 11 Figure 2-5: Average number of individuals per m² (+ standard deviation), total number of species and average diversity per season for station ZG01 for the period 1984-2007. ............................................... 12 Figure 2-6: Relative abundance of the major taxonomic groups for station ZG01. ................................. 13 Figure 2-7: MDS of the spring (left) and autumn (right) samples of station ZG01 area for the period 1979-2007................................................................................................................................................. 13 Figure 2-8: Median grain size and percentage of silt recorded for station ZG11 for the period 20042007. ......................................................................................................................................................... 13 Figure 2-9: Historical and recent records of density, species number and diversity for the very intensively extracted northern Kwintebank area. ..................................................................................... 14 Figure 2-10: Median grain size (bars) and percentage of silt (lines) for the low intensively extracted central Kwintebank area. .......................................................................................................................... 15 Figure 2-11: Average number of individuals per m² (+ standard deviation, bars), total number of species (lines) and diversity per season for station ZG04 (Central Kwintebank area). ............................ 16 Figure 2-12: Relative abundances of the major taxonomic groups for the intensively extracted central Kwintebank area. ...................................................................................................................................... 17 Figure 2-13: MDS plot with indication of the 2 different sampling periods. ........................................... 17 Figure 2-14: Median grain size and percentage of silt recorded for the low intensively extracted northern Oostdyck area for the period 1979-2007. .................................................................................. 18

iv

Figure 2-15: Average number of individuals (+ standard deviation, bars), total number of species (lines) and average Shannon-Wiener diversity index per season for the low intensively extracted northern Oostdyck area. ........................................................................................................................... 19 Figure 2-16: Relative abundance of the diffent taxa for station ZG02. .................................................... 19 Figure 4-1: Location of all macrobenthos sampling stations visited during spring and autumn between 2004 and 2008. ......................................................................................................................................... 24 Figure 4-2: Relative abundance of the different taxa collected in all macrobenthic samples. ................. 25 Figure 4-3: MDS plot with different symbols for the different cluster groups and with indication of the different communities. The arrow indicates a decrease in mud content and an increase in median grain size............................................................................................................................................................ 26 Figure 4-4: MDS plot with location of the extraction zones as source of variation and indication of the different communities............................................................................................................................... 27 Figure 4-5: Distribution of density and diversity over the different extraction zones. ............................. 28 Figure 4-6: Relative abundance of the different taxa for the different extraction zones. ......................... 29 Figure 4-7: Location of the sampling stations in zone 1 and reference stations....................................... 30 Figure 4-8: Median grain size (left panel) and depth (right) for each sampling stations. ........................ 31 Figure 4-9: Relation between depth and median grain size and sorting coefficient for Gootebank and Thornton samples. .................................................................................................................................... 31 Figure 4-10: Species range distribution.................................................................................................... 32 Figure 4-11: MDS plots from all sampling stations in the extraction zones with indication of possible sources of variation. ................................................................................................................................. 32 Figure 4-12: MDS plot with indication of the different sampling locations ............................................ 33 Figure 4-13: Average densities for the different locations and positions. ................................................ 34 Figure 4-14: Species richness and diversity N1 for the different locations and positions. ...................... 34 Figure 4-15: Distribution pattern of density (left), species richness (middle) and diversity (right) per sample on the Thornton- and Gootebank. ................................................................................................ 34 Figure 4-16: Relative abundance of the different taxa for the different positions and areas. ................... 35 Figure 4-17: Location of the sampling stations in Zone II and reference stations. .................................. 37 Figure 4-18: Median grain size (µm) for the different sampling stations (data from spring 2007). ........ 38 Figure 4-19: Correlation between depth and median grain size. .............................................................. 38 Figure 4-20: Species range distribution of extraction zone II and surroundings. ..................................... 38 Figure 4-21: MDS plots for all samples and reference samples of Zone II with Impact, Location and Position as sources of variation. ............................................................................................................... 39

v

Figure 4-22: Density, species richness and diversity per station. Values are averaged over years, seasons and replicates............................................................................................................................... 40 Figure 4-23: Distribution of density (left), species richness (S) and diversity N1 (right) for the different areas and positions within the areas. ........................................................................................................ 41 Figure 4-24: Relative abundance of the higher taxa in the different areas and on the different positions.41 Figure 4-25: Sampling locations indicating the year of sampling. ........................................................... 43 Figure 4-26: Median grain size at the different sampling stations in exploration zone IV(left). Thickness of the very coarse sand layer (Source: RCMG-UGent) (Right). ............................................. 44 Figure 4-27: Relation between depth and median grain size for exploration zone 4. .............................. 44 Figure 4-28: Species range distribution for zone 4 .................................................................................. 45 Figure 4-29: MDS plot for all samples of zone IV with position in the gully-bank system as source of variation.................................................................................................................................................... 45 Figure 4-30: Distribution of density (left), species richness and diversity N1 (right) over the different positions of the bank-gully system of the Hinderbanken. ........................................................................ 46 Figure 4-31: Relation between density (left), species richness (middle), diversity N1 (right) and depth.47 Figure 4-32: Density (left), species richness (middle) and diversity (right) per station plotted on the sediment map............................................................................................................................................ 47 Figure 4-33: Relative abundance of the higher taxa for the different positions. ...................................... 48 Figure 4-34: Indication of suitable areas for sand extraction based on the biological data and the seismic survey. BACI-design in these suitable areas as performed in autumn 2009. .............................. 49 Figure 5-1: Location of all fish tracks on the BPNS used for the general analysis. ................................. 50 Figure 5-2: Relative importance ( %) of the different taxa within the epibenthos in all samples ............ 51 Figure 5-3: MDS plots of all epifaunal samples with different sources of variation ............................... 52 Figure 5-4: Distribution of density, biomass (left), species number and diversity N1 (right) for the different locations in spring and autumn. ................................................................................................. 53 Figure 5-5: Taxonomic distribution of density and biomass over the different zones in spring and autumn. ..................................................................................................................................................... 54 Figure 5-6: Location of fish tracks sampled for baseline study of Zone I. ............................................... 55 Figure 5-7: MDS plots for all samples in zone I and the reference stations with location and year as sources of variation. ................................................................................................................................. 56 Figure 5-8: Distribution of density, biomass (right) and diversity (S and N1) over the different zones for both spring and autumn. ..................................................................................................................... 58 Figure 5-9: Relative abundance of the higher taxa in the different zones. ............................................... 58 Figure 5-10: Location of the different fish tracks sampled for the baseline study of Zone II. ................. 60 vi

Figure 5-11: MDS plot with indication of the different sources of variation in spring. ........................... 61 Figure 5-12: Distribution of density and biomass over the different stations in spring and autumn. ...... 62 Figure 5-13: Distribution of species richness (S) and diversity N1 over the different stations in spring and autumn. .............................................................................................................................................. 63 Figure 5-14: Relative abundance of density (left) and biomass (right) in the different stations in autumn and spring. ................................................................................................................................... 64 Figure 5-15: Distribution of density and biomass (left) and species richness (N0) and diversity (N1) (right) over the different stations in spring and autumn. .......................................................................... 65 Figure 5-16: Relative abundance of the higher taxa on the different stations in spring and autumn. ...... 65 Figure 5-17: Location of the different fish tracks in exploration zone 4 and the adjoining reference stations...................................................................................................................................................... 66 Figure 5-18: MDS plots with indication of the different sources of variation. ........................................ 67 Figure 5-19: Distribution of density, biomass (left) and species richness, diversity N1 (right) over the samples and the years for spring and autumn. .......................................................................................... 68 Figure 5-20: Relative abundance of the different taxa in the different areas over the years for spring and autumn. .............................................................................................................................................. 68 Figure 5-21: Epibenthos density (ind./1000m²) with relative distribution of the different taxa from spring samples between 2005 and 2008. .................................................................................................. 69 Figure 6-1: Relative importance ( %) of the different orders of demersal fish in all samples ................. 70 Figure 6-2: MDS plots with indication of the different sources of variation ........................................... 72 Figure 6-3: Distribution of density (left), species richness and diversity N1 (right) over the different zones in spring and autumn. ..................................................................................................................... 73 Figure 6-4: Distribution of the higher taxa over the different zones in spring and autumn. .................... 73 Figure 6-5: MDS plots with indication of the different significant sources of variation in spring and autumn. ..................................................................................................................................................... 75 Figure 6-6: Distribution of density (left), species richness S and diversity N1 (right) over the different areas and years in spring and autumn. ...................................................................................................... 76 Figure 6-7: Taxonomic composition of the different areas over the years in spring and autumn. ........... 76 Figure 6-8: MDS plots with indication of the different significant sources of variation in spring and autumn. ..................................................................................................................................................... 78 Figure 6-9: Distribution of density (upper graph), species richness S and diversity N1 (lower graph) over the different years in the different areas in spring and autumn. ....................................................... 79 Figure 6-10: Taxonomic composition of the different zones over the different years in spring and autumn. ..................................................................................................................................................... 80 Figure 6-11: MDS plots with indication of the different sources of variation. ........................................ 82 vii

Figure 6-12: Distribution of density (left), species richness and diversity N1 (right) over the different areas and years. ........................................................................................................................................ 83 Figure 6-13: Higher taxonomic composition of the different areas over the different years. .................. 83 Figure 7-1: Multibeam image of the central depression of the Kwintebank (FPS Economy) with the locations of the macrobenthos sampling points. (White dots: locations in and near central depression; black dots: reference stations outside closed area) ................................................................................... 86 Figure 7-2: Median grain size and percentage of coarse fraction for all stations for the different sampling periods. ..................................................................................................................................... 88 Figure 7-3: Number of species (/0.3m²) counted per station per season (top) and total number of species per season with an indication of the relative proportion of the major taxonomic groups in density (bottom). ...................................................................................................................................... 89 Figure 7-4: Shannon-wiener diversity index for the different sampling locations for the period spring 2003 – autumn 2007. ................................................................................................................................ 90 Figure 7-5: Average number of individuals (+ SD on replicates) for all sampling periods for all locations separately. ................................................................................................................................. 90 Figure 7-6: Average number of individuals per m² (+ st. dev.) with indication of the proportion of the major taxonomic groups. .......................................................................................................................... 91 Figure 7-7: Average number of ind./m² for the 6 most important species for the period Spring 2003 – Autumn 2007. ........................................................................................................................................... 92 Figure 7-8: The different proportions of juvenile and adult Nephtys cirrosa. ......................................... 92 Figure 7-9: Proportion in density of the different feeding types .............................................................. 92 Figure 7-10: Average biomass / m² (wet weight corrected to AFDW) for the different sampling periods. Left: all species included; right: excluding Echinocardium cordatum and Ensis arcuatus........ 93 Figure 7-11: Average body size of the 3 major taxonomic groups. ......................................................... 93 Figure 7-12: CA of the samples of stations ZG06, ZG07, ZG09 and ZG10 with indication of the different sampling years and associated species. ..................................................................................... 94 Figure 7-13: Density, diversity and number of species for different locations on the Kwintebank for the period spring 2003-autumn 2007........................................................................................................ 95 Figure 7-14: Average individual biomass (gAFDW) of the polychaetes for the different locations on the Kwintebank. ....................................................................................................................................... 96

viii

LIST OF TABLES Table 2-1: Overview of the historical data records used from the Marine Biology Section of Ghent University. ................................................................................................................................................ 10 Table 2-2: Comparison of different parameters between historical and recent data in the intensively extracted central Kwintebank area. .......................................................................................................... 17 Table 4-1: ANOSIM R values and significance levels for the different sources of variation. ................. 25 Table 4-2: Overview of the univariate parameters (density, species richness and diversity N1) and the top 10 most abundant species in the community for the different delineated species assemblages. ........ 29 Table 4-3: Anosim values for the different grouping factors from the multivariate analysis (KB = Kwintebank, MB = Middelkerkebank, BR = Buitenratel and OD = Oostdijck). .......................... 40 Table 4-4: Anosim values for factor position and year ............................................................................ 46 Table 5-1: ANOSIM R values for different sources of variation. ............................................................ 52 Table 5-2: Anosim R values for different sources of variation in spring and autumn for extraction zone 1. ............................................................................................................................................................... 57 Table 5-3: Anosim R values for the different sources of variation in spring and autumn. ...................... 61 Table 6-1: ANOSIM global R values for the different sources of variation and R values for the pair wise tests between the different levels within one group. ........................................................................ 71 Table 6-2: Univariate measures and taxonomic composition of 4 bank stations in spring and autumn 2007. ......................................................................................................................................................... 81 Table 7-1: Table showing the 6 most abundant species per sampling period, indicating the percentage of total density they represent. ................................................................................................................. 91

ix

EXECUTIVE SUMMARY The exploitation of sand in the Belgian Part of the North Sea (BPNS) started in 1976, and the extracted volume rapidly increased from 29,000 m³ to a current extraction of >1,800,000 m³ in 2009. In the Royal Decree of 1 September 2004, 3 zones were (re)defined as exploitation zones and one zone as exploration zone. The exploitation zones are located on the Thorntonbank and Gootebank (Zone 1 a & b), Kwintebank, Buitenratel and Oostdyck (Zone 2 a, b & c) and Sierra Ventana (Zone 3 a & b), while the exploration zone (Zone 4) is located on the Hinderbanks. Extraction activities used to be concentrated on the Kwintebank (>75 %), but after closure of the central depression in 2003, extraction activities partly shifted towards the Buitenratel and Oostdyck. Additionally, extensive extraction takes place on the Thorntonbank. The sand extraction concessions are linked to a compensation used by the authorities for the continuous monitoring of the possible impact of aggregate exploitation on the marine environment. ILVO-Fisheries has been involved in the biological monitoring since the beginning of the extraction activities; three ecosystem components (macrobenthos, epibenthos and demersal fish) are sampled each year in spring and autumn from the RV Belgica. In the beginning of the monitoring study, sampling used to be concentrated on the Kwintebank and focussed on the impact of sand extraction on commercial fish species and fisheries in general. Since the redefinition of the extraction zones in 2004, however, the sampling effort gradually increased in space over the years and the focus was directed to ecosystem-based impact assessment, which entailed an adaptation of the sampling strategy to BACI (Before/After/Control/Impact) designs. The aim of this report was to give an overview of the different results collected during monitoring in the framework of the extraction of marine sand on the BPNS. Therefore, results of both long term evaluation and short term projects were described. Furthermore, the current sampling strategy was evaluated and recommendations for future monitoring were made. In Chapter 2, the consequences of long term sand extraction on the macrobenthic community of the Kwintebank were discussed. Therefore, long term trends in the macrobenthic community were analysed and historical data were compared with recent data. The 30 years of extraction on the Kwintebank caused changes in sediment composition and bottom topography with subsequent changes in the macrobenthic community. However, changes are site-specific and depend on the intensity of extraction. The very intensively extracted areas of the Kwintebank accommodated little benthic life with a low biomass, while densities on the less intensively extracted areas were slightly higher. The dominant fauna of the Kwintebank is characteristic for highly dynamic areas and therefore better adapted to survive the disturbance of sand extraction activities. Important issues that complicated the interpretation of the results were the change in sampling and processing techniques over the years and the lack of a documented ‘before impact’ situation. The study in Chapter 2 stressed the need for sound baseline studies and comparable reference stations. Since 2004, base line studies were performed and the sampling strategy was adapted with the designation of reference stations. In Chapters 4, 5 and 6, the base line studies and current monitoring in the different extraction/exploration zones for respectively macrobenthos, epibenthos and demersal fish were described and conclusions with regard to future monitoring were formulated. For each ecosystem component, we started with a general analysis to biologically situate the extraction/exploration zones within the BPNS. These analyses were based on a multitude of samples taken between 2004 and 2008 and spread over the BPNS. Afterwards, specific analysis per extraction zone were performed. x

In Chapter 4, results of the macrobenthos were presented. The multivariate pattern on the regional scale was best explained by the coastal-offshore gradient and the different macrobenthic communities defined on the BPNS. The samples in the surroundings of zone 3 were closely associated with the coastal zone and characterised as Macoma balthica, Abra alba and transitional community between Abra and Nephtys cirrosa. Densities were lower than expected from the coastal zone, but this is probably due to the impact of intensive dumping in zone 3. Because part of this site (3b) is in use as dumping site, no extraction takes place in zone 3 and no further analyses were carried out. Within zone 2, the coastal-offshore gradient manifested itself, with 2a+b as transition between coast and off shore communities and consequently the samples could be characterised by several communities: the Abra, Nephtys, Ophelia limacina and transitional community between Abra and Nephtys. Zone 2c on the other hand, was closer related to the offshore community and was characterised by the Nephtys community. Zone 2c had lower density and diversity compared to 2a+b. Zone 1 and 4 are clearly represented by the offshore community and characterised by the Nepthys and Ophelia community. The baseline in zone 1a & 1b was performed in autumn 2004 and 95 different taxa were observed. The samples of both Thorntonbank and Goote Bank clustered together, indicating that the species composition was more or less homogeneous between both areas. This also indicated that the extensive extraction on the Thorntonbank did not result in significant changes on the level of macrobenthos. The possible reference stations on the Bank Zonder Naam, Goote Bank and Thorntonbank differed from the extraction areas, but this was mainly because of differences in abundance of interstitial species Hesionura elongata and Polygordius appendiculatus. These differences were caused by a difference in sieving procedure (reference samples were sieved alive, impact samples were sieved after fixation). When using the same sieving method, these references can be reliably used in the impact assessment. For the univariate measures density and species richness (S), higher densities and S were found in the gullies compared to the sandbank slopes and tops, but no significant differences were observed between Thorntonbank and Goote Bank. Again, due to the different sieving method, densities in the reference stations were lower compared to the impact stations. For the moment, as long as sand extraction remains extensive on the Thorntonbank, a lower sampling effort is sufficient inside the extraction area. Since September 2009, a reference area was demarcated within the concession zone, and no additional reference stations are sampled in this area. As long as extraction takes not place on the Goote Bank, no samples are taken in this area. Zone 2 is the most intensively used extraction area in the BPNS and most studies have been concentrated in this area (Chapters 2 and 7). Although many samples were collected over the years, reference samples were only collected recently. Furthermore, since the closure of the central depression on the Kwintebank, extraction activities shifted towards the Buitenratel and the Oostdyck, and since 2006 samples are also taken in these areas. In total 125 different taxa were identified in the samples analysed for zone 2. Species composition was relatively similar for all samples, but some small significant differences were observed between bank and gully samples and between the different sandbank areas. The largest difference was found between the reference stations of the Middelkerke bank and the other impact and control stations. Therefore, Middelkerke bank stations are not reliable as reference and since 2008, they are no longer sampled. The other reference stations can be used, but although species composition is quite similar, the differences in univariate measures are considerable. It is, however, difficult to distinguish between natural variation and anthropogenic disturbance due to the presence of the coast-offshore gradient. Therefore, additional reference stations on each sandbank should be designated. These will be implemented in spring 2010. Furthermore, univariate measures

