Canada

0 downloads 0 Views 2MB Size Report
Calibration of the Ammonium Electrode: An Accumet 1003 specific ion meter .... calibration procedure was performed following instructions in the Accumet 1000 .....
Biogeochemical Observations to Assess Benthic Impacts of Organic Enrichment from Marine Aquaculture in the Western Isles Region of the Bay of Fundy, 1994

B.T. Hargrave, G.A. Phillips, L.l. Doucette, M~J. White, T.G. Milligan, D.J. Wildish, and R.E. Cranston

Science Branch Maritimes Region Department of Fisheries and Oceans Bedford Institute of Oceanography P.O. Box 1006 Dartmouth, Nova Scotia B2Y 4A2 Canada

1995

Canadian Technical Report of Fisheries and Aquatic Sciences 2062

1+1

F~heries

P~ches

and Oceans

et

Oc~ans

Canada

Canadian Technical Report of Fisheries and Aquatic Sciences Technical reports contain scientific and technical information that contributes to existing knowledge but which is not normally appropriate for primary literature. Technical reports are directed primarily toward a worldwide audience and have an international distribution. No restriction is placed on subject matter and the series reflects the broad interests and policies of the Department of Fisheries and Oceans, namely, fisheries and aquatic sciences. Technical reports may be cited as full publications. The correct citation appears above the abstract of each report. Each report is abstracted in Aquatic Sciences and Fisheries Abstracts and indexed in the Department's annual index to scientific and technical publications. Numbers 1-456 in this series were issued as Technical Reports of the Fisheries Research Board of Canada. Numbers 457-714 were issued as Department of the Environment, Fisheries and Marine Service, Research and Development Directorate Technical Reports. Numbers 715-924 were issued as Department of Fisheries and the Environment, Fisheries and Marine Service Technical Reports. The current series name was changed with report number 925. Technical reports are produced regionally but are numbered nationally. Requests for individual reports will be filled by the issuing establishment listed on the front cover and title page. Out-of-stock reports will be supplied for a fee by commercial agents.

Rapport technique canadien des sciences halieutiques et aquatiques Les rapports techniques contiennent des renseignements scientifiques et techniques qui constituent une contribution aux connaissances actuelles, mais qui ne sont pas normalement appropries pour la publication dans un journal scientifique. Les rapports techniques sont destines essentiellement a un public international et ils sont distribues a cet echelon. 11 n'y a aucune restriction quant au sujet; de fait, la serie reflete la vaste gamme des interets et des politiques du ministere des Peches et des Oceans, c'est-A-dire les sciences halieutiques et aquatiques. Les rapports techniques peuvent etre cites comme des publications completes. Le titre exact parait au-dessus du résumé de chaque rapport. Les rapports techniques sont résumés dans la revue Résumés des sciences aquatiques et halieutiques, et ils sont classes dans l'index annual des publications scientifiques et techniques du Ministere. Les numeros 1 a 456 de cette serie ont ete publies a titre de rapports techniques de ]'Office des recherches sur les pecheries du Canada. Les numeros 457 a 714 sont parus titre de rapports techniques de la Direction generale de la recherche et du developpement, Service des peches et de la mer, ministere de l'Environnement. Les numeros 715 a 924 ont ete publies a titre de rapports techniques du Service des peches et de la mer, ministere des Peches et de l'Environnement. Le nom actuel de la serie a ete etabli lors de la parution du numero 925. Les rapports techniques sont produits a ]'echelon regional, mais numerotes ]'echelon national. Les demandes de rapports seront satisfaites par l'etablissement auteur dont le nom figure sur la couverture et la page du titre. Les rapports epuises seront fournis contre retribution par des agents commerciaux.

i

Canadian Technical Report of Fisheries and Aquatic Sciences 2062

1995

BIOGEOCHEMICAL OBSERVATIONS TO ASSESS BENTHIC IMPACTS OF ORGANIC ENRICHMENT FROM MARINE AQUACULTURE IN THE WESTERN ISLES REGION OF THE BAY OF FUNDY, 1994

by B.T. Hargrave, L.I. Doucette, M.J. White, G.A. Phillips, T.G. Milligan Habitat Science Division Science Branch Maritimes Region Department of Fisheries and Oceans Bedford Institute of Oceanography P.O. Box 1006 Dartmouth, Nova Scotia B2Y 4A2 Canada