xi

differed significantly between years and seasons, which indicates the importance of simultaneous sampling in impact assessment. Zone 4 is still an exploration zone, but this will probably change in the near future. Therefore, a good baseline and BACI design are necessary. During the baseline study, 116 different taxa were identified and the dominant species were Nephtys cirrosa and interstitial species like Hesionura elongata and Polygordius appendiculatus. Furthermore, some rarer species, typically found in coarse sand, were found (e.g. Syllidae species). The community structure was significantly influenced by position of the sample on the sandbank. Also for density and diversity, position was a structuring factor. Gully samples were characterised by the highest diversity and density and should thus be avoided in future extraction activities. Extraction should focus on the top of the sandbanks, since these are the areas with lowest density and diversity. Since no extraction took place yet, before-impact and before-control samples were already taken in autumn 2009 to form the basis of a BACI design. Furthermore, we strongly advise to safeguard a part of the exploration area as reference area. Chapter 5 presents the results from the baseline studies on epibenthos. In total 92 species were identified, and Ophiuroidea (44%), Bivalvia (26%) and Caridea (16%) were the dominant taxa in terms of density. Spatial variation was the most important factor determining epibenthic community structure in the BPNS, as both distance from shore (coastal versus offshore) and location of the extraction zone best explained the dispersion of the samples. The extraction zones were quite well defined as indicated by the significant R values of the pairwise tests. Temporal variation was of minor importance on a larger geographical scale, when trying to explain the observed patterns in species composition. Densities, however, differed significantly between spring and autumn with the highest densities in autumn. Similar as for macrobenthos, zone 3 was closest related to the coastal samples, as were the samples from the Kwintebank (2 a & b) in zone 2, while zone 1 and zone 4 were most similar to the offshore samples. Density was significantly higher in the coastal zone and in zone 3, while diversity N1 was significantly lower due the dominance of a few species in the coastal zone. Zone 3 was dominated by ophuiroids, while in the offshore located extraction zones, caridean shrimps, brachyuran crabs and hermit crabs were most abundant. Since no extraction takes place in zone 3 due to dumping activities, no further analyses were carried out. The extensive extraction activities in zone 1a were not expected to result in significant changes in epibenthos community. This hypothesis is confirmed by the strong resemblance between the data from zone 1a and the reference stations located on the Thorntonbank. The reference stations on the Goote Bank are not a good proxy for epibenthos communities on the Thorntonbank. This difference in species composition is probably related to the more heterogeneous habitat around the Goote Bank. Temporal variation (seasonal and interannual) is the primary structuring factor for community composition. Univariate measures were also significantly different between years. This implies that simultaneous sampling is very important in future monitoring. The position of the sample in a gully or on a bank does not influence community structure, but especially in spring, density and S were significantly higher in gully samples. Species composition in zone 2 is structured temporally with significant interannual differences in the spring samples. Furthermore, density and biomass showed a high interannual variation, and autumn values were higher than spring values. Therefore, evaluation of the impact of extraction activities in zone 2 should be done by comparing impact stations and adjoining reference stations on a year-to-year basis per season. Furthermore, due to the presence of the coastal-offshore gradient, differences in species composition were observed between zone 2c and zone 2 a & b and the reference stations. The xii

univariate measures also showed a gradual decrease from the Kwintebank towards the Oostdyck. Hence, the allocation of stations situated on the same sandbank system are strongly recommended and will be assigned in 2010. Based on data from 2007, no obvious negative impact of sand extraction on density, biomass or diversity was observed. The samples from the baseline study in zone 4 were unequally distributed over years, seasons and positions (gully-bank) and most samples were taken in spring 2008. Therefore, factors structuring the epibenthic community were difficult to representatively assess. Nevertheless, preliminary results of spring samples indicate that temporal variation and the position of the samples in a gully or on a bank are structuring factors. No differences between reference stations and exploration stations were observed, thus the reference stations are representative for future impact assessment. Gully samples showed a significantly higher density and species number than top samples. For future impact assessment, it is important to simultaneously sample the reference and impact stations, and both reference and impact should be located on a similar position on the sandbank. Both sample types should be sampled before the impact takes place. This was already done in autumn 2009 and will be repeated in spring 2010. The baseline studies on demersal fish communities are discussed in Chapter 6. In total, 69 fish species were identified with a dominance of Perciformes, Pleuronectiformes and Gobiidae. In agreement with the epibenthos analysis, spatial variation was identified as the most important factor determining species composition in the BPNS. Both distance from shore (coastal versus offshore) and location of the extraction zone best explained the observed dispersion of the samples. The stations of zone 3 and station 215 on the Kwintebank (zone 2) resembled best the coastal stations, while the other extraction zones 1, 2 and 4 were most related to the offshore stations. Differences between the different offshore extraction zones were small, indicated by the high degree of overlap and the very small R values. The coastal samples were dominated by sole, plaice and whiting, while the offshore samples were dominated by dab, lesser weever and reticulated dragonet. Although temporal variation was of minor importance for species composition on a larger geographical scale, seasonal variation was highly significant for density and diversity with highest densities, S, and N1 in autumn. Seasonal variation was the primary structuring factor in the samples related to zone 1. Different trends were observed in spring and autumn but in both seasons, interannual variation was the primary structuring factor, stressing the need for simultaneous sampling of all stations. Spatial variation was of minor importance with small differences between bank and gully samples but no differences between Thorntonbank and Goote Bank in spring. In autumn on the other hand, no differences in species composition were observed between gullies and banks but differences between Goote Bank and Thorntonbank were significant. The extensive extraction activities did not result in significant changes in fish fauna. For the univariate measures density and S, no significant seasonal variation was observed but interannual variation was significant. Gully and bank samples differed only in terms of species number S with higher values in the gullies. No differences were observed between Thorntonbank and Goote Bank for univariate measures. Currently, no extraction takes place on the Goote Bank, but additional stations should be assigned whenever extraction would start. Fish communities in zone 2 were structured by temporal (seasonal and interannual) variation and by the position of the sample in a gully or on a bank. Impact of sand extraction was subordinate to the natural variation and did not have a significant influence. For the univariate measures, temporal variation was much smaller but position of the sample did play a role with higher density and S in xiii

gullies. Bearing this in mind, evaluation of the impact in zone 2 should be done by comparing impact and adjoining reference stations on a year-to-year basis per season. Furthermore, due to the high natural variability in this area because of the coast-offshore gradient, extra reference stations should be designated in the same bank-gully system as the impact stations. This will be implemented in spring 2010. Preliminary results of spring samples taken during the baseline study in zone 4 revealed an influence of sampling year and sampling position (gully-bank) on species composition. The sandbank tops were characterised by a dominance of lesser weever. A similar fish community was observed in exploration area and in the reference area on the Bligh Bank, indicating that the selected reference stations on the BB can be used as a reliable proxy. The univariate measures were mainly temporally influenced with fluctuating densities and S between the years. Therefore, it is very important for future impact assessment that both reference and impact samples are sampled simultaneously and are located on a similar position on the sandbank. Furthermore, both sample types should be sampled before the impact takes place. This was already done in autumn 2009 and will be repeated in spring 2010. In Chapter 7, the evolution of a macrobenthic community after cessation of the extraction activities is discussed. The Kwintebank, a highly dynamic sandbank in the western BPNS, is highly extracted, mainly due to the short distance to the coast and the suitability of the sand for construction purposes. After 30 years of exploitation, a depression was formed in the central part of the Kwintebank. To allow the geomorphologic rehabilitation of the area, it was decided to cease extraction in this part of the exploitation zone in 2003. The changes of the macrobenthic communities, resulting from the closing of the area for extraction, were studied by taking samples at six locations in the area twice a year for 5 years. Density, species richness, diversity, biomass and multivariate analyses were used to evaluate the ecological status of the area. The poor macrobenthic community that was found in the central depression directly after the cessation of the extraction activities, clearly evolved to a community with higher densities, species richness, diversity and biomass within three years after the sand extraction activities had stopped. After 4 years, the species composition changed and larger, less opportunistic species started to appear. The conclusions and key issues from the different chapters are discussed in Chapter 8 with some recommendations towards future monitoring and adaptation of the monitoring strategy.

xiv

1. INTRODUCTION 1.1. Sand extraction on the Belgian Part of the North Sea The exploitation of sand on the Belgian Part of the North Sea (BPNS) started in 1976 with an annual exploitation of about 29,000 m³. In 1977, exploitation increased by about one order of magnitude and became more and more important. In the mid 1990’s already 1,700,000 m³ yr-1 was extracted, reaching a maximum of more than 1,900,000 m³ yr-1 in 2001. In 2007-2008, subtidal sand extraction yielded about 1,600,000 m³ yr-1 (ICES, WGEXT 2008). Till 2008, extraction activities on the BPNS were mainly concentrated on the Kwintebank (> 75 %), due to the presence of suitable sand and its close location to the harbour, enabling a cost-effective exploitation (Degrendele et al., in press). In 2000, the formation of a depression in the central Kwintebank was observed while investigating the bathymetric and the morphological evolution of the Kwintebank by single beam profiles (Degrendele et al., press). Since federal legislation prohibits further exploitation when a deepening of > 5 m occurs with respect to the most recent hydrographical charts, this area had to be closed for extraction activities in February 2003. Until 2004 extraction was allowed in two areas on the BPNS. A new Royal Decree (RD) of September 1 of 2004 described new conditions, geographic limits and procedures for granting licenses for the extraction of sand on the BPNS. In this procedural decree, 3 sand extraction zones were defined, divided in sectors for which a concession can be issued:    

Exploitation Zone 1: Thorntonbank (Sector 1a) en Gootebank (Sector 1b) Exploitation Zone 2: Kwintebank (Sectors 2a en 2b, from 2010 changing to one zone 2ab) and Buitenratel and Oostdyck (Sector 2c) Exploitation Zone 3: Sierra Ventana (divided in sectors a en b) An exploration Zone 4 (Hinderbanken) has been defined. Based on the results of the exploration research, new sectors for exploitation will be defined in this zone. The maximum total surface will be 46 km².

The accessibility for the zones is defined as follows:   



sectors 1a, 2c and 3a are open for exploitation all year round; sector 1b is only open for exploitation during the months March, April and May; sectors 2a and 2b are open for exploitation alternately for a period of 3 years. Between 2005 and 2008, sector 2a was open and 2b was closed. In 2009, 2a was closed and 2b was opened, except for the central depression which remained closed. From 2010 onwards, there is no alternation, since both sectors are combined as 2ab. The sector 2ab is open, except for the central and the northern depression. sector 3b is closed for exploitation as long as the sector is still being used as a dumping site for dredged material.

The concessions are linked to a compensation used by the authorities for the continuous monitoring of the influence of the exploitation on the marine environment and to prepare for the Environmental Impact Assessments (EIA’s) (Federal Government for economic affairs, ILVO and MUMM)

1

Figure 1-1: Overview of the concession zones for the exploitation and exploration of sand extraction on the Belgian Continental Shelf (source: MUMM).

The blackboxes, installed on the different vessels operating on the BPNS since 2006, give information on the positioning as well as the evolution of the amount and intensity of extraction activities (Figure 1-2). These blackbox data allow for exact locations of the extraction activities.

2

2002

2003

2004

2005

2006

2007

Figure 1-2: Blackbox data of the extraction activities on the BPNS from 2002 to 2007 (source: FOD Economie – Fonds voor Zandwinning).

3

1.2. Monitoring the benthos Since the beginning of the extraction activities, the Environmental Monitoring section of ILVOFisheries has been involved in the monitoring. The aim of this monitoring is to evaluate the impact of the sand extraction activities on the soft sediment fauna. The biological monitoring section of ILVO has been sampling the three main components of the soft sediment fauna over de last 30 years: Macrobenthos: all organisms, larger than 1 mm, that live most part of their life in the sediment of the sea bottom. Macrobenthos is sampled using a Van Veen grab (0.1 m²) and sieved over a 1 mm sieve after fixation to retain the animals. Main groups of species are bristle worms (Polychaeta), shellfish (Mollusca), crustaceans (Crustacea) and echinoderms (Echinodermata).

Epibenthos: all organisms living most part of their life on top or just above the bottom of the sea. They are sampled using an 8 m beam trawl rigged with a 22 mm shrimp net. Main groups are crustaceans, echinoderms, shell fish and cnidarians.

Demersal fish: all fish species living most part of their life in close association with the bottom of the sea. They are sampled using an 8 m beam trawl rigged with a 22 mm shrimp net. Main groups are perciforms, gadoids (Gadiformes), clupeids (Clupeiformes) and gobies (Gobiidae).

The macrobenthos is well suited as an indicator for assessing the state of the marine environment and to detect possible changes caused by human and natural impacts. Macrobenthic organisms are closely associated with the seafloor for the major part of their life (Snelgrove & Butman, 1994). Most species have a limited mobility and are therefore forced to adapt to seasonal fluctuations and anthropogenic influences (Boyd et al., 2003). Climatologic circumstances (e.g. cold winters, heavy storms, ...), food supply, predation, as well as human impacts (such as dumping of dredge spoils, sand extraction, beam trawl fisheries) can have serious repercussions on the resident populations (Kröncke & Bergfeld, 2001). Both macrobenthos and epibenthos species are a major food source for demersal fish. Every possible change in the invertebrate communities might cause a change in the demersal fish 4

communities. Therefore, a complete and ecologically meaningful study should incorporate the three ecosystem components. Since the spatial and temporal variation of the infaunal species and communities is closely linked to the physical environment, some abiotic parameters were also studied.

1.3. General description of sampling strategy and methods used The monitoring campaigns of ILVO-Fisheries take place each year in spring and autumn. In general, most sampling is done from the RV Belgica, although in some cases also other vessels, like the RV Zeeleeuw, are used. The macrobenthos is sampled with a Van Veen grab with a sampling surface of 0.1 m², and its contents are sieved after fixation with formaldehyde over a 1 mm sieve. From each sample, a subsample for grain size analysis was taken with a 3.6 cm Ø core. Additionally, depth and position were determined during the sampling campaign. In the lab, the samples are then sorted, and the species are identified, counted and weighed wet. The epibenthos and the demersal fish are sampled using an 8 m beam trawl with a fine-meshed shrimp net (stretched mesh width 22 mm in the codend) and a bolder-chain but no tickler chains to minimise the environmental damage. Each trawl has lasts about 30 minutes, covering a distance of approximately 3 nautical miles (trawling at an average speed of 3.5 knots). All animals caught are sorted on board and counted. Epibenthic species are weighed and the length of all fish species is measured. For a number of locations within the sand extraction zones, long term data series are available. However, most locations were sampled only once or a few times in the framework of designated studies. Sampling in the past aimed at the impact of sand extraction on fisheries, and hence focused on commercial fish species. Recently (last 5 years), the focus shifted towards ecosystem-based impact assessment, and research was more environment oriented. Therefore, Before/After Impact/Control designs were set-up during the last 5 years. In total 1,668 macrobenthos samples between spring 2004 and autumn 2008 were taken into account for this report, and with these a general analysis was performed to situate the extraction zones within the BPNS (paragraph 4.1). Afterwards, more detailed analyses per extraction zone were performed with a selection of the samples (described in the separate paragraphs). In Chapter 2 and 7, 223 samples from the last 25 years were used for long-term impact analysis on the Kwintebank. The general analysis of the epibenthos and demersal fishes is based on data from 274 samples spread over the whole BPNS over the last 5 years. As for the macrobenthos, more detailed analyses were performed per extraction zone with a selection of the samples. Diverse univariate and multivariate statistics were used during the analysis of the data of all three ecosystem components. Parameters are density, species richness, diversity and biomass (the latter only for epibenthos).

5

Figure 1-3: Overview of the sampled locations for macrobenthos (left) and epibenthos and demersal fish (right)

1.4. The aim of this report According to the agreement between the Federal and Flemish government (Belgisch Staatsblad, N 2006-261), ILVO-Fisheries is expected to deliver a report to the advisory committee (Art. 2). According to Art. 4, ILVO-Fisheries is responsible for:     

The biological monitoring of the impact of sand- and gravel extraction on the benthic marine ecosystem. The chemical monitoring of the impact of sand and gravel extraction on the sediment and associated fauna. The sampling of reference areas decided on in cooperation with the federal government The analysis of sediment samples taken by the Fund for Sand extraction of the Ministry of Economic affairs and by ILVO-Fisheries, in the framework of this agreement. Report annually of the advances of the above mentioned activities.

This report aims at giving an overview of the different results collected during monitoring in the framework of the extraction of marine sand on the BPNS. Results of both long term evaluation and of short term projects are described in the different chapters. Recently (since 2004), the number of sampling stations has increased in order to include reference stations or to designate more stations in the impact area. The current monitoring strategy is investigated and evaluated and suggestions are made for future monitoring. Based on the results, some recommendations for policy makers and stakeholders are made regarding a sustainable use of the marine aggregates on the BPNS.