D.J. Wildish Aquaculture Division Science Branch Maritimes Region Department of Fisheries and Oceans St. Andrews Biological Station St. Andrews, New Brunswick EOG 2XO Canada

and R.E. Cranston Geological Survey of Canada Atlantic Bedford Institute of Oceanography P.O. Box 1006 Dartmouth, Nova Scotia B2Y 4A2 Canada

11



c Minister of Supply and Services Canada 1995 Cat. No. Fs 97-6/2062E ISSN 0706-6457

Correct Citation for this publication: Hargrave, B.T., G.A. Phillips, L.I. Doucette, M.J. White, T.G. Milligan, D.J. Wildish, and R.E. Cranston. 1995. Biogeochemical observations to assess benthic impacts of organic enrichment from marine aquaculture in the Western Isles region of the Bay of Fundy, 1994. Can. Tech. Rep. Fish. Aquat. Sci. 2062: v + 159 p.

iii

TABLE OF CONTENTS ABSTRACT

v

REsUME .............................................

v

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1

METHODS............................................

1

STATION LOCATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SAMPLE COLLECTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ANALYTICAL METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5 5

Vertical Profiles for Redox, Ammonium, and Sulfide . . . . . . . . . . .

5

1

. . . .

5

.

9

. . . .

9 9 9 9

Pore Water Salinity, Ammonium, and Sulfate . . . . . . . . . . . . . . . .

10

Calibration of the Redox Electrode . . . Calibration of the Ammonium Electrode Calibration of the Sulfide Electrode . . . Core Profile Sampling . . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

Sediment Water Content, Grain Size, Organic Carbon, and Nitrogen Water Content . . . . . . . . . . . . . Inorganic Grain Size . . . . . . . . . Organic Carbon and Nitrogen . . . Benthic Enrichment Index . . . . . .

Pore Water Extraction Salinity . . . . . . . . . Ammonium . . . . . . . Sulfate . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

6

7 8

10 10 10 10

Benthic Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Macrofauna Biomass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Water Movement Measurements . . . . . . . . . . . . . . . . . . . . . . . . 13 DATA PRESENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 WATER CONTENT . . . . SALINITY . . . . . . . . . . . INORGANIC GRAIN SIZE REDOX POTENTIALS . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

16 16 16 35

iv TABLE OF CONTENTS (continued)

TOTAL SULFIDES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DISSOLVED SULFATE AND AMMONIUM IN PORE WATER AND DERIVED AMMONIUM GRADIENTS AND CARBON BURIAL RATES A~ONIUM~LECTRODE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . ORGANIC CARBON AND NITROGEN . . . . . . . . . . . . . . . . . . . . . . BENTHIC ENRICHMENT INDEX . . . . . . . . . . . . . . . . . . . . . . . . . BENTHIC OXYGEN UPTAKE AND CARBON DIOXIDE RELEASE . . . MACROFAUNA BIOMASS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WATER MOVEMENT MEASUREMENTS . . . . . . . . . . . . . . . . . . . . RELATIVE SENSITIVITY OF VARIABLES FOR DETECTING DIFFERENCES BETWEEN CAGE AND REFERENCE SITES . . . . . .

36 36 37 37 38 38 39 41 42

ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 REFERENCES

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

APPENDIX A. SEDIMENT GEOCHEMICAL PROFILES . . . . . . . . . . . . . .

47

APPENDIX B. PORE WATER CHEMISTRY . . . . . . . . . . . . . . . . . . . . . .

39

APPENDIX C. BENTHIC METABOLISM . . . . . . . . . . . . . . . . . . . . . . . .

40

APPENDIX D. BENTHIC MACROFAUNA BIOMASS . . . . . . . . . . . . . . . .

41

APPENDIX E. SEDIMENT INORGANIC GRAIN SIZE SPECTRA . . . . . . . .

42

v ABSTRACT

Hargrave, B.T., G.A. Phillips, L.I. Doucette, M.J. White, T.G. Milligan, D.J. Wildish, and R.E. Cranston. 1995. Biogeochemical observations to assess benthic impacts of organic enrichment from marine aquaculture in the Western Isles region of the Bay of Fundy, 1994. Can. Tech. Rep. Fish. Aquat. Sci. 2062: v + 159 p. Sediment profiles of modal grain size, water content, salinity, redox potentials, ammonium, sulfides, sulfate, organic carbon and nitrogen, benthic dissolved oxygen and carbon dioxide flux, and macrofauna biomass were determined at 22 stations in the Western Isles region of the Bay of Fundy between June and August 1994. Eleven sites were located under salmon net-pens and 11 sites at distances >50 m from net-pens served as reference (control) locations. Totals= and redox potentials in surface sediment were the most sensitive indicators of benthic organic enrichment associated with salmon aquaculture.