6

2. EVALUATION OF THE CONSEQUENCES OF LONG TERM SAND EXTRACTION ON THE STRUCTURAL CHARACTERISTICS OF THE MACROBENTHOS COMMUNITIES1 2.1. Introduction Several studies have been investigating the effects of marine aggregate extraction on the benthic fauna (de Groot, 1979; Poiner & Kennedy, 1984; Pagliai et al., 1985; Jones, 1986; Van Moorsel, 1994; Kenny & Rees, 1994; Desprez, 2000; Sarda et al., 2000; Van Dalfsen et al., 2000; Van Dalfsen & Essink, 2001; Boyd & Rees, 2003; Boyd et al., 2003; Guerra-Garcia et al., 2003; Sanchez-Moyano et al., 2003; Boyd et al., 2004; Newell et al., 2004a; Newell et al., 2004b; Robinson et al., 2005; Simonini et al., 2005). However, most of these studies are limited to the initial effects of extraction, and thus do not address the effects of dredging over the life time of a typical commercial extraction licence. Most serious physical impacts of sand extraction are related to substratum removal, alteration of the bottom topography and sediment composition, changes in depth and current strength and the creation of plumes due to the disturbance by the drag head and due to screening (Newell et al., 1998; Boyd et al., 2003; Hacking, 2003). The direct removal of species and individuals is considered as the main biological impact. In most cases, it is found that dredging causes an initial reduction in abundance, species diversity and biomass of the benthic community in the extraction area (Pagliai et al.,1985; Jones, 1986; Van Moorsel, 1994; Kenny & Rees, 1994; Desprez, 2000; van Dalfsen et al., 2000; Sardá et al., 2000; van Dalfsen and Essink, 2001; Guerra-Garcia et al., 2003; Newell et al., 2004b; SanchezMoyano et al., 2004; Simonini et al., 2005). However, the impact of sand extraction is not limited to these short term effects, in casu the immediate removal of fauna, but also causes long term changes in the structure of the benthic community due to habitat changes (Desprez, 2000). Outside the boundaries of the extraction area, some studies found an effect mainly due to screening operations (Poiner & Kennedy, 1984; Desprez, 2000; Newell et al., 2004a, b). Increased sedimentation and resuspension due to dredging of clean (mobile) sand are generally thought to be of less concern, as the fauna inhabiting such deposit areas tends to be adapted to naturally high levels of suspended sediments caused by wave and tidal current action (Newell et al., 2004b). For this study on the long term impact of sand extraction on the Kwintebank and surrounding areas, different locations on the Kwintebank and on the Oostdyck with different intensities of extraction are considered. The comparison of historical and recent data gives an idea of the evolution of the macrobenthos on a heavily extracted sandbank on the Belgian Continental Shelf. We aimed at clarifying the impact of the long term extraction activities on the macrobenthos. So far, no paper on long term effects of sand extraction on the BCS was published. Therefore, the available historical and recent data of the Kwintebank of ILVO-Fisheries were analysed. In addition, historical data from different studies from Ghent University were used for comparison. Analysis of this dataset allowed assessing the impact of sand extraction on the macrobenthos.

1

Moulaert I., Hillewaert H. and Hostens K. (2005) Analysis of the long term consequences of sand extraction on the macrofauna communities. In: Rapport SPEEK UGent, CLO, AZTI – Belspo: 25-36 7

2.2. Compilation of historical data 2.2.1. Data gathered by ILVO-Fisheries Since 1979, macrobenthic samples have been collected at different locations on the Belgian Continental Shelf. Some of these stations are located on the Kwintebank and the Oostdyck in the area licensed for sand extraction (Figure 2-1). Samples from the north of the Kwintebank (ZG01) have been collected since 1978, although not continuously (Figure 2-2). Only few of the old samples were well preserved. Since 2004, a new location in the north (ZG11) was sampled in the newly formed depression. In the central part of the Kwintebank, samples have been collected since 1998, just south of the central depression (ZG04), and since 2003 in the central depression (ZG05-10). Only the samples from 2003 of the central depression will be used for comparison as later samples have higher densities due to the process of 'recovery' (see chapter 7). According to the black box data, all these sampling stations on the Kwintebank were located in the most intensively extracted areas (Figure 2-3). In the northern part of the Oostdyck, sampling has been conducted since 1978 (ZG02), although the macrobenthos samples have only been processed from 1996 onwards.

Figure 2-1: Map of the Belgian Continental Shelf with location of the old and new sand extraction areas.

At every location, twice a year (February/March and September/October), 4 Van Veen grabs were collected: 3 for identification of the macrobenthos and 1 for analysis of the sediment. Although, the same sampling and processing technique (Van Veen grab, fixation before sieving and a 1 mm sieve) has been used throughout these 25 years, it has to be mentioned that the positioning of the ship has been very variable. This might hamper the long term interpretation, as sampling on top of the bank or on the slope can have different results in the sediment characteristics as well as in the macrobenthic community. 8

Figure 2-2: Detailed map of the sampling locations on the Kwintebank. White dots: ILVO locations; red dots: locations of the historical records from UGent/Marine Biology. (Background Multibeam images: Fund for Sand Extraction)

Figure 2-3: Black box records indicating extraction intensity summarised over the period 1996-2005 (Fund for Sand Extraction).

9

2.2.2. Data gathered by UGent/Marine Biology Only few locations have been sampled by ILVO-Fisheries in the early stages of sand extraction activities. As mentioned above, the historical samples from only one location on the Kwintebank have been processed so far. Therefore, some historical data from the Kwintebank and surrounding areas, sampled in the framework of other studies, were used to compare with more recent data (Table 2-1). Vanosmael & Heip (1986) used a 1 mm sieve after fixation, producing biological data comparable to the data available from ILVO-Fisheries. Waeterschoot (1980) used an 870 µm sieve, so the comparison of the biological data needs to be done with caution. The mesh size used by Vanosmael et al. (1982) was a lot smaller (250 µm), so only the sedimentological data from this study were used for comparison. Data from two sampling locations in the north of the Kwintebank were available from the Macrobel2 database, although no information was available on the mesh size used. For the Oostdyck, data was available from the thesis study of Meheus (1981). In this study, a 870 µm sieve was used after fixation. The same locations were sampled in the following year and data from this study were available from the Macrobel database. Table 2-1: Overview of the historical data records used from the Marine Biology Section of Ghent University.

Kwintebank

Oostdyck

Vanosmael & Heip, 1986 Waterschoot, 1980 Vanosmael, 1982 Macrobel databank Meheus, 1981 Macrobel databank

sampling year 1980-1984 1979-1980 1978 1979-1980 1980 1980-1981

mesh size station comparable to 1 mm 40004 ZG06 & ZG09 870 µm 40006 ZG04 250 µm SB05 ZG06 & ZG09 ? ZB43-44 ZG11 870 µm ZS2 ZG02 ? ZS2 ZG02

2.2.3. Data analyses The different sediment fractions as well as the median grain size were used for comparison of the sediment characteristics. For the macrobenthos, the parameters density (ind./m²), species number (in # species / 0.3 m²) and the Shannon-Wiener diversity index were used. The number of species is presented per 0.3 m² (the content of 3 Van Veen grabs) since for some historical data, only a sum of the 3 replicates was available. The multivariate technique non-metric multidimensional scaling (MDS) was used to evaluate changes in the community over time. The MDS was performed using the square root transformed abundance data, after elimination of rare species (present in less than 3 replicates) by means of Primer v6 software (Plymouth Routines in Multivariate Ecological Research; Clarke and Gorley, 2006). Data from only one season were used in order to minimise seasonal variation. Sediment grain size was determined using the sieving tower method with six sieve sizes (63 µm, 125 µm, 250 µm, 500 µm, 1 mm and 2 mm). Sediment grain size of the samples since spring 2007 was determined by laser diffraction using the Malvern Master Sizer after sieving the samples over a 1.6 mm sieve to determine the coarser part. This change in method might introduce possible differences in the results. The definition of coarse sediment thus changed from larger than 2 mm to larger than 1.6 mm.

2

http://www.vliz.be/Vmdcdata/macrobel/index.php 10

2.3. Results 2.3.1. Northern part of the Kwintebank ZG01, 1980-2007 The long term sediment data series revealed no major changes in the past 27 years (Figure 2-4), although a variation between the different samples could be noticed (225-390 µm). Samples from the late 1970's and the beginning of the 1980's showed less variation than samples from the last 10-15 years. The high median grain size and absence of silt fraction in 2007 probably results from the new method of analysing sediment samples. Median grain size (µm)

10

Silt fraction (%)

400

8

350

6 300

4 250

silt fraction (%)

median grain size (µm)

450

2

200

150 S A S A S A S A S A S A S A S A S A S A 80

81

82

83

84

85

86

87

88

89

S A S A S A S A S A S A S A S A S A S A S A S A S A _

95

96

97

98

99

00

01

02

03

04

05

06

0

07

Figure 2-4: Median grain size and percentage of silt recorded for station ZG01 for the period 1980-2007.

No major evolution could be found in any of the different macrobenthic community parameters, i.e. density, diversity or number of species, over the last 25 years (Figure 2-5). Samples from 1985 and 1988, as well as from 2000 and 2001, showed high density values. Autumn density values of these years were twice as high compared to the values in other years. Generally, densities at the beginning of the 80’s were comparable to densities of the last 10 years. Peaks occurring in autumn 1985 and 1988 were also found in autumn 2000 and 2001. The peaks of 1985, 1988 and 2000 were caused by the increase of similar species (mainly Spiophanes bombyx, Scoloplos armiger, Nephtys spp. and some amphipods like Bathyporeia guilliamsoniana and Urothoe poseidonis). Juvenile individuals of the echinoderm Echinocardium cordatum were responsible for the high density found in autumn 2001.

11

# ind / m² 2000 1600 1200 800 400 0 S A S A 84

85

S A _

88

S A S A S A S A S A S A S A S A S A S A S A S A S A _

95

96

97

98

99

00

01

02

03

04

05

06

07

# species / 0.1m² 25 20 15 10 5 0 S A S A 84

85

S A _

88

S A S A S A S A S _

95

96

97

98

A S A S A S A S

99

00

01

02

A S A S A S A S A

03

04

05

06

07

Diversity 4,0 3,5 3,0 2,5 2,0 1,5 1,0 0,5 0,0 S A S A 84

85

S A _

88

S A S A S _

95

96

A S A S

97

98

A S A S

99

00

A S A S

01

02

A S A S A

03

04

05

S A S A 06

07

Figure 2-5: Average number of individuals per m² (+ standard deviation), total number of species and average diversity per season for station ZG01 for the period 1984-2007.

Polychaetes were the most important taxonomic group (Figure 2-6). In most samples, they represented more than 60 % of the total density. Crustaceans (mainly amphipods) were the second most important group of species. Bivalves were never of much importance for the macrobenthic density of this area. In autumn of 2001, there was a sudden increase of juvenile echinoderms.

12

Polychaeta

Mollusca

Echinodermata

Crustacea

Others

100% 80% 60% 40% 20% 0% S A

S A

84

85

S _

A

88

S A S _

95

A S

96

A S

97

A S

98

A S

99

A S

00

A S

01

A S

02

A S A

03

S A

S A

S A

05

06

07

04

Figure 2-6: Relative abundance of the major taxonomic groups for station ZG01.

The MDS analysis of the autumn samples indicated a small shift in species composition over the years (Figure 2-7). This could be attributed to a shift from long living, more sessile species towards smaller species that prefer coarser sediment. S tress: 0,25 S tress: 0,23

S tress: 0,2

Before11990

1

1

2 2 1990-1999

2

3

3

3 After 2000

Figure 2-7: MDS of the spring (left) and autumn (right) samples of station ZG01 area for the period 1979-2007.

Very intensively extracted western zone: ZB43&ZB44, 1979-1980 vs. ZG11, 2004-2005 No historical sediment data were available from ZB43 and ZB44. Sediment data of the last 4 years indicate a slight coarsening of the sediment (Figure 2-8). median grain size

coarse fraction silt fraction

700 600 500 400 300 200 100 0

10 8 6 4 2 0 S

A 2004

S

A 2005

S

A 2006

S

A 2007

Figure 2-8: Median grain size and percentage of silt recorded for station ZG11 for the period 2004-2007.

Samples from 1979 and 1980 had low faunal densities, comparable to 2004, spring 2005 and 2006. In autumn 2005 and especially in 2007, high densities were found (Figure 2-9). The extremely low values found in spring 2006 might be a result of the high extraction activities in the area in 2005. The 13

evolution found in density was mainly caused by the interstitial polychaetes Polygordius appendiculatus and Hesionura elongata, resulting in a fast recovery concerning density of the area. The macrobenthic species composition of the older samples was partly similar to the samples from the last four years. However, some bivalves and bigger sessile bristle worms were replaced by mobile and interstitial species. Ophelia limacina, Scoloplos armiger, Spio spp. and Spisula spp. were more abundant in the historical samples, whereas Polygordius appendiculatus and Microphthalmus spp. were more abundant in the recent samples. Polychaetes are the most important group present in this area. Only few molluscs and crustaceans were found, which is an indication that these are less resistant to disturbance.

# ind / m² 2000

ZB43

ZB44

ZG11

1600 1200 800 400 0 S

A

S

1979

A 1980

S _

A

S

2004

A

S

2005

A

S

2006

A 2007

# species

ZB43 spec/0,3m²

ZB44 spec/0,3m²

30

ZG11 spec/0,3m²

ZG11 spec/0,1m²

25 20 15 10 5 0 S

A

S

1979

A 1980

S _

A

S

2004

A

S

2005

A

A 2007

ZB43 /0,3m² ZG11 /0,3m²

Diversity

S

2006

ZB44 /0,3m² ZG11 /0,1m²

4 3 2 1 0 S

A 1979

S

A 1980

S _

A 2004

S

A 2005

S

A 2006

S

A 2007

Figure 2-9: Historical and recent records of density, species number and diversity for the very intensively extracted northern Kwintebank area.

14

2.3.2. Central part of the Kwintebank Intensively extracted zone: 40006, 1979-1980 vs. ZG04, 1998-2007. The data from 1979 and 1980 were gathered very close to the location of the more recent samples. Waeterschoot (1980) used a smaller sieving mesh size (870 µm). The historical samples had a slightly coarser grain size and a smaller percentage of silt (Figure 2-10). Sediment data indicated a small increase in median grain size from 1998 to 2007. Density, species richness and diversity were lower in 1979-80 compared to 1998-2001 (Figure 2-11). In autumn 2001, a peak in density, diversity and species richness was recorded, but since then all values decreased. Since 2005, the values were low (< 400 ind./m²), comparable to the historical samples. The peak in autumn 2001 was caused by the increase of different species. Med grain size (µm)

Silt fraction (%)

400

3,0

350

2,5

300

2,0

250

1,5

200

1,0

150

0,5

100

0,0 S 79

A

S 80

A

S _

A 98

S

A 99

S 00

A

S 01

A

S

A 02

S

A 03

S 04

A

S 05

A

S

A 06

S

A 07

Figure 2-10: Median grain size (bars) and percentage of silt (lines) for the low intensively extracted central Kwintebank area.

Polychaetes were the dominant taxonomic group present, but their proportion was smaller than in the northern part of the Kwintebank (Figure 2-12). Bivalves, amphipods and echinoderms were almost absent from the historical samples, but were clearly more important in the samples from 1998 to 2003. Since autumn 2003, bivalves and echinoderms have almost disappeared again. Polychaetes and amphipods remained the dominant taxonomic groups. Spio spp., Ophelia limacina, Hesionura elongata and Nepthys cirrosa were the most abundant species in the samples from 1979 and 1980. Urothoe poseidonis, Spiophanes bombyx and Scoloplos armiger were the most important species between 1998 and 2003. Also several bivalve species like Abra alba, Kurtiella bidendata and Tellina fabula where present in this area during this period. After 2003, a clear shift in species composition occurred (visualised by the plots of the multivariate analysis (Figure 2-13)): Nephtys cirrosa (adult and juvenile individuals) became the most abundant species during the last 4 years, followed by the amphipod species Bathyporeia guilliamsoniana, B. elegans, Urothoe poseidonis and U. brevicornis. Interstitial species were hardly found during the last 4 years. The proportions of adult and juvenile individuals remained relatively stable over the whole period (60 % juveniles/40 % adults).

15

# ind / m² 3500 3000 2500 2000 1500 1000 500 0 S

A

S

1979

A

S

1980

_

A

S

1998

A

S

1999

A

2000

S

A

2001

S

A

2002

S

A

S

2003

A

2004

S

A

2005

# species

Rep 1

Rep 2

60

# spec / 0.1m²

# spec / 0.3m²

S

A

2006

S

A

2007

Rep 3

50 40 30 20 10 0 S

A

1979

S

A

1980

S _

A

S

A

S

A

S

A

S

A

S

A

S

A

S

A

S

A

S

A

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

S

S

S

S

S

S

S

S

S

S

Diversity 4,0 3,5 3,0 2,5 2,0 1,5 S

A

1979

S

A

1980

_

A

1998

A

1999

A

2000

A

2001

A

2002

A

2003

A

2004

A

2005

A

2006

A

2007

Figure 2-11: Average number of individuals per m² (+ standard deviation, bars), total number of species (lines) and diversity per season for station ZG04 (Central Kwintebank area).

16

Polychaeta

Mollusca

Echinodermata

Crustacea

Other

100%

80%

60%

40%

20%

0% S

A 79

S

A 80

S _

A 98

S

A 99

S

A 00

S

A 01

S

A 02

S

A 03

S

A 04

S

A

S

05

A 06

S

A 07

Figure 2-12: Relative abundances of the major taxonomic groups for the intensively extracted central Kwintebank area. S tress: 0,22

S 98-S 03

A 03-A 07

Figure 2-13: MDS plot with indication of the 2 different sampling periods.

Very intensively extracted zone: SB5, 1978 vs. ZG09, 2003&2007 and 40004, 1980-1984 vs. ZG06, 2003&2007 Historical data from stations SB5 and 40004 were compared with data from 2003 and 2007 for stations ZG09 and ZG06, respectively. There was a large difference in median grain size and depth between SB5 sampled in 1978 and ZG09 sampled in 2003 (Table 2-2). However, in 2007, the sediment was coarser, and again more similar to 1978. The macrobenthos data were not compared since a much smaller sieving mesh size was used in 1978. Table 2-2: Comparison of different parameters between historical and recent data in the intensively extracted central Kwintebank area.

1978 (SB5) depth (m) median grain size (µm) mud content (%) gravel content (%) organic material (%) skewness average density (ind/m²) average number of species (#/0.3m²) average diversity

17

13,5 517 0 0,24 2,92 -0,11 ----

2003 (ZG09) 16 272 0,79 1,16 0,62 -0,01 ----

2007 (ZG09) 16 420 0 5,07 ------

1980-1984 (40004) -516 0,55 7,19 --780 ±236 21 ±3 2.95 ±0.31

2003 (ZG06) 15,5 329 0,75 4,16 0,99 -0,14 143 13 1,75

2007 (ZG06) 15,5 497 0 6,01 --1210 24 2,07

For stations 40004 and ZG06, a much smaller median grain size was recorded in the 80’s compared to 2003 (Table 2-2). But again, the sediment recorded in 2007 was more similar to the beginning of the 1980’s. Macrobenthic density, species number and diversity were lower in 2003. Historical samples were mainly characterised by a high percentage of interstitial species. Some interstitial species (Hesionura elongata, Polygordius appendiculatus) still occurred in the 2003 samples, although at lower densities, but were again highly abundant in 2007. Several other species (Pisione remota, Streptosyllis arenae, Sphaerosyllis bulbosa) were no longer found after many years of intensive extraction in the area. 2.3.3. Northern part of the Oostdyck: ZS2, 1980-1981 vs. ZG02, 1996-2005 The long term sediment data from the northern part of the Oostdyck showed a comparable pattern. No clear increase or decrease of the median grain size was recorded over the last 30 years, except for 2004 (Figure 2-14). During the 1980's, sediment data was relatively stable, but more variation was found in median grain size during the last decade. Median grain size (µm)

med grain (µm)

siltfraction (%)

silt fraction (%)

600

1,4

500

1,2 1,0

400

0,8 300 0,6 200

0,4

100

0,2

0

0,0 S A S A S A S A S A S A S A S A S A S A S A S A 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989

S A S A S A S A S A S A S A S A S A S A S A S A S A _

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Figure 2-14: Median grain size and percentage of silt recorded for the low intensively extracted northern Oostdyck area for the period 1979-2007.