Hargrave, B.T., G.A. Phillips, L.I. Doucette, M.J. White, T.G. Milligan, D.J. Wildish, and R.E. Cranston. 1995. Biogeochemical observations to assess benthic impacts of organic enrichment from marine aquaculture in the Western Isles region of the Bay of Fundy, 1994. Can. Tech. Rep. Fish. Aquat. Sci. 2062: v + 159 p. On a etabli des proflls de sediments sur la grosseur de grain typique, la teneur en eau, la salinite, les potentiels d'oxydoreduction, !'ammonium, les sulfures, les sulfates, le carbone et l'azote organiques, I' oxygene dissous et le flux du gaz carbonique benthiques ainsi que la biomasse de 1a macrofaune a 22 stations de la region des lles Western, dans la baie de Fundy, entre juin et aout 1994. Onze de ces stations se trouvaient sous des pares a saumon. Les onze autres, situees a des distances >50 m des cages a saumon, servaient de sites-temoins. Les donnees obtenues revelent que le potentiel d'oxydoreduction, le s= total et le carbone organique sont les indicateurs les plus sensibles des impacts de 1' enrichissement organique benthique associe a l'aquiculture du saumon. Un indice d'enrichissement benthique (produit du potentiel d'oxydoreduction et de la teneur en carbone organique) a ete etabli pour apprecier l'ampleur des impacts benthiques sur divers sites.

1

INTRODUCTION Observations in coastal inlets used for finfish aquaculture have shown that the benthic response to organic enrichment through enhanced sedimentation of particulate organic waste products from net-pen aquaculture is site specific, spatially limited, and highly dependent on physical factors such as water current speed, sediment composition, and seasonal storm-related resuspension (Findlay and Watling 1994; Gowen et al. 1994; Hargrave 1994a; Silvert 1994). While it is known that physical and biological variables affect geotechnical and sedimentary properties (Snelgrove and Butman 1994), relationships between the rate of organic loading and resulting changes in sediment properties are not well known. It is thus not presently possible to predict the impacts of increased organic matter deposition at sites presently occupied or potentially available for aquaculture development. There is also often a small or nonexistent database for comparing present sedimentary conditions with those in the past to assess the impacts of finfish aquaculture to allow assessment of long-term changes in ecological conditions within benthic habitats of inlets used for the expanding aquaculture industry. Finally, there is an urgent need to compare various methods that might be used to assess long-term environmental changes associated with aquaculture development and to choose the method best suited for monitoring changes in environmental conditions. The rapid development of salmonoid aquaculture in the Western Isles (Quoddy region) area of the Bay of Fundy has resulted in several studies directed toward assessment of environmental impacts of this new industry (Hargrave et al. 1993a; Milligan 1994; Strain et al. 1994; Trites and Petrie 1995; Wildish et al. 1988; 1990; 1993). The present report adds to this growing base of marine environmental information in the Western Isles region by comparing different methods that may be used to assess changes in geotechnical, geochemical, and biological properties in subtidal sediment under salmon aquaculture sites. Sediment profiles of modal inorganic grain size, water content, salinity, redox potentials, ammonium, sulfate, organic carbon and nitrogen, benthic fluxes of dissolved ~ and C02 , and macrofauna biomass were measured between June and August 1994 at 22 stations located in Letang Inlet, Passamaquoddy Bay, Deer Island, Campobello Island, Beaver Harbour, and Deadman's Harbour, N.B. Eleven sites were under salmon net-pens, and 11 control stations were located >50 m from these sites. The data can be used to assess the degree of sensitivity of different variables to detect benthic impacts of net-pen aquaculture at different sites.

METHODS STATION WCATIONS Sediment cores and bottom grab samples were collected between June 3 and August 15, 1994, at 22 stations in Letang Inlet (Table 1; Fig. 1). Latitude and longitude were recorded for each site at the time of sampling using a hand-held Sylva GPS. As sediment cores were collected by diver, sampling time during daylight hours was selected to coincide approximately with the time of low tide.