In both historical and recent samples, the species community found was extremely poor with low densities, diversity and species number (Figure 2-15). In the spring of 1980, a higher density was found due to the presence of the interstitial species Hesionura elongata, which represented more than 80 % of the total number of individuals. Since a smaller mesh size was used in these historical samples, no further conclusions can be made regarding the higher abundance of this species. On the other hand, the interstitial species Polygordius appendiculatus was not found in the historical samples. Hesionura elongata, Polygordius appendiculatus, Protodrilus spp., Ophelia limacina and Nephtys cirrosa were the most important species present in the northern part of the Oostdyck in the recent samples. In spring 2003, an increase was found in the number of species and diversity. Six months later also density had increased as a result of the high abundance of Hesionura elongata and Polygordius appendiculatus. In 2004 and spring 2005, a clear decrease in species number and diversity was found which can be linked to the higher sediment grain size found in this period. The only species present during this period were small interstitial polychaetes such as H. elongata, P. appendiculatus and Pisione remota. Since spring 2005, density remained stable, but both species number and diversity have clearly increased.

18

Density (ind/m²) 1500 1200 900 600 300 0 S A S A

S A S A S A S A S A S A S A S A S A S A S A S A

1980 1981 _ 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

# species / 0,1m² 12 10 8 6 4 2 0 S A S A

S A S A S A S A S A S A S A S A S A S A S A S A

1980 1981 _ 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Diversity 3,0 2,5 2,0 1,5 1,0 0,5 0,0 S A S A

S A S A S A S A S A S A S A S A S A S A S A S A

1980 1981 _ 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Figure 2-15: Average number of individuals (+ standard deviation, bars), total number of species (lines) and average Shannon-Wiener diversity index per season for the low intensively extracted northern Oostdyck area. Polychaeta

Mollusca

Echinodermata

Crustacea

Other

100% 80% 60% 40% 20% 0% S A S A 1980 1981

S A S A S A S A S A S A S A S A S A S A S A S A _ 1996 1997 1998 1999

2000 2001 2002 2003

Figure 2-16: Relative abundance of the different taxa for station ZG02.

19

2004 2005 2006 2007

2.4. Discussion When comparing historical with recent data or when using long term biological data records, many aspects have to be taken into account. Next to differences in sampling locations, sampling techniques and sample processing, there is also natural variation. To investigate the variation caused by anthropogenic impacts, reference data are needed to distinguish naturally caused changes from anthropogenic ones. A good reference area should be determined through a base line study, before any dredging activities have taken place (Boyd et al., 2003). The Kwintebank has been a sand extraction area for 30 years. Although some historical data on the macrobenthos from the early years of extraction exist, no base line data from before that period are available. Both short term studies and models emphasise the importance of the initial period in relation to the main changes in the community structure of the macrobenthos (Kenny & Rees, 1996; Desprez, 2000; van Dalfsen et al., 2000; Sardá et al., 2000; van Dalfsen and Essink, 2001; Guerra-Garcia et al., 2003; Newell et al., 2004b; Sanchez-Moyano et al., 2004; Simonini et al., 2005). The initial impact of sand extraction on the macrobenthos of the Kwintebank will remain unknown. Due to differences in sampling techniques and sample processing, caution has to be taken when interpreting the data from different datasets of the Kwintebank. It is shown that higher densities of smaller species are found when using a smaller mesh size (James et al., 1995). Change in the exact position of the samplings might also hamper the long term interpretation as sampling on top of the bank or on the slope can have different results regarding the sediment characteristics as well as in the macrobenthic species composition. The impact of sand extraction on macrobenthic communities depends on numerous factors: extraction method and intensity, sediment type and mobility, bottom topography and current velocity. The impact also depends on the type of the original macrobenthic community and is thus site specific (Newell et al., 1998; Desprez, 2000; van Dalfsen et al., 2000; Boyd et al., 2004). Sand extraction area II is located in a highly dynamic area. Results from the sediment analysis indicated a high variation between samples and seasons. Therefore, this unstable environment is occupied by macrobenthic communities composed of species that are mobile, have a short life span and are adapted to living in disturbed areas. Bigger, long living species are seldom found on the top of the sandbanks. A species typically found in both historical and recent samples, and on all sandbank stations was Nephtys cirrosa. This rapidly swimming bristle worm constructs temporary burrows and lives in wave and current exposed locations (Budd, 2005). Furthermore, the interstitial species Hesionura elongata is a typical inhabitant of these sandbank systems. Since no baseline data are available, the question remains whether these species were the pristine inhabitants of the sandbank or whether they became more abundant with increasing disturbance due to sand extraction. According to Hiscock et al. (2005), Nephtys cirrosa and Hesionura elongata are both tolerant to extraction activities. Boyd et al. (2003) also found higher abundances of Hesionura elongata, Spiophanes bombyx and juvenile Nephtys spp. in intensively dredged sediments compared to the reference areas. Furthermore, these species are typical for highly dynamic sand bank systems (Moulaert et al., 2007). Results from the sediment analysis of the central depression of the Kwintebank indicated a clear decrease of the median grain size. This was not found for the area just south of the depression, nor in the north of the Kwintebank. Sediment data from the north of the Kwintebank showed a high variation in median grain size. This is mainly due to the highly dynamic character of the area whereby small variations in position of the ship relative to the sampling point could contribute to this variation. 20

Samples from the late 1970's and the beginning of the 1980's showed less variability than samples from the last 10 years. The same results were found for the north of the Oostdyck. Positioning was probably more correct in the last decade so the greater variation might be solely attributed to an increased disturbance of the area. Whether this disturbance is caused by increasing sand extraction activities, which disturb the bottom and create a less stable environment, or by natural variation is open for discussion. The designation of comparable reference stations could help to distinguish between natural variation and human disturbance. In the very intensively extracted zone in the north of the Kwintebank, a clear depression is being formed (Figure 2-2). This is mainly due to the fact that most of the extraction activities have taken place in this area since the closure of the central part of the Kwintebank (Degrendele et al., 2005). This probably has its repercussions on the local hydrodynamics, the grain size distribution and the benthic community. A clear difference is found between the very intensively extracted western zone and the less intensively extracted eastern zone of the northern part of the Kwintebank. In the former, a larger grain size was found and macrobenthic density and species richness were much lower. Diversity on the other hand did not differ much. Macrobenthic density and species number were higher in 1979 and also in autumn 2005, whereas samples from 1980 were poor, and comparable to the samples from 2004 and spring 2005. The cause of these peaks is unclear; it could be the result of the anthropogenic disturbance but more probably is that the conditions in spring 1979 and autumn 2005 were more suitable for macrobenthic recruitment. Interstitial species (Hesionura elongata, Polygordius appendiculatus and Microphthalmus spp.) represented a higher relative abundance of the density in the very intensively extracted western part. The change in species composition (more interstitial species) indicated a slight coarsening of the sediment. In 1978, a smaller median grain size was found in the north of the Kwintebank (Vanosmael et al., 1982). In the same study, high numbers of interstitial species were found as well but mainly because a 250 µm sieve was used. Long term macrobenthic data from the less intensively extracted eastern part of the north of the Kwintebank, showed no negative evolution in density or diversity. According to the sediment data, the environment was less stable in the last decade, while the macrobenthos community remained more or less the same. A higher variation was found in the 1980's. Some peaks in density occurred (1985, 1988, 2000 and 2001), but most likely these were not a result of the extraction activities since comparable peaks were found for other locations on the BPNS, which are not influenced by sand extraction. For the intensively extracted central depression of the Kwintebank, the comparison of historical data with data from 2003 (only one month after closure of that area for sand extraction) indicated a clear decrease of median grain size. Additionally, the macrobenthic density had decreased mainly due to the reduction or even disappearance of interstitial species. This decrease was not related to different processing techniques, because for both historical and recent samples, a 1 mm sieve was used after fixation of the samples. The reduction of interstitial species was probably caused by the decrease in median grain size. In a following chapter (Chapter 7), the results from the recovery of the macrobenthic community in that area are presented. It will be interesting to see if the community evolves to a community similar to the one found in the historical samples, or to a different community. The area south of the central depression did not show any changes between 1979-1980 and 2004-2005 (comparable univariate indices and species composition), but in between both periods, a higher number of species and a higher diversity were found. The peaks in 2001 of density, diversity and species number were also found for other, non-extracted areas on the BPNS. The negative evolution 21

from 2001 to 2005 is probably due to an increase in extraction activities in this area as a result of the closure of the central depression. On the Oostdyck, extraction activities only sporadically occurred due to the more offshore location and a less suitable grain size for the extraction industry. The Oostdyck has always hosted a relatively poor macrobenthic community as a result of the larger grain size. Little changes were found between the historical and the recent data.

2.5. Conclusion Many scientific studies concerning the monitoring of anthropogenic impacts suffer from the absence of a base line study. Another problem is the change in sampling and processing techniques over the years, which makes it difficult to compare historical and recent data. Therefore, conclusions should be drawn with care. After 30 years of extraction on the Kwintebank, changes in sediment and depth were observed, with subsequent changes in the macrobenthic communities. The results from the sediment analyses indicated that the effects of extraction on the sediment composition are site specific. They mainly depend on the sediment composition of the sandbank, the grain size of the extracted sediment and the intensity of extraction, as such the effects of extraction cannot be generalised. The very intensively extracted zones of the Kwintebank accommodated little benthic life with a low biomass. Consequently, these zones only offer a small food supply for other organisms. For the less intensively extracted areas of the Kwintebank, density was slightly higher. Long term data series from the less intensively extracted area in the northern part of the Kwintebank indicated no major changes in species composition, density and diversity over the last 25 years. Data series from the area just south of the central depression, where extraction activities have increased, showed a decrease in macrobenthic number of species and diversity during the last 4 years. These results indicate that sand extraction has a negative effect on the macrobenthic community. However, Nephtys cirrosa, Hesionura elongata, Polygordius appendiculatus, Spiophanes bombyx, Urothoe brevicornis, Urothoe poseidonis and Bathyporeia spp., the main species present on the Kwintebank, are able to survive in dynamic areas, and are probably adapted to survive the disturbance of sand extraction activities.

22

3. THE NEED FOR BASELINE STUDIES One of the major conclusions of the previous chapter was the need for sound baseline studies. Ideally, to be able to assess the impact of extraction activities, a BACI (Before, After, Control, Impact) design should be applied (Quinn and Keough, 2002). Therefore, a control site(s) should be designated which is similar in terms of habitat characteristics to the impact site, and both control and impact site(s) should be monitored before and after the impact took place to be able to statistically demonstrate the possible effect of impact. Unfortunately, as already mentioned in the previous chapter, past sampling in the framework of impact monitoring started after the initial impact. However, extraction is so far mainly concentrated in zone 2, extraction in zone 1 is extensive, zone 3 is not exploited and zone 4 is only an exploration area for the moment. By setting a good baseline for these extensive extraction zones and especially the exploration zone, it will still be possible to detect possible changes due to intensified or changing extraction activities. Furthermore, the designation of comparable adjoining reference stations is another measure for impact assessment, preferably before the impact took place but even after the start of the impact. Therefore, since 2004, ILVO-Fisheries started to monitor more intensively and several reference stations were designated for the different (potential) impact areas. It is very important that the reference stations closely resemble the impact stations in terms of physical environment and community structure to be able to detect effects of the impact. In the following chapters, the baseline studies for the three ecosystem components are described. The designated reference stations were evaluated and their suitability for impact assessment was tested. Furthermore, the most important sources of variation in the different areas were tested and some recommendations on future monitoring for impact assessment were formulated.

3.1. Data analysis First of all, the species data sets for all ecosystem components were reduced; for macrobenthos, all species occurring in only one sample were excluded; for epibenthos and demersal fish, all species observed in less than two fish tracks and occurring with a mean density of less than 0.01 individuals per 1000 m² were excluded from the analysis. The three ecosystem components were dealt with separately concerning density, biomass (epibenthos only), diversity and community structure. The number of individuals per sample and per species was converted to number of individuals per m² (macrobenthos) or per 1000 m² (epibenthos and demersal fish) (abundance). Biomass was expressed as grams of wet weight (WW) per 1000 m² and diversity was evaluated based on species richness (S) and on Hill’s diversity index N1 (Hill, 1973). Differences were tested with main effects ANOVA, followed by Tukey’s Post Hoc test when appropriate, and all analyses were done in Statistica 9. When assumptions were not met, values were natural log or square root transformed to meet the assumption of homoscedasticity (Levene’s test). The community structure of all ecosystem components was analysed using the multivariate techniques non-metric multidimensional scaling (MDS), ANOSIM (Analysis of Similarities) and SIMPER (Similarity Percentages Procedure) available in Primer v6 (Plymouth Routines in Multivariate Ecological Research; Clarke & Gorley, 2001). These analyses were based on 4th root transformed and reduced datasets of frequency of occurrence and density. Furthermore, for the overview of the macrobenthos data, cluster analysis by group average sorting based on a Bray-Curtis similarity dataset was performed. This cluster analysis was complemented with a SIMPROF test to define the significantly different groups within the cluster analysis. The clustering was visualised using a nonmetric Multi-Dimensional Scaling analysis (MDS).

23

4. BASELINE STUDIES MACROBENTHOS 4.1. General overview: situating the extraction zones within the BPNS The general overview of the macrobenthos is based on 1668 samples from 521 stations visited between 2004 and 2008 both in spring and autumn, spatially spread over the BPNS (Figure 4-1).

Figure 4-1: Location of all macrobenthos sampling stations visited during spring and autumn between 2004 and 2008.

A total of 169 species were identified from all samples (species list in Annex I). In terms of density, the species groups Polychaeta (56 %), Oligochaeta (20 %) and Mollusca (12 %) were most abundant (Figure 4-2).

24

Crustacea Echinordermata Mollusca Oligochaeta Polychaeta Other

Figure 4-2: Relative abundance of the different taxa collected in all macrobenthic samples.

Based on a cluster analysis of all samples, 6 large groups were delineated on the ca. 20 % level. These groups were further split up on the 30-40 % level and they were used as factors in the MDS, ANOSIM and SIMPER analysis (Figure 4-3). Other sources of variation taken into account were sampling year, sampling season and location on the BPNS (coastal versus offshore stations and the different extraction zones as factor). Table 4-1: ANOSIM R values and significance levels for the different sources of variation.

Source of variation

Global R

p

Year

0.047

0.001

Season

0.034

0.001

Location

0.144

0.001

Cluster group

0.812

0.001

Based on the results of the MDS and ANOSIM analyses of all data, R-values for temporal variation (both seasonal and annual) were negligibly small (Table 4-1) and could not explain the observed multivariate pattern when considering the variation on the BPNS. Location of the samples had a higher, but still quite small R value (0.144), indicating that location alone was a poor descriptor for the observed pattern. The pattern could best be explained by the sublevels determined by the 30-40 % level of the cluster analysis (Table 4-1, Figure 4-3). When taking a closer look at the species composition of these different clusters (SIMPER analysis), the four macrobenthic communities defined by Van Hoey et al. (2004) for the Belgian Continental Shelf could be distinguished: Macoma balthica community, Abra alba community, Nephtys cirrosa community and Ophelia limacina community with in between some transitional assemblages and some outlier samples (Figure 4-3).

25

Figure 4-3: MDS plot with different symbols for the different cluster groups and with indication of the different communities. The arrow indicates a decrease in mud content and an increase in median grain size.

Van Hoey et al. (2004) and Degraer et al. (1999) described the different communities as follows: 1) The Macoma balthica community is characterised by moderate densities and very low species number. It is present in sediments with high mud concentration (> 20-25 %) and low median grain size ( 75 % of the samples. These were: Oligochaeta (94 %), Hesionura elongata (86 %), Nepthys juv (81 %), Polygordius appendiculatus (79 %), Spiophanes bombyx (77 %).

Number of species restricted to x samples

30

25

20

15

10

5

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 16 17 18 25 26 28 36 43 18 55 59 61 62 66 72

Number of samples

Figure 4-10: Species range distribution

4.2.4. Community structure A very small Anosim R (0.099, p = 0.002) was found for the factor location, indicating that no real differences in community structure exist between Gootebank and Thorntonbank. A similar result was found for the position on the banks (R = 0.09, p = 0.001), indicating that Thornton and Gootebank are more or less homogeneous in terms of species composition.

Figure 4-11: MDS plots from all sampling stations in the extraction zones with indication of possible sources of variation.

32

Figure 4-12: MDS plot with indication of the different sampling locations

When including the reference stations in the multivariate analysis (Figure 4-12), the factor location is significant and higher compared to the analysis of the two extraction zones (Anosim R = 0.269, p = 0.001). Pair wise tests revealed that station 330 is similar in species composition with both Thorntonbank and Gootebank (non-significant differences). For the other reference stations, significant differences in species composition were observed with both extraction areas. SIMPER analysis revealed that the differences between BZN, WGR and WT and the Gootebank and Thorntonbank samples were mainly caused by differences in abundance of Oligochaeta, Hesionura elongata and Polygordius appendiculatus. Samples of stations BZN, WGR and WT were sieved alive, while all other samples were sieved after fixation. So, the interstitial species were able to escape in these reference samples causing differences in abundance. Although station 340 was sieved after fixation, the species composition differed significantly with Thorntonbank and Gootebank samples (pair wise test R = 0.535, p = 0.002 and R = 0.441, p = 0.005, resp.), indicating that station 340 needs to be reconsidered as reference station. 4.2.5. Density and diversity Density and species richness differed significantly between banks and gullies (Anova, resp. p = 0.00025 and p = 0.019) with highest density and species richness in the gullies (mean resp. 737 ± 65 ind. /m² and 12.5 ± 0.6 species) and lowest on the banks (mean 597 ± 51 ind./m² and 11 ± 0.5 species). Diversity was not influenced by the position of the samples. On the geo referenced maps showing the different sampling stations in the extraction area, these trends of lower density and species richness on the banks are clearly visible (Figure 4-15). Diversity seems to be lower on the Thorntonbank compared to the adjoining gully, while on the Gootebank, diversity seems a bit higher on the bank than in the gully. However, this is not confirmed statistically. Furthermore, no differences in density and diversity were found between Thorntonbank and Gootebank, which confirms the multivariate pattern. Average density on the Gootebank was 931 ± 83 ind./m² and avg species richness 13 ± 1. On the Thorntonbank, density was 805 ± 73 ind./m² and species richness was 14 ± 1. However, differences in density were observed between Gootebank and BZN (p = 0.011) and WT (p = 0.007) samples, these were probably caused by the differences in methodology of sieving the samples (fixed versus alive) (Figure 4-14).

33

1400

1200

Density (Ind/m²)

1000

800

600

400

200

0 Goote (bank)

Goote (gully)

Thornton Thornton 330 (bank) 340 (gully) (bank) (gully)

BZN (bank)

WGR (bank)

WGR (gully)

WT (bank) WT (gully)

Figure 4-13: Average densities for the different locations and positions. 18 Species Richness N1

16 14 12 10 8

6 4 2 0 Goote (bank)

Goote Thornton Thornton 330 (gully) (bank) (gully) (bank)

340 (gully)

BZN (bank)

WGR (bank)

WGR (gully)

WT (bank)

WT (gully)

Figure 4-14: Species richness and diversity N1 for the different locations and positions.

Figure 4-15: Distribution pattern of density (left), species richness (middle) and diversity (right) per sample on the Thorntonand Gootebank.

Bristle worms (Polychaeta) were relatively more abundant on the banks, while crustaceans and echinoderms were more abundant in the gullies of both Thorntonbank and Gootebank. The lancelet (Branchiostoma lanceolatum) only occurred in the gullies of Thorntonbank and Gootebank. 34

100% 90% 80% Other 70%

Polychaeta Oligochaeta

60%

Mollusca Echinodermata

50%

Crustacea 40%

Cephalochordata

30% 20% 10% 0% G o o te

G o o te

(b an k )

(g u lly )

T h o r n to n T h o r n to n (b an k )

(g u lly )

330

340

B ZN

W GR

W GR

WT

WT

(g u lly )

(g u lly )

(b an k )

(g u lly )

(b an k )

(b an k )

(g u lly )

Figure 4-16: Relative abundance of the different taxa for the different positions and areas.

4.2.6. Conclusion In both extraction zones 1a and 1b, a baseline study was performed in autumn 2004. The extensive extraction activities on the Thorntonbank were not expected to result in significant changes on the level of the macrobenthos. This hypothesis was confirmed by the strong resemblance between Zone 1a, where extraction has already taken place and zone 1b, where no extraction activities were performed. The current sampling strategy aims more at control/impact designs and therefore possible reference stations were included in the analyses. From the reference stations included, station 330 is a very good reference, strongly resembling both extraction zones in terms of species community. Station 340 differed too much in species composition to be treated as reference and should no longer be used as reference for zone 1. The other stations (WGR, BZN, WT) seem promising, but the same sieving method (after fixation) should be used as for the other stations. Preferably, stations should be monitored in the same year to exclude interannual variation, especially concerning densities. As long as the extraction remains extensive, no impact is expected and a lower sampling effort is sufficient. Since spring 2008, 8 to 10 impact stations are sampled in zone 1a (where extraction is present), and 7 reference stations are monitored to be able to assess the impact. In zone 1b, no further samples were taken, since no extraction takes place. Since, September 2009, a reference area within the concession zone is demarcated and additionally, the samples taken in this area can be used as control samples.

35

4.3. Zone II (Kwintebank, Buitenratel and Oostdyck) 4.3.1. Sampling strategy Zone II is the most intensively used sand extraction area on the BPNS, and the activities used to be mainly concentrated on the Kwintebank. However, black box data clearly show a shift with increasing extraction activities on the Buitenratel since 2005 after the closure of extraction zone 2b. To be able to accurately monitor the future impact of sand extraction in Zone II, ILVO-Fisheries started to sample more intensively in different areas within Zone II in 2006. Samples were taken on:    

Northern part of the Kwintebank (grid of 12 locations, KBN stations) Central part of the Kwintebank (grid of 6 locations, ZG stations, see previous chapter) Northern part of the Buitenratel (grid of 9 locations, BRN stations) Northern part of the Oostdijck (transect over bank of 6 locations, ODN stations)

Since autumn 2008, another two transects were sampled to be able to assess the impact when zone 2b reopens and to follow the second shift in extraction activities on the Buitenratel. These data are, however, not included in this report:  

Central part of the Buitenratel (transect of 6 locations, BRC stations) Southern part of the Kwintebank (transect of 6 locations, KBZ stations)

Reference stations were designated to allow for comparison with the impacted stations, and to be able to assess the potential impact of sand extraction. One reference transect over the bank of 7 stations was situated on the southern part of the Buitenratel (BRZ stations). Another transect of 4 locations was sampled on the northern part of the Middelkerkebank (MBN stations) (Figure 4-17). Furthermore, the longer term (since 1996) monitoring stations 315 (north of Oostdijck) and ZG03 (south of Kwintebank) were also used as reference stations.

36

2a

2c

2b

Figure 4-17: Location of the sampling stations in Zone II and reference stations.

4.3.2. Abiotic data Sediment grain size differed significantly between the different sand bank systems (p = 0.0017). Overall, median grain size was smallest on the Middelkerkebank (avg 295 ± 22 µm) and largest on the Oostdyck (384 ± 8 µm). For Kwintebank and Buitenratel, similar grain sizes were observed (resp. 363 ± 9 µm and 344 ± 14 µm). However, grain size on the southern part of the Buitenratel (BRZ stations) was much smaller compared to the other stations (Figure 4–18). Furthermore, a significant difference was observed between bank and gully samples (p = 0.0039) with overall higher median grain size on the banks (364 ± 8 µm) compared to the gullies (323 ± 13 µm). Nevertheless, no correlation was found between depth and median grain size (Figure 4-19).

37

Figure 4-18: Median grain size (µm) for the different sampling stations (data from spring 2007). 900 800

Median grain size (µm)

700 600

R2 = 0.0008

500 400 300 200 100 0 0

5

10

15

20

25

30

35

Depth (m)

Figure 4-19: Correlation between depth and median grain size.

4.3.3. In general In total 125 different taxa were identified in extraction zone 2 and the reference stations. Eighteen taxa occurred in only one sample, and another 18 taxa were found in only 2 samples (Figure 4-20). Only 4 species occurred in > 60 % of the samples and most species were present in 5-45 % of the samples. The 4 most common species were 1) Nephtys juvenile (85 %), 2) Nephtys cirrosa (81 %), 3) Hesionura elongata (69 %) and 4) Oligochaeta species (61 %). 20

Number of species restricted to x samples

18 16 14 12 10 8 6 4 2 0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 19 20 22 23 25 26 27 28 29 30 32 36 37 49 57 61 63 65 73 78 82 88 92 98 115 116 161182 214223

Number of samples

Figure 4-20: Species range distribution of extraction zone II and surroundings. 38

4.3.4. Community structure Species composition was relatively similar for all samples, which is reflected by a clustering of the samples in the MDS plot (Figure 4-21). Both interannual and seasonal variation was negligible, with very small R values (resp. Anosim R = 0.043 and R = 0.051) (Table 4-3). Differences between extraction zones (2a, 2b and 2c) and reference stations were small as well (R = 0.098), and smaller than differences between the different bank areas (R = 0.157) and the differences between bank and gully samples (R = 0.18) (Table 4-3). However, even the differences in species composition between the bank and gully samples and the different bank areas were small. Largest species composition differences were found between Middelkerke bank and the other bank systems (Table 4-3). This difference, however, is rather a difference between coastal and offshore fauna, than a difference caused by impact. Thus, MBN samples are not reliable as reference stations for Zone II, and therefore, they are no longer sampled since autumn 2008.

Figure 4-21: MDS plots for all samples and reference samples of Zone II with Impact, Location and Position as sources of variation.

39

Table 4-3: Anosim values for the different grouping factors from the multivariate analysis (KB = Kwintebank, MB = Middelkerkebank, BR = Buitenratel and OD = Oostdijck).

Global R

p

Bank system

0.16

0.001

Impact

0.098

0.001

Position(gully-bank) Year Season

0.18 0.043 0.051

Pairwise test KB-MB KB-BR KB-OD BR-MB BR-OD OD-MB Ref-2a Ref-2b Ref-2c 2a-2b 2a-2c 2b-2c

R 0.358 0.108 -0.04 0.381 0.24 0.322 0.084 0.191 0.125 0.067 0.094

0.023

p 0.001 0.003 0.8 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.1

0.001 0.001 0.001

4.3.5. Density and diversity

Figure 4-22: Density, species richness and diversity per station. Values are averaged over years, seasons and replicates.

The effect of impact was not significant for density in 2006 and 2007 (main effect ANOV A, p = 0.13). Location and position on the banks, on the other hand were highly significant (resp. p = 0.005 and < 0.0001); densities from the Buitenratel were significantly lower compared to the other areas and densities in the gullies were much higher than on the banks (Figure 4-23). In contrast with the species 40

composition, densities did differ between years (p = 0.0003) and between seasons (p = 0.002). In 2006, overall densities were lower compared to 2007, and they were also lower in spring compared to autumn. For species richness (S) and diversity (N1), we again did not find an effect of impact. Position on the banks was significant for both S (p =< 0.0001) and N1 (p = 0.007) with higher values in the gullies (Figure 4-23). Location in the area was only significant for N1 (p = 0.025), not for S (p = 0.09). Again, year and season were significant factors with the lowest S and N1 in 2006 and in spring. 20 1800

S N1

18 1600

16

Density (Ind/m²)

1400

14

1200

12

1000

10

800

8

600

6

400

4

200

2

0

0 MB bank Ref

MB gully Ref

KB a bank

KB a gully

KB b bank

BR bank

BR bank Ref

BR gully Ref

OD bank

OD gully

OD gully Ref

MB bank Ref

MB gully Ref

KB a bank

KB a gully

KB b bank

BR bank

BR bank Ref

BR gully Ref

OD bank

OD gully

OD gully Ref

Figure 4-23: Distribution of density (left), species richness (S) and diversity N1 (right) for the different areas and positions within the areas.

Overall, bristle worms (Polychaeta) were relatively more abundant on the banks compared to the gullies. Echinoderms, molluscs and crustaceans were more abundant in the gullies. Some small differences in taxon distribution were apparent between the bank areas. Molluscs were best represented on the Middelkerkebank and Kwintebank, probably because these areas still have coastal characteristics with higher contents of organic material and phytoplankton. Oligochaeta were considerably more abundant in the Oostdyck area (Figure 4-24). 100% 90% 80% 70%

Other

60%

Polychaeta Oligochaeta

50%

Mollusca

40%

Echinodermata

30%

Crustacea

20% 10% 0% MB MB KB a KB a KB b BR BR BR OD OD OD bank gully bank gully bank bank bank gully bank gully gully Ref Ref Ref Ref Ref

Figure 4-24: Relative abundance of the higher taxa in the different areas and on the different positions. 41

4.3.6. Conclusion Zone II is the most intensive extraction area in the BPNS and most studies have been concentrated in this area (chapter 2 and 7). Many samples were collected over the years, but most reference stations were only sampled relatively recently. The community analysis revealed that the reference locations on the Middelkerkebank are not reliable as reference, since this is still a more coastal dominated sandbank system. Therefore, these stations were no longer sampled in 2008. The other references BRZ, 315 and ZG-03 can be used as reference, but although species composition is quite similar between these reference stations and the stations in the different areas, the differences in univariate measures are considerable. Since there is a natural gradient (coast-offshore) present in this system, it remains difficult to differentiate between natural variation and variation caused by anthropogenic disturbance. Therefore, it would be recommendable to assign additional reference stations on each sand bank, which are located closer to the effective extraction areas. Furthermore, species composition is not temporally influenced, but univariate measures are seasonally and interannually influenced. This asks for simultaneous sampling of impact and reference samples.

4.4. Zone III As long as extraction zone 3b is in use as dumping site Br & W S1, no extraction activities take place. Since this site is characterised by intensive dumping activities, the sediment consists of fine muddy sand and is not suited for extraction. Therefore, no samples in the framework of sand extraction monitoring were taken. However, ILVO takes several samples in this area in the framework of monitoring the effects of the disposal of dredged material in the BPNS. The reader is thus referred to the report by Van Hoey et al. (2010) for recent biological information of this area. Also extraction area 3a is not in use for extraction, however 4 stations are sampled since autumn 2008 but no results are available yet. However, as soon as one intends to start sand extraction in this area, ILVO fisheries should be notified in time to be able to set up a BACI design to reliably assess the impact.

4.5. Zone IV (Hinderbanks) 4.5.1. Sampling strategy In the new Decree of 2004, the area of the Hinderbanken was appointed as an area of exploration to evaluate the possibility for sustainable sand or gravel extraction. In the exploration area, 129 locations were sampled in a regular grid, over a period of four years (2004-2008) with a Van Veen grab (0.1 m² sample) (Figure 4-25). All sampling was done in spring to exclude seasonal patterns in the interpretation of the results.

42

Figure 4-25: Sampling locations indicating the year of sampling.

4.5.2. Abiotic data The exploration zone of the Hinderbanken is a sandbank-gully system, with depth varying from 10 m to more than 35 m. The sediment of the area mainly consists of particles with a 300-450 µm median grain size, with the exception of some areas where a high percentage of mud (in the gullies) or gravel and larger cobbles were found (Figure 4-26). A very weak correlation was found between the composition of the sediment and depth (i.e. bank or gully) (Spearman r = 0.23, p = 0.01) (Figure 4-27). Especially the very coarse sand (300-420 µm) is of interest for potential extraction, since this is essential for beach replenishment. A seismic survey performed by the RCMG-UGent shows the thickness of these layers which are of potential interest (Figure 4-26).

43

Figure 4-26: Median grain size at the different sampling stations in exploration zone IV(left). Thickness of the very coarse sand layer (Source: RCMG-UGent) (Right).

1000

Median Grain Size (µm)

900 800 700 600 500 400 300 200 100 0 0

10

20

30

40

50

Depth (m)

Figure 4-27: Relation between depth and median grain size for exploration zone 4.

4.5.3. In general In total 116 different taxa were found, of which most taxa occurred in a restricted number of samples. Thirty-three species were restricted to only one sample and 16 species were restricted to two samples. Only 5 species occurred in >75 % of the samples (Figure 4-28). These common species were: 1) Hesionura elongata (in 87 % of the samples), 2) Nepthys cirrosa (81 %), 3) Nepthys juvenile (80 %), 4) Polygordius appendiculatus (78 %) and 5) Oligochaeta sp. (78 %).

44

Number of species restricted to x samples

35 30 25 20 15 10 5 0 1

2

3

4

5

6

7

9

10 12

13 14 16

18 19 20 22 23 25 33 34 36 41 42 44 45 46 47 70 100 103 105 112

Number of samples

Figure 4-28: Species range distribution for zone 4

4.5.4. Community structure Community structure was significantly influenced by position (gully/slope/top) of the sample in the bank-gully system (Anosim R = 0.27, p = 0.001). Interannual variation did not influence species composition (Anosim R = 0.027, p = 0.185) (MDS). Pair wise tests discerned differences in species composition between top and gully samples, with species composition of the slope samples being transitional (Figure 4-29), (Table 4-4). The multivariate pattern was best explained by the factor depth (Bio-env, r = 0.3). Median grain size was less important as explanatory factor (Bio-env, r = 0.198).

Figure 4-29: MDS plot for all samples of zone IV with position in the gully-bank system as source of variation.

45

Table 4-4: Anosim values for factor position and year

Global R

p

Position

0.27

0.001

Year

0.027

0.185

Pairwise test gully – slope gully – top slope – top

R 0.22 0.46 0.24

p 0.001 0.001 0.002

The SIMPER top 3 contribution percentages of species to the different positions can be listed as follows: Top: Nepthys cirrosa (22 %), Hesionura elongata (18 %), Polygordius appendiculatus (13 %) Slope: Nephtys cirrosa (20 %), Hesionura elongata (18 %), Nephtys juvenile (15 %) Gully: Oligochaeta (17 %), Polygordius appendiculatus (16 %), Hesionura elongata (14 %) 4.5.5. Density and diversity Differences between the different positions in the bank-gully system were also observed in terms of density, species richness and diversity (Figure 4-30). Density differed significantly between the different positions (Anova, p 350 µm, with some very high peaks, probably caused by a high number of shell fragments. The median grain size of ZG05 fluctuated around an average of 250 µm with lower grain sizes in spring compared to the autumn samples (except for a coarser grain size in spring 2007). The two stations located in the middle of the depression (ZG06 and ZG09) had a similar sediment composition by the end of the study period (350400 µm), although the temporal evolution found was different. The sediment of station ZG06 was more or less stable (average median grain size of 350 µm). A clear coarsening of the sediment, from 200 µm to 400 µm, was found for Station ZG09. For the stations of the eastern slope of the depression (ZG07 and ZG10) no increase nor decrease of the median grain size was found. Both locations showed a variation between the subsequent samplings with an average median grain size around 270 µm (Figure 7-2).

87

Median Grain size (µm)

1230 1220 870

500

Mar/03

450

Sep/03

400

Mar/04

350

Sep/04

300

Mar/05

250 Sep/05

200

Mar/06

150 100

sep/06

50

mar/07

0

sep/07

ZG05

ZG06

ZG07

ZG08

ZG09

ZG10

Coarse fraction (%) Mar/03

35

Sep/03

30

Mar/04

25

Sep/04

20

Mar/05 sep/05

15

mar/06 sep/06

10

Mar/07

5

Sep/07

0

ZG05

ZG06

ZG07

ZG08

ZG09

ZG10

Figure 7-2: Median grain size and percentage of coarse fraction for all stations for the different sampling periods.

7.3.2. Macrobenthos - Species richness In total 124 different macrobenthic species were found in the period March 2003 - September 2007. For each station a different pattern in the number of species was found. For most stations the number of species was lowest in the samples from spring 2003. Species number in station ZG05 in September 2006 was extremely high (50) (Figure 7-3). Taking all stations into account, the highest species richness was found in September 2007 (67) whereas only 32 different species were counted in March 2003. The proportion of the different taxa in the total amount of species did not substantially vary over the different sampling periods, except for a slight increase of crustaceans. Polychaetes or bristle worms accounted on average for 43 % of the species composition, crustaceans (mainly amphipods) 30 % and molluscs (mainly bivalves) 15 % (Figure 7-3).