2 Table 1. Location of 22 sampling sites from 1994 illustrated by name in Figure 1 in the Western Isles region of the Bay of Fundy. Chart datum depth, from Canadian Hydrographic Service Charts 4124, 4331, and 4114, is for the lowest normal tide which at Letang Harbour is 3. 7 m below mean water level. DFA (New Brunswick Department of Fisheries and Aquaculture) statistics on site capacity and area were provided by B. Chang (pers. comm.).

Site No.

Name/Location

Depth (m)

DFA Licence Capacity

Cage Area (hectares)

240 000 40 000 150 000 160 000 200 000 320 000 320 000 80 0001 150 0001 60 0001 320 000 60 000

9.63 6.40 13.05 10.10 17.21 22.73 22.73 1.801 6.901 6.801 20.36 2.39

Cage Sites: 1 2 3 4 5 6 6 7 8 9 10 11

Beaver Harbour Deadman's Harbour Reserve Cove Bar Island Hills Island Frye Island (June) Frye Island (September) Clam Cove Clam Cove Letang Head Harbour de Lute Head Harbour

8 10 14 14.5 20 11 11 8 14 18 15 8

Reference Sites: 6A

12 13 14 15 16 17 18 19 20 21 22 1

Frye Island (September) Birch Point Goss Point Granger Point Scotch Bay Lime Kiln Bay Back Bay Deadman's Harbour Digdeguash Pendleton West Pendleton East Head Harbour Island

Based on data from July 1991

12 6 12 17 6 7 12 12 7 15 12 9

3 SAMPLE COLLECTION

Six cores were collected at each station by diver. Three short cores containing approximately 15 em of sediment used for benthic gas flux measurements were taken at each sampling site using acrylic plastic tubes (5.7 em inside dia., 20 em length) which were inserted into the sediment to preserve an intact sediment-water interface leaving a water-filled head space of approximately 2 em. Plastic caps or rubber stoppers were used to close the top and bottom of each core. Three long cores containing approximately 30 em of sediment used for geochemical and geotechnical profiles were taken using longer core tubes (6.5 em inside dia., 50 em length). Plastic caps or rubber stoppers were used to close the top. and bottom of each core. These cores were used to determine vertical profiles of sediment properties by insertion of electrodes (for oxidation reduction potentials, ammonium, and sulfide) and for withdrawing samples for determinations of water content, grain size, organic carbon and nitrogen, and pore water analysis of salinity, sulfate, and ammonium as described below. Holes (1.9 em dia.) drilled in the tube at 2-cm intervals over the length of the long core tubes were covered with duct tape to prevent water and sediment loss. Cores were kept upright, handled as gently as possible to maintain an undisturbed sediment-water interface, and transported to the laboratory in a chilled cooler within a few hours after collection. They were stored in a refrigerator (2 to 4 °C) prior to subsampling (within 48 h of collection). Sediment for macrofauna biomass determinations was collected using a Hunter 0.1 m2 grab (Hunter and Simpson 1976). The grab functions well in mud and sandy sediments. The side jaws are extended to form "cheeks," which prevents jamming of the jaws by small stones and subsequent loss of material. Five grabs were taken at each station. All sediment in each grab was placed in a 20-L plastic bucket, covered with a tight-fitting lid and stored at 2 to 4 oc. The volume of sediment was determined by measuring the height of sediment in each bucket, and all sediment in each grab was used for macrofauna biomass determinations.

ANALYTICAL METHODS Vertical Prof"IIes for Redox. Ammonium. and Sulfide Calibration of the Redox Electrode: An Orion® (9678BN) combination redox (platinum) electrode and an Accumet 1003 millivolt meter were used to determine oxidation-reduction (redox) potentials in two of the three long cores collected at each sample location. Reagents were prepared 12 to 24 h before use and held refrigerated. Redox Standard A (0.1 M potassium ferrocyanide and 0.05 M potassium ferricyanide) was prepared by weighing 4.22 g K,.Fe(CN) 6 .3H20 and 1.65 g K3Fe(CN) 6 into a 100-ml volumetric flask. Approximately 50 m1 of distilled water was added with swirling to dissolve the solids. The solution was diluted to volume with distilled water. Standard B (0.01 M potassium ferrocyanide, 0.05 potassium ferricyanide, and 0.36 M potassium fluoride) was prepared by weighing 0.42 g K,.Fe(CN)6.3H20, 1.65 g K3Fe(CN) 6 , and 3.39 g KF.2H20 into a 100 m1 volumetric flask. 50 m1 of distilled

4



Figure 1. Map of place names for 22 sampling sites in the Western Isles region of the Quoddy region of the Bay of Fundy. Uncircled numbers indicate operating aquaculture sites; circled numbers are reference (control) sites.