88

Number of species / 0.3m² 50

mar/03

45

jun/03

40

sep/03

35

mar/04

30

sep/04

25

mar/05

20

sep/05

15

mar/06

10

sep/06

5

mar/07

0

sep/07 ZG05

ZG06

ZG07

ZG08

ZG09

ZG10

Number of species 100%

70 60

80% Polychaeta

50

60%

40

Crustacea Mollusca

30

40%

Echinodermata Other groups

20

20% 10

0%

0 S03 Su03

A03

S04

A04

S05

A05

S06

A06

S07

A07

S03 Su03 A03 S04 A04 S05 A05 S06 A06 S07 A07

Figure 7-3: Number of species (/0.3m²) counted per station per season (top) and total number of species per season with an indication of the relative proportion of the major taxonomic groups in density (bottom).

7.3.3. Macrobenthos - Diversity The Shannon-Wiener diversity index showed a similar slightly increasing trend, from < 2 in spring 2003 to > 2 in autumn 2007, except for station ZG09 where diversity was already higher in the first month after the closure of the area (Figure 7-4). For station ZG05, diversity was not only low in spring 2003, but also in autumn 2003, although since then a major increase was recorded, with a diversity of 4 in autumn 2006. Station ZG08 had the lowest diversity overall. The diversity of stations ZG06 and ZG07 showed a slightly increasing trend from 2003 to 2005, but slightly decreased again over the last two years. Diversity of the macrobenthos at station ZG10 initially increased, but decreased again towards spring 2005 and fluctuated since then with high values in autumn compared to lower values in spring (Figure 7-4).

89

Average diversity 4,0 3,5 3,0

ZG05

2,5

ZG06

2,0

ZG07

1,5

ZG08

1,0

ZG09

0,5

ZG10

0,0 S03

S03

A03

S04

A04

S05

A05

S06

A06

S07

A07

Figure 7-4: Shannon-wiener diversity index for the different sampling locations for the period spring 2003 – autumn 2007.

7.3.4. Macrobenthos - Density For all stations, an increasing trend in density was found: extremely low densities in March 2003 and higher values towards the end of the sampling period. Clearly higher densities were found for station ZG05 in autumn 2006 and for ZG06 and ZG08 in autumn 2007. Taking into account all stations, a minimum average of 90 ind./m² was found in spring 2003 and a maximum of more than 710 ind./m² in autumn 2007. Station ZG08 had overall the lowest densities (except for 2007) (Figure 7-5). Spring samples mar/04 sep/05 mar/07

Average density / m² (2003-2007)

1800

Summer sample sep/04 mar/06 sep/07

Autumn samples mar/05 sep/06

1600 1400 1200 1000 800 600 400 200 0 ZG05

ZG06

ZG07

ZG08

ZG09

ZG10

Figure 7-5: Average number of individuals (+ SD on replicates) for all sampling periods for all locations separately.

Polychaetes and crustaceans represented the largest fraction of the overall density (Figure 7-6). The relative abundance of the polychaetes clearly increased compared to a decreasing relative abundance of the crustaceans. In 2003, only 50 % of the overall density was taken up by polychaetes, whereas by the end, in autumn 2007, Polychaetes represented almost 90 %. The most dominant species present in the central depression were adult and juvenile Nephtys cirrosa, Urothoe brevicornis, Hesionura elongata, Polygordius appendiculatus and Spiophanes bombyx (Table 7-1). Most of these species showed a clear increase until spring 2005 and slightly decreased again after. Hesionura elongata and Nephtys spp. increased over the whole time period. For the interstitial species, Hesionura elongata and Polygordius appendiculatus, density only showed an increase beginning 1 year after the cessation of dredging. Some other interstitial species, like Pisione remota, Microphthalmus spp. and other species that prefer coarser sediment (e.g. Phoronis pallida), were also present and increasing since 2004. The density increase of Spiophanes bombyx and Urothoe brevicornis was only found in the first two years (Figure 7-7). 90

The proportion between the total amount of adult and juvenile individuals of Nephtys cirrosa slightly changed from 2003 to 2006 in favour of the adults (Figure 7-8). In 2007, again a higher number of juvenile species were found. The main feeding types were predators and deposit feeders. The percentage of omnivore/predators clearly increased over the whole sampling period, taking up more than 80 % of total density in 2007 (Figure 7-9). Av ind / m² 100%

800 700

80%

600 500

Polychaeta Crustacea

60%

Mollusca

400

Echinodermata

40%

300 200

Other groups

20%

100 0

0%

S03 Su03 A03 S04 A04 S05 A05 S06 A06 S07 A07

S03 Su03 A03 S04 A04 S05 A05 S06 A06 S07 A07

Figure 7-6: Average number of individuals per m² (+ st. dev.) with indication of the proportion of the major taxonomic groups. Table 7-1: Table showing the 6 most abundant species per sampling period, indicating the percentage of total density they represent. March 2003

June 2003

September 2003

Nephtys spp.

21,5

Nephtys spp.

16,4

Nephtys spp.

19,1

Urothoe brevicornis

14,8

Spiophanes bombyx

15,1

Urothoe brevicornis

15,7

Bathyporeia elegans/pelagica

8,1

Bathyporeia elegans/pelagica

10,5

Spiophanes bombyx

13,8

Nephtys cirrosa

7,4

Scoloplos armiger

10,5

Nephtys cirrosa

10,5

Ophelia limacina

6,0

Nephtys cirrosa

7,9

Microphthalmus spp.

7,2

Pariambus typicus

6,0

Spio spp.

6,4

Hesionura elongata

6,1

March 2006

September 2006

Nephtys spp.

24,2

Hesionura elongata

Hesionura elongata

21,8

Atylus swammerdami

Nephtys cirrosa

12,0

March 2007 23,9

September 2007

Hesionura elongata

33,8

Hesionura elongata

40,7

8,3

Nephtys spp.

18,0

Nephtys spp.

15,0

Nephtys spp.

6,5

Nephtys cirrosa

8,2

Microphthalmus spp.

10,1

Polygordius appendiculatus

5,6

Nephtys cirrosa

6,1

Microphthalmus spp.

4,3

Nephtys cirrosa

5,1

Ophelia limacina

4,2

Microphthalmus spp.

5,3

Glycera spp.

4,1

Glycera spp.

3,1

Urothoe brevicornis

4,0

spio spp.

4,7

Urothoe brevicornis

4,9

Glycera lapidum

3,0

March 2006

September 2006

March 2007 23,9

September 2007

Nephtys spp.

24,2

Hesionura elongata

Hesionura elongata

33,8

Hesionura elongata

40,7

Hesionura elongata

21,8

Atylus swammerdami

8,3

Nephtys spp.

18,0

Nephtys spp.

15,0

Nephtys cirrosa

12,0

Nephtys spp.

6,5

Nephtys cirrosa

8,2

Microphthalmus spp.

10,1

Polygordius appendiculatus

5,6

Nephtys cirrosa

6,1

Microphthalmus spp.

4,3

Nephtys cirrosa

5,1

Ophelia limacina

4,2

Microphthalmus spp.

5,3

Glycera spp.

4,1

Glycera spp.

3,1

Urothoe brevicornis

4,0

spio spp.

4,7

Urothoe brevicornis

4,9

Glycera lapidum

3,0

91

Number of individuals counted

518

300 250 200 150 100 50 0 S03

Su03

A03

Hesionura elongata Urothoe brevicornis

S04

A04

S05

A05

S06

Nephtys cirrosa Polygordius appendiculatus

A06

S07

A07

Nephtys spp. Spiophanes bombyx

Figure 7-7: Average number of ind./m² for the 6 most important species for the period Spring 2003 – Autumn 2007. Adults

Juveniles

100% 80% 60% 40% 20% 0% S03 Su03 A03

S04

A04

S05 A05

S06

A06

S07

A07

Figure 7-8: The different proportions of juvenile and adult Nephtys cirrosa. 100% 90% 80% 70% 60%

suspension feeder

50%

omnivoor/predator interface feeder

40%

deposit feeder

30% 20% 10% 0% S03 Su03 A03 S04 A04 S05 A05 S06 A06 S07 A07

Figure 7-9: Proportion in density of the different feeding types

7.3.5. Macrobenthos - Biomass Average biomass increased from March 2003 to September 2004 with again lower values in 2005 and 2006, although when excluding the biomass of E. cordatum and Ensis arcuatus, an increase in biomass was found (Figure 7-10). The total biomass of polychaetes, crustaceans and molluscs increased, although the average individual body size (biomass/number of individuals) per taxonomic 92

group did not show a similar trend: the highest value was recorded in March 2003 for polychaetes. Crustaceans and molluscs did have a higher individual body size in 2005 and 2006 (Figure 7-11). Average biomass (gAFDW / m²)

Average biomass (gAFDW / m²)

6

1,0

5

0,8 Polychaeta Crustacea 0,6 Mollusca 0,4 other groups 0,2 Echinodermata

4 3 2 1 0

Polychaeta Crustacea Mollusca other groups Echinodermata

0,0

M03 J03 S03 M04 S04 M05 S05 M06 S06

M03 J03 S03 M04 S04 M05 S05 M06 S06

Figure 7-10: Average biomass / m² (wet weight corrected to AFDW) for the different sampling periods. Left: all species included; right: excluding Echinocardium cordatum and Ensis arcuatus Polychaeta

Mollusca 0,3500

0,0050

0,3000 0,2500

0,0040

0,2000 0,0030 0,1500 0,0020

0,1000

0,0010

Average bodysize mollusca

Average bodysize crustacea and polychaeta

Crustacea 0,0060

0,0500

0,0000

0,0000 M03

J03

S03

M04

S04

M05

S05

M06

S06

Figure 7-11: Average body size of the 3 major taxonomic groups.

7.3.6. Community analysis Correspondence Analysis, using data from all six stations, indicated that station ZG08 had a different faunal pattern, possibly caused by the different sediment composition (higher proportion of coarse material). Also stations ZG05 and ZG06 had a slightly different species composition than the other stations. To have better insight into the temporal evolution in the community structure after the cessation of extraction, these 3 stations were analysed separately from the other 3 stations. For location ZG05, clear changes were found between the samples of 2003, the samples of 2004 and spring 2005 and the samples of autumn 2005. The first year was characterised by few species and a high dissimilarity between the samples. The samples taken in 2004 and spring 2005 were characterised by mainly interstitial, opportunistic and fast growing predator species. The similarity between the samples of autumn 2005 and later, was clearly higher than for the earlier seasons. The community structure of station ZG08 was clearly different in 2003, but samples from 2004 and 2005 clustered more or less together. The species that were more abundant in the samples of 2003 were mainly Ophiura albida, Urothoe brevicornis and Aonides paucibranchiata. The dissimilarity was also highest for the samples of 2003. The CA of the other 3 stations together indicated a shift in community structure, with the samples of 2003 and 2004 on the right side of axis 2 and the other years on the left side (Figure 7-12). The 2003 93

and 2004 samples were partly separated from the rest due to the higher abundance of Bathyporeia spp. (in 2003) and Urothoe brevicornis (in 2003 and 2004) and the higher abundance of interstitial species Polygordius appendiculatus, Microphthalmus spp. and Hesionura elongata and the juvenile Glycera spp. from 2005 on.

Figure 7-12: CA of the samples of stations ZG06, ZG07, ZG09 and ZG10 with indication of the different sampling years and associated species.

7.3.7. Comparison with other sampling locations Density, number of species and diversity of the six stations from the central depression were compared to three stations, equally located on the Kwintebank and to one station located on a nearby extractionfree sandbank. Only one month after the cessation of dredging, macrobenthic density, species number and diversity in the central depression were much lower than in station ZG01 and ZG04 (Figure 7-13). Since March 2004 values recorded in both the central depression and the other stations were comparable. ZG04 and ZG01 even showed a decrease in species number and diversity. Densities recorded in station ZG11, situated in the zone of currently very high extraction, were as low as the ones found in March 2003 in the central depression, but increased in autumn 2005 and after a decrease in 2006, high abundances were found in 2007 (Figure 7-13). The species composition in the central depression was best comparable to the community found in the north of the Kwintebank (Nephtys cirrosa, Hesionura elongata, Polygordius appendiculatus and Urothoe brevicornis). In the samples of ZG04 Nephtys cirrosa was one of the main species found, but Hesionura elongata was almost absent and Urothoe poseidonis is much more abundant than Urothoe brevicornis, which is the most abundant one in the central depression. CA indicated only a small similarity between the samples from the central depression and stations ZG01 and ZG04. For the period 2003-2005 station ZG06 showed a good similarity with station ZG11. Due to the increase of small polychaetes towards 2005, also several other stations from the central depression were found closer to the ZG11 samples in the CA. 94

Total biomass of the macrobenthos community in the central depression was smaller than the biomass recorded in the other locations on the Kwintebank, although the average biomass per individual was comparable. An example of the average individual biomass is given for the polychaetes in Figure 7-14. 1800 1600 1400

ZG05 ZG06

ind/m²

1200

ZG07 1000

ZG08 ZG09

800

ZG10 600 ZG01

400

ZG04 200

ZG11

0 Mar/03

Sep/03

Mar/04

Sep/04

Mar/05

Sep/05

Mar/06

Sep/06

Mar/07

Sep/07

30

Number of species

25

ZG05 ZG06

20

ZG07 ZG08

15

ZG09 ZG10

10 ZG01 5

ZG04 ZG11

0 Mar/03

Sep/03

Mar/04

Sep/04

Mar/05

Sep/05

Mar/06

Sep/06

Mar/07

Sep/07

4,0 3,5 ZG05

Diversity

3,0

ZG06

2,5

ZG07 ZG08

2,0

ZG09

1,5

ZG10

1,0 ZG01 0,5

ZG04 ZG11

0,0 Mar/03

Sep/03

Mar/04

Sep/04

Mar/05

Sep/05

Mar/06

Sep/06

Mar/07

Sep/07

Figure 7-13: Density, diversity and number of species for different locations on the Kwintebank for the period spring 2003autumn 2007.

95

gAFDW 0,010 ZG04

0,008

ZG05-10

0,006

ZG11

0,004

ZG01

0,002 0,000 M03

S03

M04

S04

M05

S05

M06

S06

Figure 7-14: Average individual biomass (gAFDW) of the polychaetes for the different locations on the Kwintebank.

7.4. Discussion Samples taken in the first three years after the closing of the central depression showed an increase in macrobenthos density, species number and diversity. For most stations in the central depression, the lowest density was measured directly after the cessation of dredging. After a few months, a higher density was found for all stations. Almost 3 years after the closing of the area, the average density had clearly increased and reached a level comparable to or higher than other locations on the Kwintebank. The species responsible for this general increase in abundance were small interstitial and fast growing predatory species. These species were responsible for 50 % of total density throughout the first 3 years of the study period. These species were already found in the early stages of extraction in this area (Waeterschoot, 1980; Meheus, 1981; Vanosmael et al., 1982; Vanosmael and Heip, 1986; Moulaert, 2007) and are the prominent species on other sandbanks of the BPNS (De Maersschalck et al., 2006; Moulaert et al., 2007). Similar results (the early settlement of species that are typical for the area and the absence of species belonging to families that were of low abundance or rare before dredging) were also found in other studies (Kenny & Rees, 1996; Sanchez-Moyano et al., 2000; Simonini et al., 2005). Long living species like bivalves were not recorded in large numbers in the samples of this study, in contrast to van Dalfsen et al. (2000), who found that bivalve recruitment was favoured at several North Sea extraction sites. It is known that bivalves are not highly abundant on the offshore sandbanks of the BPNS (Van Hoey et al., 2004; De Maersschalck et al., 2006; Moulaert, 2007). Even in the early years of extraction on the Kwintebank, only small numbers of bivalves (mainly Spisula spp.) were recorded (Waeterschoot, 1980; Meheus, 1981; Vanosmael et al., 1982; Vanosmael and Heip, 1986). The proportion between the total amount of adult and juvenile individuals of Nephtys cirrosa changed slightly, first in favour of the adults, whereas in the last year of sampling the proportion of juveniles re-increased. This indicates that the community re-established a state, where juveniles can reach maturity (increase in adults) and are able again to reproduce (increase in juveniles). The main feeding types were predators and deposit feeders. The percentage of omnivore/predators clearly increased over the whole sampling period, taking up more than 80 % of total density in 2007. Results from the Correspondence Analyses showed some differences between the samples taken in the first year after cessation of extraction and samples taken in the second and third year. Samples taken in the first year after extraction showed a slightly greater variability. Warwick & Clarke (1993) suggested that a higher variability may be due to perturbations. As such the smaller variability in year 2 and 3 of the study period might be another sign of improvement. A rapid increase in density and species number was found in several studies, although these mainly concerned the recovery after a short term impact of sand extraction (Pagliai et al., 1985; Sarda et al., 2000; Guerra-Garcia et al., 2003; Sanchez-Moyano et al., 2004; Simonini et al., 2005). In contrast to 96

the density and number of species, the biomass and the structure of the community were in most of these studies not yet restored by the end of the sampling period. In some other studies, the ‘recovery’ of an extraction area took several years (Kenny & Rees, 1994, 1996; Kenny et al., 1998; Desprez, 2000; Boyd et al., 2004; van Dalfsen et al., 2000; Cooper et al., 2005; Boyd et al., 2005). Many of these studies concern the effects of dredging operations that lasted only a few months or less. These studies generally have a good BACI (Before/After/Control/Impact) approach that give clear results. A rapid increase of certain species is usually found within one or several months, but a general conclusion is often that the recovery is not fully completed within the time span of the study. A good example is the experimental study off Norfolk, UK that continued for a couple of years and where different conclusions were drawn after each study period (Kenny & Rees, 1994, 1996; Kenny et al., 1998). Few studies have addressed the consequences of long-term dredging operations on the recolonisation of the macrobenthos (Desprez, 2000; Boyd et al., 2004; Newell et al., 2004a; Newell et al., 2004b; Boyd et al., 2005; Cooper et al., 2005). The recovery in areas of prolonged continuous impact (>10 years) is more difficult to study compared to areas with only a short term impact as in a lot of these studies, base line data is not available. It can be that the community at the long-term impact site has evolved to a new structure that is adapted to regular disturbances. These sites may not be totally defaunated by the end of the extraction period, in contrast to the short term impact sites. Another problem is the difficulty to find a good reference area, as the area for sand extraction covers the whole Kwintebank. Furthermore, it is an area with a highly diverse character. The interpretation of our data was clearly hampered by the lack of knowledge of the pre-dredged status of the area and information on the pure natural variation of the area. Data from other locations on the Kwintebank were used as reference locations, but as these other stations were also located in a sand extraction area, their seasonal pattern is probably not only naturally caused. As such these stations cannot be used as a reference area according to the definition by Boyd et al. (2003): 'a good reference area should be identical in all respects to the extracted sites, save for the impact of extraction activities'. This reference area should not be representative of the pre-dredged status, as a community does not necessarily have to lead to a situation similar to the one that existed before the disturbance. The environmental conditions on the impact site may have changed irreversibly, leading to a new macrobenthic community which is also stable, but no longer comparable to the pre-dredged status. Kenny et al. (1998) en van Dalfsen et al. (2000) found changes in community composition both in the disturbed and undisturbed areas in the North Sea that were similar to each other, but dissimilar to the original community. When assessing 'recovery' rate, it is important to draw a distinction between (1) 're-colonisation', which is the settlement of new recruits from the plankton or immigration of adults from outside the area, and (2) 'restoration', which can be considered as the re-establishment of the community structure (Boyd et al., 2003). It is not clear whether the Kwintebank central depression has reached a full 'recovery', in the sense of restoration, after 3 years, but the process of 're-colonisation' of the area has clearly started. Density, diversity and species number have reached a level that is higher than other locations on the Kwintebank, but it is not clear whether these levels are equal to the pre-extraction level as no base line records from the area are available. Total biomass values have only slightly increased and are still smaller compared to other areas on the Kwintebank, but the average individual biomass was stable and comparable to the other stations on the Kwintebank and with stations on other sandbanks on the BPNS. Wilson (1998) defined 're-colonisation' as a temporal change in biological variables following a perturbation, and 'recovery' as a lack of difference in the temporal change of biological variables at impact sites relative to reference sites. In other words, 'recovery' is thought to be complete when temporal trends in the benthos of the impact site run parallel with those at the 97