5

assamaquoddy Bay

0I

2000 I

meters

6

water was added to dissolve the solids, and the solution was diluted to 100 ml with distilled water. A KCL filling solution for the internal reference electrode was prepared by placing 2.5 m1 of 4 M KCL in 50 ml and diluting with distilled water to give a 0.2 M KCl solution. Redox standards were used to calibrate electrodes at room temperature (20°C) at the start and end of measurements on each core. Standard A was transferred to a 150-ml beaker, and the electrode was placed in the solution until the reading stabilized with stirring (1 to 2 min.). The potential of Standard A was approximately + 147 + 9 mV. The electrode was rinsed and the measurement repeated with Standard B (potential of +216±9 mV). The potential in Standard A was approximately +66 mV greater in Standard B. Quinhydrone (pH 7.0, +31 mV) was also used as a standard. The potential of the reference electrode ( +244 mV at 20°C), corrected for the average difference between measured potentials of standard solutions and their calibration values, was added to the mV reading to determine the actual Eh potential in sediment samples. Eh potentials of approximately +300 to +350 mV are typical of oxygenated seawater. The redox electrode was rinsed with distilled water after use and stored for short periods (a few weeks) in tap water. For longer periods, the electrode was drained, rinsed with distilled water, and stored dry.

Calibration of the Ammonium Electrode: An Accumet 1003 specific ion meter with a NH4 + electrode (Fisher 13-620-505) and a combination pWreference electrode (Fisher 13-620-111) was used to determine ammonium proflles in sediment withdrawn from long sediment cores following redox measurements. Reagents were prepared 12 to 24 h before use and held refrigerated. The NH4 + electrode filling solution (Fisher 13-620-803) containing 0.01 M ammonium chloride/silver chloride was toppedup every few days and renewed if readings beeame erratic or unstable. A stock 10M NaOH solution was prepared by dissolving 40 g of NaOH in 100 ml distilled water. A 0.1 M NaOH was prepared by transferring 10 ml of this stock solution to a 1-L volumetric flask and filling to the mark with distilled water. The solution was well mixed by shaking for several minutes. Stock solutions for preparing NH4 + standards were kept in the refrigerator and diluted just before use. Dilute standards are stable for up to 3 h, so solutions were prepared in the morning and at mid-day on the day of use for calibration of electrodes used in sediment samples processed during morning and afternoon periods. Standards were diluted to yield two concentrations (100 and 1000 J.tM). The standards are unstable when exposed to air, thus measurements of samples were made as rapidly as possible to allow a calibration check to be performed immediately after samples from each core were processed. The ammonium electrode was rinsed, blotted, and dried after removal from standards prior to placement in sediment samples described below. The ammonium electrode was standardized using the two freshly diluted stock ammonium solutions. A stock 0.1M NJI4Cl standard was prepared by weighing 0.535 g NH4Cl into a Biotight jar containing 100 ml of distilled water. This standard

7 was stored in a refrigerator and freshly prepared for each series of analyses. A 1000 pM dilute ~Cl standard was prepared by transferring 1 ml of 0.1 M ~Cl stock to a Biotight jar and diluting to 100 ml with distilled water. A 100 pM dilute ~Cl standard was prepared by transferring 10 ml of the 1000 pM standard solution to a Biotightjar and diluting to 100 ml with distilled water. As mentioned above, these standards were prepared every 3 to 4 h as they are unstable. The final standards for calibration of the ammonium electrode were prepared by adding 0.9 ml of 10M NaOH to the 1000 pM standard and 1.0 ml of 10 M NaOH to the 100 pM standard. The solutions were well mixed and used as standards following the procedure for a twopoint calibration using the Accumet 1003 specific ion meter. The ammonium and pH reference electrodes were rinsed with distilled water after use and stored for short periods (a few days) in a 0.01M NH4Cl solution without NaOH added. For longer periods, the electrode was drained, rinsed with distilled water, and stored dry.

Calibration of the Sulfide Electrode: An Accument 1003 specific ion meter and Orion silver/sulfide electrode (9416BN) with a combination pH/reference electrode (Fisher 13-620-111) was used to determine total sulfide in sediment profiles. SAOB (sulphide anti-oxidant buffer solution) was prepared in 250-ml plastic screw-top jars by adding 20.00 g of NaOH and 17.90 g EDTA (Na2C100 8N2.2H20) and diluting to 250 ml with distilled water. The 2M NaOH and 0.2 M Na2EDTA solution was stored in a refrigerator. Immediately before sample analysis, 8. 75 g L-ascorbic acid (preweighed in scintillation vials) was added to each 250 ml of the NaOH-EDTA solution. The SAOB buffer solution is stable for up to 3 h after addition of L-ascorbic acid. As 5 ml of this solution was used for each sample analyzed (described below), 250 ml was sufficient for 50 samples. Stock and diluted S = solutions used as standards for electrode calibration standards were kept in the refrigerator and diluted just before use. As with NJ4 + standards, dilute S = standards are unstable when exposed to air. Diluted standards are stable for up to 3 h, thus they were made up in the morning and at mid-day on the day of sample analysis. The s= electrode was calibrated before and after analysis of each core. Two s= standards (100 and 1000 pM) were used for a two-point electrode calibration. A stocks= solution of 0.01 M Na2S was prepared by weighing 0.2402 g Na2S.9H20 into a Biotight jar and diluting to 100 ml with distilled water. Na2S.9H20 is hygroscopic and should be handled with rubber gloves in a fume hood. The solution was made fresh every 48 hand stored refrigerated in a dark bottle. A 1000 pM s= standard (l0-3 M) was prepared by transferring 10 ml of the 0.01 M Na2S stock solution into a Biotight jar and diluting to 100 ml with distilled water to give a 1000 pM (10"3 M) solution. A 100 pM s= standard (104 M) was prepared by transferring 10 ml of the 1000 pM standard to a Biotight jar and diluting to 100 ml with distilled water. Both diluted standards were mixed thoroughly before use. A two-point (100 and 1000 pM) calibration procedure was performed following instructions in the Accumet 1000 series instruction manual. A three-point (10 pM, 100 pM, and 1000 pM) calibration procedure is more precise for sediment samples with low sulphide concentrations (1 pM

8 to 10 #-'M). Just before calibration of the s= electrode, 25 ml of each standard was transferred to a dark bottle and 25 m1 of SAOB (containing ascorbic acid) was added. The combined solution was kept tightly capped until used for standardizing the s= electrode. The S = and pH reference electrodes were rinsed with distilled water after use and stored in pH 7.0 buffer for periods of a few days. For longer periods, the electrodes were stored dry. Core Prof"Ile Samplin&: Redox,~+, and s= electrodes were attached to three separate Accumet 1003 specific ion meters; and calibration procedures were applied for each electrode prior to analysis of two long sediment cores collected at each site. Duct tape was removed from individual holes starting at the top of a core. The Eh electrode was inserted horizontally in the first hole to the centre of the sediment column and a mV reading recorded after 1 to 2 min. The electrode was then moved (without cleaning) to the next lower hole. A-5 m1 subsample for determination of sediment water content, grain size, and particulate organic carbon and nitrogen (described below) was taken with a cut-off 10-ml syringe after removal of the electrode. The 10-ml cut-off syringe was re-inserted for taking a second horizontal subcore for ammonium and sulfide determinations. Sediment was expelled as two 5-ml subsamples (using a flat-tip spatula to slice the sediment at the 5-ml point) into two 100-ml Biotight wide-mouth plastic jars. 30 m1 of 0.1 M NaOH was added to one subsample for ~ + determination, and 5 m1 of SAOB buffer (to which L-ascorbic acid had been added) was added to the second subsample for s= determination. The SAOB solution was used within 3 h of preparation. A magnetic stirrer was used to continuously stir the slurry of the NH4 + sample containing 30 m1 of 0.1 M NaOH. The ~ + electrode was mounted in a stand for positioning in the sample during stirring. A flat-tip stainless steel spatula was used to mix and homogenize sediment samples with the SAOB and NaOH solutions in each Biotight jar. Following this, the s= electrode was used to stir sediment. The sediment slurry with only 5 ml of SAOB added usually contained insufficient liquid to allow use of a magnetic stirring bar. Electrode drift and time for stable readings varied between electrodes. The ~ + and S = electrodes usually stabilized in 2 to 4 min., while the Eh electrode stabilized in less than 1 min. The "auto" feature on the Accumet 1003 meter was used as a stability indicator to lock the display when the mV reading was unchanged for 5 sec. If there was a large variation between successive mV readings with increasing depth in a core, a check was performed by repeating the measurement on the preceding sample. Electrodes were not cleaned between readings, unless it was necessary to measure the standards. In this case, sediment was removed from the electrode using a Kimwipe towel, and electrodes were rinsed with distilled water and dried before immersion in standard solutions. Electrodes were usually transferred directly from one sample to the next without cleaning, rinsing, or calibration, going successively down the core from

9 hole to hole. Calibration checks with standards were usually made before the first and after the last sample from one core. Processing time was approximately 2 h for one core (approximately 15 samples at 2 em depth intervals over a 30 em long sediment column). Measurements were made as rapidly as possible to allow calibration checks (at least every 2 h). This minimized errors due to drift between readings during the time required for one set of measurements on a core.