reference sites. According to these definitions, 're-colonisation' did occur, as an increase of the biological variables was found for the central depression, but it can be assumed that 'recovery' has not been fulfilled. Although the temporal trends run parallel with those found for reference station ZG11, it has to be reminded that this station is located in a highly extracted area. For stations ZG01 and ZG04, which are under less influence of the extraction activities, the temporal trend was different from the central depression. Also for the Thorntonbank and Goote bank the seasonal variation was different from the study area (De Maersschalck et al., 2006). By reviewing the different studies on the recovery of the macrobenthos after dredging, it is clear that recovery is site specific, mainly depending on the habitat type of the dredged area, as well as on the scale and duration of disturbance, hydrodynamics, etc. (for review see Newell et al., 1998). In a study on the long term effects of extraction on the BPNS, a fining of the sediment was found for the central part of the Kwintebank from samples of 1980-1984 compared to 2003 (Moulaert, 2007). The sediment sampled in the area after cessation of dredging showed an increase of the median grain size for most of the sampling locations, although a high variability between locations and sampling periods was found. This large variability in the sediment characteristics are probably a combination of the dynamic character of the sandbank as well as the uneven impact of the extraction. In this highly dynamic area, it is not surprising that a relatively rapid re-colonisation occurs. Highly disturbed sediments in dynamic coastal areas and estuaries, which are dominated mainly by opportunistic, rapidly colonising, fast-growing species, have a rapid rate of recovery. The recovery time increases in more stable habitats that are dominated by long-lived components with complex biological interactions controlling community structure (Newell et al., 1998). The recovery of the macrobenthos also depends on the comparability of the extracted area with the surrounding area. If the habitat type of the impact area and the neighbouring areas is similar, a short distance, rapid migration may occur. Results from the June 2003 samples indicate an increase in the density for some species and species richness, probably due to the migration from areas close by.

7.5. Conclusions and recommendations Within this project, we aimed at assessing the possible biological recovery of the central depression on the Kwintebank after its closure for extraction activities in February 2003. In general, it can be concluded that 're-colonisation' of the central depression has occurred. The poor macrobenthic community that was found directly after the cessation of extraction, evolved in 2-3 years to a community with higher densities, number of species and diversity. Moreover, the macrobenthic community was characterised by small interstitial and mobile species, which is typical for other sandbanks on the Belgian Part of the North Sea with a comparable sediment composition. So far, it remains unclear whether this situation is stable and whether 'recovery' s.s. of the macrobenthos of the central depression has been fulfilled, due to the lack of baseline data and the practical problems related to defining a suitable reference area. As such, the observed positive trend in the different macrobenthic variables could besides being related to the cessation of dredging, as well be partly related to natural variability. We recommend that the size of the area of impact should be so that the impacted area and the distance for potential re-colonisation from unimpacted areas are small. On the other hand, the intensity of dredging per unit area increases with decreasing size and higher dredging intensities may eliminate a higher number of species (Boyd & Rees 2003). Therefore, an optimum size of extraction area should be found to reach equilibrium between the above mentioned factors.

98

7.6. Acknowledgements This study was part of the project SPEEK (EV/38) funded by the Belgian Science Policy. We would like to thank the different partners within the project and the FPS Economy, cell Continental Shelf for the cooperation and the multibeam images of the Kwintebank.

99

8. GENERAL CONCLUSIONS AND RECOMMENDATIONS 8.1. Conclusion In the Royal Decree of 1 September 2004, 3 zones were (re)defined as exploitation zones and one zone as exploration zone. The exploitation zones are located on the Thorntonbank and Gootebank (Zone 1 a & b), Kwintebank, Buitenratel and Oostdyck (Zone 2 a, b & c) and Sierra Ventana (Zone 3 a & b), while the exploration zone (Zone 4) is located on the Hinderbanken. To evaluate the biological value of the exploitation and exploration areas on the BPNS, data were collected on three ecosystem components: organisms living in the sediment (macrobenthos) and organisms living on or in close association with the seafloor (epibenthos and demersal fish). Next to the extraction and exploration zones, several corresponding reference areas were sampled. On a regional scale (BPNS), the coastal-off shore gradient is the dominant structuring factor for all three ecosystem components. The macrobenthos communities of the BPNS were strongly related to the sediment type and the coastal-offshore gradient, with the Abra alba and Macoma balthica community situated in the coastal area, whereas the Nephtys cirrosa and Ophelia limacina community were dominant offshore (Van Hoey et al., 2004). The epifauna and demersal fish communities of the BPNS were structured by salinity, sedimentology, depth and temperature, which results in 3 more coastal community types and 2 offshore types (Vandendriessche et al., in prep). There is a strong similarity in the structure and distribution of the macro- and epi and fish communities. The fauna in extraction zone 1 and exploration zone 4 was comparable with the typical off shore communities. Extraction zone 3 was more closely associated with the coastal zone community. Within zone 2, the coastal-offshore gradient manifested itself, with zone 2 a + b closer related to the coastal communities and 2 c to the off shore community. In general, densities were higher in the coastal area for all three ecosystem components compared to off shore samples. As the 4 extraction/exploitation zones are situated on different sandbank systems, the differences in density, biomass, species richness and diversity for the three ecosystem components between the different sandbank systems were primarily due to topography, sediment composition and distance from the coastline. Therefore, it is difficult to use data from one sandbank system and the according extraction zone as reference for the other extraction zones. Consequently, it is recommended to allocate reference zones, free from extraction activity, for each sandbank system in the integrated spatial management of the BPNS. On the local scale (extraction zone), the importance of the structuring factors is different for the different ecosystem components. For epibenthos and demersal fish, seasonal and interannual effects primarily determined the community structure, while topography (gully-bank position) of the samples was of secondary importance. The community structure of the macrobenthos was primarily structured by topography on the local scale, and seasonal and interannual variation was of lower importance. Therefore, it is important that the sampling for epibenthos and demersal fish is temporally elaborated, and is carried out simultaneously within one extraction zone and the adjoining reference stations to exclude temporal variability. For macrobenthos, sampling effort should focus on a high spatial coverage of the extraction zone and the adjoining reference stations, while temporal monitoring could be reduced to once per year, preferably in the autumn after the recruitment period, when populations are at their maximum. Nevertheless, macrobenthos density does show temporal variation, and therefore simultaneous sampling of impact and reference stations is important in impact assessment. Sand extraction activities in the BPNS are generally quite extensive, except in zone 2, and the analyses indicated no dramatic impact of sand extraction on macrobenthos, epibenthos and demersal fish. However, from the detailed analyses of the Kwintebank, it was concluded that sand extraction caused 100

a reduction in species number and densities for macrobenthos. This impact was limited in time and space, because after cessation of the activities, rapid colonisation by small, opportunistic and mobile species was observed. For demersal fish fauna, impact was limited to a slight reduction in densities and an increased dominance of fishes, while for epibenthos, density, biomass and diversity appeared to be higher in the extraction zone, possibly due to an increased supply of dead organic matter that becomes available for epibenthic scavengers, which are found in higher numbers in the impact zones. There is, however, no certainty that this was solely caused by the effect of sand extraction, since natural variability is high in this area, which again supports the need for representative reference areas to exclude natural variability as much as possible. The results on macrobenthos further indicate that it is not necessary to rotate the extraction activities in the different areas, but that it is important to carefully monitor in order to be able to detect large changes timely and to take mitigating measures, such as the temporal closure of certain areas in order to fully restore the biological value of these impacted areas. A good example is the closure of the central depression on the Kwintebank between 2003 and 2009 and the subsequent recovery of the benthic system. Although sand extraction in the other zones showed no significant impact, future intensification of extraction activities in these zones, to levels similar or even higher than the current activities on the Kwintebank area, may result in long-term changes following the removal of benthic fauna and changes in sediment characteristics and the topography of the seabed. The response of the ecosystem components to the upcoming extraction activities in the Hinderbanken area, for example, may be completely different than in the Kwintebank area, as it harbours a different species community and even some species unique for the BPNS. So, when extraction starts in the Hinderbanken, this should be closely investigated. The currently exerted baseline studies provide a good T0 status for future impact assessments and allow for the evaluation and adaptation of the current monitoring strategy. There clearly is a need for control areas within most extraction zones. Furthermore, it would be ideal to receive the blackbox data at regular time intervals (e.g. 4 times/year) to be able to concentrate the monitoring sampling schemes on the most intensively extracted zones and to examine recovery in the recently abandoned extraction areas. In the light of commercially exploited species (sole, plaice, dab, shrimp) and the preservation of biodiversity occurring in the extraction and exploration zones, it is therefore imperative to keep all three ecosystem components included in the existing monitoring program. For optimal results, however, some changes in the selection of stations should be considered based on the discussed analyses.

8.2. Recommendations. In general, the most important results that will be incorporated, continued or adapted in the future monitoring: 

The extraction is mainly concentrated on top of the sand banks, therefore we will concentrate the impact sampling in these areas with transects following the direction of the sand bank ridge. Gully samples for macrobenthos will be reduced as the impact is not directed towards these gullies.



Accurate reference stations on the same sand bank system as the extracted area are needed. Extra reference stations for the different extraction zones will be allocated in 2010.



Simultaneous sampling of impact and reference samples has been repeatedly addressed, as it is imperative in impact assessment.

101



Adaptation of the sampling design and number of samples to make impact assessment with benthic indicator tools possible (e.g. Benthic Ecosystem Quality Index; BEQI).

This results in the following approach for the impact assessment and monitoring in each area: Zone 1: Macrobenthos: Sand extraction on the Thorntonbank is extensive and no significant impact on the macrobenthos was observed. The impact of extensive extraction on the Thorntonbank continues to be monitored and the analyses showed that the reference stations currently used on the Bank Zonder Naam, Goote Bank and on the Thorntonbank outside the extraction zone were good references provided the same sieving method is used for impact and control samples. Additionally, as of September 2009, a reference area was designated in the extraction area, which is an ideal reference zone located on the same sandbank. Epibenthos and demersal fish: The extensive extraction did not result in significant changes in epibenthos or demersal fish fauna. The current monitoring on the Thorntonbank is sufficient and the chosen references are good proxies for the impact stations. When extraction on the Goote Bank would start, at least one extra fish tracks in the impact area in the gully should be assigned. Zone 2: Macrobenthos: Since this is the most intensively used sand extraction area, and a heterogeneous area due to the natural coastal-off shore gradient, additional reference stations on each sand bank system should be sampled. As of 2010, extra reference stations outside the extraction zone will be sampled on the Oostdyck, Buitenratel and Kwintebank simultaneously with the impact samples. The follow-up of the recovery of the benthos in the closed areas on the Kwintebank will be continued. Epibenthos and demersal fish: Sand extraction did not have a negative impact on the epibenthos and fish community. However, it was difficult to draw reliable conclusions because of the natural gradient present in this extraction zone. Therefore, it is recommended to include additional reference stations on the same sand bank system as the impact stations (i.e. extra reference station on Oostdyck and Kwintebank) and simultaneous sampling is imperative to exclude temporal variability. These changes will be implemented in 2010. Zone 3: Currently, zone 3b is in use as dumping site and therefore not used for extraction. Consequently, no samples in the framework of sand extraction monitoring are taken. For biological information of this area the reader is referred to the report on disposal of dredged material by Van Hoey et al. (2010). Since 2009, 4 stations are sampled in zone 3a for macrobenthos. Zone 4: Macrobenthos: From the baseline study, it could be concluded that the gullies are the richest areas (highest density, diversity and species richness) and therefore extraction should be avoided there. In autumn 2009, two potential extraction areas were sampled and simultaneously, reference samples were taken on the Bligh Bank. Ideally, a small area on the Hinderbanken (northern part of the sandbanks) should be demarcated as reference area. A BACI design monitoring will be executed from 2010 onwards in the proposed and designated extraction areas in zone 4. Epibenthos and demersal fish: Temporal variation was important as structuring factor for the epibenthos and fish communities, emphasising the need for simultaneous sampling of impact and reference stations. Furthermore, also for epibenthos and fish the gullies were observed as the richest areas. In autumn 2009, these two ecosystem components were included in the BACI design.

102

9. REFERENCES ALSF (2005) MEPF 04/04 Coupling Physical and Ecological Models: A New Approach to Predicting the Impacts of Aggregate Extraction on Biological Recoverability. Report 2004-5 Birklund J & Wijsman JWM (2005) Aggregate extraction: a review on the effect on ecological functions. DHI Water and Environment and WL/Delft Hydraulics, report Z3297.10. 56pp Boyd SE, Rees HL (2003) En examination of the spatial scale of impact on the marine benthos arising from marine aggregate extraction in the central English Channel. Est Coast Shelf Sci 57: 1-16 Boyd SE, Limpenny DS, Rees HL, Cooper KM, Campbell S (2003) Preliminary observations of the effects of dredging intensity on the re-colonisation of dredged sediments off the southeast coast of England (Area 222). Est Coast Shelf Sci 57: 209-223 Boyd SE, Cooper KM, Limpenny DS, Kilbride R, Rees HL, Dearnaley MP, Stevenson J, Meadows WJ, Morris CD (2004) Assessment of the rehabilitation of the seabed following marine aggregate dredging. Sci Ser Tech Rep 121, CEFAS Lowestoft, 154pp Boyd SE, Limpenny DS, Rees HL, Cooper KM (2005) The effects of marine sand and gravel extraction on the macrobenthos at a commercial dredging site (results 6 years post-dredging). ICES J Mar Sci, 62: 145-162 Brey T (2001) Population dynamics in benthic invertebrates. A virtual handbook. Version 01.2. Alfred Wegener Institute for Polar and Marine Research, Germany. http://www.awibremerhaven.de/Benthic/Ecosystem/FoodWeb/Handbook/main.html Budd GC (2005) Nephtys cirrosa. White catworm. Marine Life Information Network: Biology and Sensitivity Key Information Sub-programme [on-line]. Plymouth: Mar Biol Ass UK http://www.marlin.ac.uk/species/Nephtyscirrosa.htm Cooper KM, Eggleton JD, Vize SJ, Vanstaen K, Smith R, Boyd SE, Ware S, Morris CD, Curtis M, Limpenny DS, Meadows WJ (2005) Assessment of the rehabilitation of the seabed following marine aggregate dredging-part II. Sci Ser Tech Rep 130, CEFAS Lowestoft, 82pp Degraer S, Vincx M; Meire P, Offringa H (1999) The macrozoobenthos of an important wintering area of the common scoter (Melanitta nigra). .J. Mar. Biol. Ass. UK 79(2):243-251 De Groot SJ (1979) An assessment of the potential environmental impact of large-scale sand-dredging for the building of artificial islands in the North Sea. Ocean Management 5: 211-232 De Maersschalck V, Hostens K, Wittoeck J, Cooreman K, Vincx M, Degraer S (2006) Monitoring van de effecten van het Thornton windmolenpark op de benthische macro-invertebraten en de visfauna van zachte substraten - Referentietoestand. Rapport ILVO-Visserij/Monitoring/2006-01 Degrendele K, Roche M, Schotte P (2005) Monitoring van de impact van de extracties. in: 2005. Balans van het duurzaam beheer van zand en grindexploitatie op het Belgische Continentaal Plat en toekomstperspectieven. Studiedag 10 Oktober 2005 Degrendele K, Roche M, Schotte P, Van Lancker V, Bellec V, Bonne W (in press) Morphological evolution of the Kwinte Bank central depression before and after cessation of aggregate extraction. J Coast Res

103

Desprez M (2000) Physical and biological impact of marine aggregate extraction along the French coast of the Eastern English Channel: short- and long-term post-dredging restoration. ICES J Mar Sci 57: 1428-1438 Guerra-Garcia JM, Corzo J, Garcia-Gomez JC (2003) Short-term benthic re-colonisation after dredging in the harbour of Ceuta, North Africa. P.S.Z.N.: Mar Ecol 24(3): 217-229 Hacking N (2003) A review of the ecology of offshore ocean sediments with particular reference to marine aggregate resources for beach nourishment in New South Wales. Centre for Natural Resources. Department of Infrastructure, Planning and Natural Resources Hiscock K, Langmead O, Warwick R, Smith A (2005) Identification of seabed indicator species to support implementation of the EU Habitats and Water Framework Directives. Second edition. Report to the Joint Nature Conservation Committee and the Environment Agency from the Marine Biological Association. Plymouth: Marine Biological Association. JNCC Contract F90-01-705 77 pp James FC, McCulloch CE (1990) Multivariate analysis in ecology and systematics: panacea or Pandora's box? Ann Rev Ecol Syst 21: 129-166 James RJ, Lincoln Smith MP, Fairweather PG (1995) Sieve mesh-size and taxonomic resolution needed to describe natural spatial variation of marine macrofauna. Mar Ecol Prog Ser 118: 187198 Jones AR (1986) The effects of dredging an spoil disposal on macrobenthos, Hawkesbury Estuary, NSW. Mar Poll Bull 17: 17-20 Jongman RHG, Ter Braak CJF, Van Tongeren OFR (1995) Data analysis in community and landscape ecology. Camebridge University Press, 299 pp Josefson AB Forbes TL, Rosenberg R (2002) Fate of phytodetritus in marine sediments: functional importance of macrofaunal community. Mar Ecol Prog Ser 230: 71-85 Kenny AJ, Rees HL (1994) The effects of marine gravel extraction on the macrobenthos: early post dredging re-colonisation. Mar Poll Bull 28: 442-447 Kenny AJ, Rees HL (1996) The effects of marine gravel extraction on the macrobenthos: results 2 years post-dredging. Mar Poll Bull 32: 615-622 Kenny AJ, Rees HL, Greening J, Campbell S (1998) The effects of marine gravel extraction on the macrobenthos at an experimental dredge site off north Norfolk, U.K. (Results 3 years postdredging). ICES CM 1998/V:14, 1–8 Kröncke I, Bergfeld C (2001) Review of the current knowledge on North Sea benthos. Berichte aus dem Zentrum für Meeres- und Klimaforschung Z 12: 112 pp MACROBEL Long term trends in the macrobenthos of the Belgian Continental Shelf. Research action SPSD-I. MN/DD2/013 http://www.vliz.be/Vmdcdata/macrobel/index.php McCune B, Mefford MJ (1999) Multivariate Analysis of Ecological Data Version 4.25 MjM Software, Gleneden Beach, Oregon, U.S.A. Meheus L, (1981) Vergelijkende studie van het macrobenthos van enkele zandbanken in de zuidelijke bocht van de Noordzee. Lic thesis, University of Gent, Belgium 104