Sediment Water Content. Grain Size. Organic Carbon. and Nitrogen Water Content: Pre-weighed scintillation vials, filled with approximately 5 m1 of wet sediment withdrawn from each sample depth in each core, were weighed as soon as possible ( +0.1 g) after sample collection. Vials were placed in a drying oven (60°C) for 48 h. Sediment water content was determined as the weight loss on drying expressed as a percentage of the original sediment wet weight. Inorganic Grain Size: Dry sediment used to determine water content was pulverized in a porcelain mortar and pestle, and 2 to 4 mg subsamples were taken for analysis of inorganic grain size using methods described in Milligan and Kranck 1991. Samples were weighed dry in tared 30-ml pyrex beakers on a Mettler AE163 balance ( +0.1 mg) and treated with excess 30% H20 2 to remove organic matter. Samples were dried, reweighed, resuspended in electrolyte, sonified, and counted for disaggregated inorganic grain size using a Coulter Multisizer liE. The data were plotted as smoothed histograms of log10 equivalent weight percent determined from the volume of particles in each size class using a specific gravity of 2.65 kg m·3 , against log 10 diameter in micrometres. Each particle size spectrum, which ranges in value from 0.1 to 30%, has been offset by one decade on a dimensionless y axis to show variations in grain size with core depth. Modal diameters were determined from the plots. Organic Carbon and Nitrogen: Dry sediment subsampled for grain size was also used for determinations of organic carbon and nitrogen using a Perkin Elmer 240 analyzer with a combustion temperature of 600°C. Ignition temperatures of > 700°C are required to remove carbonates (P. MacPherson, unpubl.), and thus it was assumed that a combustion temperature of 600°C avoided loss of carbonates. The measures of total carbon were assumed to represent organic carbon. Benthic Enrichment Index: Data for water content (W, percent of sediment wet weight), Eh (R, mV), and organic carbon (C, percent sediment dry weight, 0 to 2 em layer) were combined to calculate an index of benthic enrichment (BEl) (units of mol C m·2 x mV) as described in Hargrave 1994b by: BEl = [{(100-W)/100}

X

let x {(C/100)/12}1 x R

10

Pore Water Salinity. Ammonium. and Sulfate Pore Water Extraction: The third long core collected at each sampling site was used for pore water extraction for analysis of salinity, dissolved ammonium, and sulfate. Duct tape was removed from individual holes starting at the top of a core as described above. A 10-ml cut-off syringe was used to withdraw two horizontal subcores at 1- to 2-cm depth intervals down the length of the core (depending on sediment water content). Each sample contained from 10 to 15 m1 of wet sediment which was placed in 50-ml tubes. These were immediately capped and refrigerated until pore water extraction. A refrigerated (5°C) centrifuge was used to separate pore water from sediment. A top-loading pan balance was used to tare samples in tube holders to balance the centrifuge head. Centrifugation time was determined by anticipated sediment water content (sediment at the bottom of cores with 70% water for 10 min.). Supernatant water was removed using a plastic syringe with a Luer-lock end. A fllter disc (1.2 I'm pore diameter filters Cameo 25NS) was attached to the syringe, and the flltered pore water sample was placed into a clean 5-ml vial. Vials containing pore water samples were capped tightly and kept frozen (-18°C) until analysis for salinity, dissolved ammonium, and sulfate. Salinity: A 50-J.'l subsample of flltered pore water was diluted with 200 J.'l of deionized water, and conductivity was measured with a Horiba Model B-173 conductivity meter. Standard IAPSO sea water was used as a calibration standard. Precision and accuracy limits resulted in an uncertainty range of +0.3%o. Ammonium: A 200-,.d subsample of pore water was placed in a 15-rnl test tube containing 1 m1 of deionized water, 0.5 ml of phenol alcohol solution, 0.5 rn1 of nitroprusside solution, and 1 ml of oxidizing solution (Solarzano 1969). Samples and standards were allowed to stand for 2 h for a blue colour to develop. The colour intensity was measured at 640 nm using a Milton-Roy Mini20 colorimeter. Standards were prepared from ammonium chloride. Precision and accuracy limits are +0.05 mM. Sulfate: A 50-J.'l subsample of pore water was placed in a 7-ml cuvette. 50 J.'l of 300 mM BaC~ was added to precipitate the sulfate in the sample. 4 ml of deionized water was added to the solution, and the turbidity was read on a Milton-Roy Mini20 spectrophotometer fitted with a turbidity attachment. Standards were prepared from open-ocean surface seawater. Precision and accuracy limits are + 1 mM.