Moulaert I (2007) Evaluation of the consequences of long term sand extraction on the structural characteristics of the macrobenthos communities. In: Vanaverbeke J, Bellec V, Bonne W, Deprez T, Hostens K, Moulaert I, Van Lancker V & Vincx M (2007) Study of post-extraction ecological effects in the Kwintebank sand dredging area (SPEEK): final report. Belgian Science Policy: Brussels, Belgium. 92pp Moulaert I, Hostens K, Hillewaert H, Wittoeck J (submitted) Spatial variation of the macrobenthos species and communities of the Belgian Continental Shelf and the relation to environmental variation. Journal of Sea Research. Newell RC, Seiderer LJ, Hitchcock DR (1998) The impact of dredging works in coastal waters: a review of the sensitivity to disturbance and subsequent recovery of biological resources on the sea bed. Oceanogr Mar Biol 36: 127-178 Newell RC, Seiderer LJ, Robinson JE, Simpson NM, Pearce B, Reeds KA (2004a) Impacts of overboard screening on seabed and associated benthic biological community structure in relation to marine aggregate extraction. Tech Rep to ODPM and MIRO. Project No SAMP.1.022. MES Ltd, St. Ives, Cornwall. 152 pp Newell RC, Seiderer LJ, Simpson NM, Robinson JE (2004b) Impacts of marine aggregate dredging on benthic macrofauna of the south coast of the U.K. J Coast Res 20: 115-125 Pagliai AMB, Varriale AMC, Crema R, Galletti MC, Zunarelli RV (1985) Environmental impact of extensive dredging in a coastal marine area. Mar Poll Bull 16(12): 483-488 Poiner R, Kennedy R (1984) Complex patterns of change in the macrobenthos of a large sandbank following dredging. Marine Biology 78: 335-352 Quinn G and Keough M (2002) Experimental design and Data analysis for Biologists. Cambridge 561 University press, Cambridge, United Kingdom, pp. 302-338 Robinson JE, Newell RC, Seiderer LJ, Simpson NM (2005) Impacts of aggregate dredging on sediment composition and associated benthic fauna at an offshore dredge site in the southern North Sea. Mar Env Res 60: 51-68 Sanchez-Moyano JE, Estacio FJ, Garcia-Adiego EM, Garcia-Gomez JC (2004) Dredging impact on the benthic community of an unaltered inlet in southern Spain. Helg Mar Res 58: 32-39 Sardá R, Pinedo S, Gremare A, Taboada S (2000) Changes in the dynamics of shallow sandy-bottom assemblages due to sand extraction in the Catalan Western Mediterranean Sea. ICES J Mar Sci 57: 1446-1453 Simonini R, Ansaloni I, Bonvicini Pagliai AM, Cavallini F, Iotti M, Mauri M, Montanari G, Preti M, Rinaldi A, Prevedelli D (2005) The effects of sand extraction on the macrobenthos of a relict sands area (northern Adriatic Sea): results 12 months post-extraction. Mar Poll Bull 50: 768-777 Snelgrove P, Butman C (1994) Animal-sediment relationships revisited: cause versus effect. Oceanogr Mar Biol Ann Rev 32: 111-117 Vanaverbeke J, Bellec V, Bonne W, Deprez T, Hostens K, Moulaert I, Van Lancker V, Vincx M (2007) Study of post-extraction ecological effects in the Kwintebank sand dredging area (SPEEK): final report. Belgian Science Policy: Brussels, Belgium. 92pp

105

van Dalfsen J, Essink K, Toxvig Madsen H, Birklund J, Romero J, Manzanera M (2000) Differential response of macrozoobenthos to marine sand extraction in the North Sea and the Western Mediterranean. ICES J Mar Sci 57: 1439-1445 van Dalfsen J, Essink K (2001) Benthic community response to sand dredging and shore face nourishment in Dutch coastal waters. Senck marit 31: 329-332 Vandendriessche S, Hostens K and Wittoeck J (2009) Monitoring of the effects of the Thorntonbank and Bligh Bank windmill parks on the epifaunca and demersal fish fauna of soft-bottom sediments: Thorntonbank: status during construction (T1), Bligh Bank: reference condition (T0): 93-150 in Degraer S and Brabant R (Eds.) (2009) Offshore windfarms in the Belgian Part of the North Sea: State of the art after two years of monitoring. Royal Belgian Institute for Natural Sciences, Management Unit of the North Sea Mathematical Models. Marine ecosystem management unit, pp 287 + annexes Vandendriessche S, De Backer A, Wittoeck J and Hostens K (In prep.) Variability within communities of demersal fish and epibenthos in the Belgian Part of the North Sea, and implication for impact monitoring. Van Hoey G, Degraer S, Vincx M (2004) Macrobenthic community structure of soft-bottom sediments at the Belgian Continental Shelf. Est Coast Shelf Sci 59: 601-615 Van Hoey G, Hostens K, Parmentier K, Robbens J, Bekaert K, De Backer A, Derweduwen J, Devriese L, Hillewaert H, Hoffman S, Pecceu E, Vandendriessche S and Wittoeck J (2010) Biological and chemical effects of the disposal of dredged material in the Belgian Part of the North Sea (Period 2007-2008). ILVO-report, pp 97 Van Moorsel GWNM (1994) The Klaverbank North Sea, geomorphology, macrobenthic ecology and the effect of gravel extraction. Report 94.24. Culemborg, The Netherlands, Bureau Waardenburg bv. 65 pp Vanosmael C, Willems K A, Claeys D, Vincx M, Heip C (1982) Macrobenthos of a sublittoral sandbank in the southern bight of the north sea. J Mar Biol Ass U.K. 62: 521-534 Vanosmael C, Heip C (1986) A comparative study of the macrobenthos of three sandbanks in the Belgian coastal waters in 1980-1984. Has sand exploitation an influence on the macrobenthos? In: Heip C, Coomans A (Ed) (1986) Ecology, ecotoxicology and systematics of marine benthos. 125144 Waeterschoot H (1980) Macrobenthos van de Kwintebank (1979-1980): studie in het kader van de zand- en grindexploitatie voor de Belgische kust. LicThesis University of Gent, Belgium Warwick RM, Clarke KR (1993) Increased variability as a symptom of stress in marine communities. J Exp Mar Biol Ecol 172: 215-226 Wilson G (1998) A Post-impact Monitoring Study of Benthic Fauna in Areas Dredged for the Third Parallel Runway in Botany Bay. Australian Museum Marine Invertebrates Section

106

ANNEX I: Species list of macrobenthos PHYLUM ANNELIDA Ordo Polychaeta Aonides oxycephala Aonides paucibranchiata Aricidea sp. Capitella sp. Capitellidae sp. Cirratulidae sp. Eteone flava Eteone foliosa Eteone longa Eumida Eunereis longissima Euzonus flabelligerus Exogone naidina Exogone hebes Glycera sp. Glycera alba Glycera lapidum Goniada maculata Harmothoe sp. Hesionura elongata Heteromastus filiformis Lanice conchilega Macrochaeta helgolandica Magelona filiformis Magelona johnstoni Malmgreniella castanea Malmgreniella glabra Mediomastus fragilis Magelona sp. Microphthalmus sp. Myrianida sp. Nephtys sp. Nephtys assimilis Nephtys caeca Nephtys cirrosa Nephtys hombergii Nephtys longosetosa Nereis sp. Notomastus latericeus Ophelia limacina Ophelia rathkei Owenia fusiformis Parougia eliasoni Pectinaria sp. 107

Phyllodoce sp. Phyllodoce groenlandica Phyllodoce lineata Phyllodoce rosea Phyllodoce maculata/mucosa Pisione remota Podarkeopsis sp. Poecilochaetus serpens Polycirrus sp. Polydora sp. Polygordius appendiculatus Pomatoceros sp. Protodorvillea kefersteini Protodriloides chaetifer Protodrilus sp. Psammodrilus balanoglossoides Pygospio elegans Sabellaria spinulosa Scolelepis squamata Scolelepis bonnieri Scoloplos armiger Sigalion mathildae Sphaerodorum gracilis Sphaerosyllis hystrix Spio sp. Spiophanes bombyx Sthenelais boa Streblospio benedicti Streptodonta pterochaeta Streptosyllis websteri Syllidae sp. Syllidia armata Syllis armillaris Syllis cornuta Syllis hyalina Travisia forbesii

Ordo Oligochaeta Oligochaeta sp.

PHYLUM ARTHROPODA Subphylum CRUSTACEA Ordo Amphipoda Abludomelita obtusata Ampelisca brevicornis Amphilochus neapolitanus Aora sp. Apherusa sp. 108

Atylus falcatus Atylus swammerdami Bathyporeia sp. Bathyporeia elegans Bathyporeia gracilis Bathyporeia guilliamsoniana Bathyporeia pelagica Bathyporeia sarsi Cheirocratus sundevalli Corophium sp. Gammarus sp. Jassa sp. Leucothoe incisa Leucothoe lilljeborgi Megaluropus agilis Microprotopus maculatus Pariambus typicus Perioculodes longimanus Pontocrates altamarinus Pontocrates arenarius Stenothoe marina Tryphosella sarsi Urothoe brevicornis Urothoe poseidonis

Ordo Cumacea Bodotria arenosa Bodotria scorpioides Cumopsis goodsir Diastylis bradyi Diastylis rathkei Pseudocuma sp. Pseudocuma longicorne Pseudocuma simile

Ordo Decapoda Callianassa sp. Diogenes pugilator Ebalia sp. Pagurus sp. Thia scutellata

Ordo Isopoda Eurydice sp.

Ordo Mysida Gastrosaccus spinifer

Ordo Nebaliacea Nebalia bipes

Ordo Tanaidacea Tanaidacea sp. Tanaissus lilljeborgi 109

PHYLUM CHORDATA Subphylum CEPHALOCHORDATA Branchiostoma lanceolatum

PHYLUM CNIDARIA Classis Anthozoa Actinaria sp.

PHYLUM ECHINODERMATA Classis Echinoidea Echinocardium cordatum Echinocyamus pusillus

Classis Ophiuroidea Acrocnida brachiata Amphipholis squamata Amphiura sp. Ophiocten affinis Ophiura sp. Ophiura albida Ophiura ophiura

PHYLUM MOLLUSCA Classis Bivalvia Abra alba Aequipecten opercularis Donax vittatus Ensis sp. Goodallia triangularis Kurtiella bidentata Lutraria lutraria Macoma balthica Mactra sp. Mya truncata Mytilus sp. Petricola pholadiformis Pholoe sp. Spisula sp. Spisula elliptica Spisula solida Spisula subtruncata Striarca lactea Tellimya ferruginosa Tellina pygmaeus Tellina fabula Venerupis senegalensis

Classis Gastropoda Caecum sp. Crepidula fornicata Epitonium clathrus Euspira sp. 110

Euspira catena Nassarius reticulatus Opisthobranchia sp.

PHYLUM PHORONIDA Phoronis sp.

111

ANNEX II: Species list of epibenthos PHYLUM ANNELIDA Ordo Polychaeta Aphrodite aculeata Arenicola marina Chaetopterus variopedatus Lanice conchilega Nephtys sp. Nereis sp. Ophelia limacina Owenia fusiformis Pectinaria koreni Travisia forbesii

PHYLUM ARTHROPODA Subphylum CRUSTACEA Ordo Cirripidia Balanus sp.

Ordo Decapoda Callianassa tyrrhena Cancer pagurus Carcinus maenas Corystes cassivelaunus Crangon allmanni Crangon crangon Diogenes pugilator Ebalia cranchii Eurynome aspera Hemigrapsus penicillatus Hyas coarctatus Liocarcinus navigator Liocarcinus depurator Liocarcinus holsatus Liocarcinus marmoreus Liocarcinus pusillus Liocarcinus vernalis Macropodia rostrata Necora puber Pagurus bernhardus Pagurus cuanensis Palaemon serratus Pandalus montagui Philocheras trispinosus Pinnotheres pisum Pisidia longicornis Portumnus latipes Thia scutellata Upogebia deltaura 112

Ordo Isopoda Idotea linearis

PHYLUM CHORDATA Subphylum Tunicata Ascidiacea sp.

PHYLUM CNIDARIA Classis Anthozoa Alcyonium digitatum Anthozoa sp. Tealia felina

PHYLUM ECHINODERMATA Classis Asteroidea Asterias rubens

Classis Echinoidea Echinocardium cordatum Echinocardium flavescens Psammechinus miliaris Spatangus purpureus

Classis Ophiuroidea Amphiura brachiata Ophiothrix fragilis Ophiura albida Ophiura ophiura

PHYLUM MOLLUSCA Classis Bivalvia Abra alba Aequipecten opercularis Cerastoderma edulis Chlamys varia Crassostrea gigas Diplodonta rotundata Donax vittatus Dosinia exoleta Ensis arcuatus Ensis directus Ensis ensis Glycymeris glycymeris Lutraria lutraria Macoma balthica Mactra stultorum Mya truncata Mytilus edulis Petricola pholadiformis Spisula elliptica Spisula solida Spisula subtruncata Striarca lactea Tellina fabula 113

Tellina tenuis Venerupis senegalensis

Classis Cephalopoda Loligo subulata Loligo vulgaris Sepia officinalis Sepiola atlantica

Classis Gastropoda Buccinum undatum Crepidula fornicata Epitonium clathrus Euspira catena Euspira pulchella Nassarius reticulatus Onchidoris bilamelata

PHYLUM PORIFERA Porifera sp.

PHYLUM SIPUNCULA Subphylum Sipunculida Phascolion strombi

114

ANNEX III: SPECIES LIST OF DEMERSAL FISHES Ordo and Species ANGUILLIFORMES

Dutch Name

English Name

Anguilla anguilla

paling

eel

schrapper

sand smelt

geep

garpike

gevlekte gladde haai gladde haai hondshaai

starry smooth-hound smooth-hound dogfish

fint haring ansjovis sprot

twaite shad herring anchovy sprat

vijfdradige meun noordse meun vierdradige meun kabeljauw driedradige meun wijting steenbolk dwergbolk leng

five bearded rockling northern rockling four beard rockling cod three bearded rockling whiting pouting poor cod ling

ATHERINIFORMES Atherina presbyter

BELONIFORMES Belone belone

CARCHARHINIFORMES Mustelus asterias Mustelus mustelus Scyliorhinus canicula

CLUPEIFORMES Alosa fallax Clupea harengus Engraulis encrasicolus Sprattus sprattus

GADIFORMES Ciliata mustela Ciliata septentrionalis Enchelyopus cimbrius Gadus morhua Gaidropsarus vulgaris Merlangius merlangus Trisopterus luscus Trisopterus minutus Molva molva

GASTEROSTEIFORMES Gasterosteus aculeatus

driedoornige stekelbaars three spined stickelback

OSMERIFORMES Osmerus eperlanus

spiering

European smelt

zandspiering glasgrondel pitvis rasterpitvis diklipharder zeebaars kleine pieterman zwarte grondel / smelt gevlekte lipvis

sandeel transparent goby dragonet reticulated dragonet mullet sea bass lesser weever black goby smooth sandeel great sandeel ballan wrasse

PERCIFORMES Ammodytes tobianus Aphia minuta Callionymus lyra Callionymus reticulatus Chelon labrosus Dicentrarchus labrax Echiichtys vipera Gobius niger Gymnammodytes semisquamatus Hyperoplus lanceolatus Labrus bergylta 115

Mullus surmuletus Parablennius gattorugine Pomatoschistus lozanoi Pomatoschistus microps Pomatoschistus minutus Pomatoschistus pictus Scomber scombrus Sparus aurata Symphodus melops Trachinus draco Trachurus trachurus

mul gehoornde slijmvis Lozano's grondel brakwatergrondel dikkopje kleurige grondel makreel goudbrasem zwartooglipvis grote pieterman horsmakreel

mullet tompot blenny Lozano's goby common goby sand goby painted gony mackerel gilthead seabream corkwing wrasse greater weever horse mackerel

schurftvis dwergtong lange schar schar tongschar Franse tong bot pladijs tarbot griet tong

scaldfish solenette Americain plaice dab lemon sole Dover sole flounder plaice turbot brill sole

stekelrog

thornback ray

harnasmannetje Engelse poon rode poon snotolf grauwe poon slakdolf Montagu's slakdolf zeedonderpad groene zeedonderpad

hooknose red gurnard tub gurnard lumpsucker grey gurnard striped sea-snail Montagu's sea-snail scorthorn sculpin longspined bullhead

adderzeenaald kortsnuitzeepaardje grote zeenaald kleine zeenaald

snake pipefish short-snouted seahorse greater pipefish Nilsson's pipefish

PLEURONECTIFORMES Arnoglossus laterna Buglossidium luteum Hippoglossoides platessoides Limanda limanda Microstomus kitt Pegusa lascaris Platichthys flesus Pleuronectes platessa Psetta maxima Scophthalmus rhombus Solea solea

RAJIFORMES Raja clavata

SCORPAENIFORMES Agonus cataphractus Aspitrigla cuculus Chelidonichthys lucernus Cyclopterus lumpus Eutrigla gurnardus Liparis liparis Liparis montagui Myoxocephalus scorpius Taurulus bubalis

SYNGNATHIFORMES Entelurus aequoreus Hippocampus hippocampus Syngnathus acus Syngnathus rostellatus

116

ANNEX IV: IN PREP RESULTS PRESENTED AT THE 10TH VLIZ YOUNG SCIENTISTS DAY 27 NOVEMBER 2009

117