Benthic Metabolism Benthic metabolism was measured as oxygen uptake and C~ release in three short sediment cores collected with overlying water from each sampling site. Sediment



11

filled from 75 to 80% of the core length. Cores were stored upright in a cool place at or below ambient temperature, avoiding vibration and disturbance of the sediment-water interface. If there was excess sediment in a core, sediment was removed from the bottom of the core to increase the supernatant volume with care not to disturb the surface sediment layer. With the upper end of the core tightly capped, the upright core was fully immersed in a bucket of freshly collected seawater from the sampling site. The bottom end cap of the core was gently removed and the upper cap loosened to allow sediment to be removed from the bottom. The desired amount of sediment in a core was reached when the position of the sediment surface was about 2 em below the upper core rim. The lower core cap or rubber stopper was reinserted and the upper cap was replaced, avoiding air spaces. Prior to sealing cores for incubation, it was necessary to replace the supernatant water with filtered seawater from the study site with and without the addition of a metabolic poison. When the procedure described above to adjust the amount of sediment in a core was used to rapidly replace supernatant water, resuspension of surface sediment often occurred. A slower but more gentle method was to use a plastic squeeze bottle to replace supernatant water slowly. A 50-ml disposable syringe was used to remove as much supernatant water as possible (taking care not to remove any sediment or suspended particles). If this water was turbid, a core was allowed to stand for 30 to 60 min. to allow resuspended particles time to settle. Supernatant water over undisturbed sediment in the cores was replaced with glassfibre filtered seawater which had been aerated for 15 to 20 min. to increase the dissolved oxygen concentration. Seawater was placed in a modified wash bottle fitted with small-diameter tubing leading to a 1-ml plastic pipette taped to a spatula. When seawater was squeezed gently from the bottle, it was deflected from the pipette by the spoon end of the spatula which was held just above the sediment surface. The initial incubation for measurements of dissolved oxygen and C02 flux was carried out with fresh seawater added to cores for measures of total gas exchange. Following this incubation (30 to 90 min.), the seawater was replaced with seawater prepared by dissolving 1 g of HgC12 in 1 L of filtered seawater. Seawater was heated to - 80°C, and a magnetic stirrer was used to ensure complete dissolution of HgC12 • The final concentration of HgC12 in supernatant water was 0.1% (w/v). Methods of dissolved oxygen and C02 analysis and calculations of rates of gas exchange are described in Hargrave and Conolly 1978 and Hargrave and Phillips 1981. Initial samples were taken from the supernatant water in each core for dissolved oxygen and C02 analysis using plastic disposable or glass syringes. A Radiometer gas analyzer (polarographic electrode system fitted with a thermostated jacket and small volume injection assembly) was used for measuring dissolved oxygen in 1-ml water samples. Zeroing the instrument was carried out using a fresh solution of 0.2 g Na2S03 added to 10 ml of distilled water. This removes all dissolved oxygen from the water and allows the instrument zero value to be set after drift has stabilized (about 5 to 10 min.). It was necessary to check the zero setting only once per week, or less. An air blank (water saturated with bubbling air) was used to provide an upper calibration value.

12 This value was set at 130.5 mm Hg after rinsing the electrode with fresh distilled water several times to remove all trace of Na2S03 • The air blank was checked after each sample for the first several samples analyzed after zeroing the instrument. The zero value was found to be very stable, but the air blank was checked frequently. After sampling for initial values of dissolved oxygen and C02 , cores were capped with tight-fitting plexiglass tops. A small amount of petroleum jelly was used to make an airtight seal. The caps were fitted on the inside with magnetic stirring bars held in plastic rotating sleeves. Magnets were positioned in the center of each cap and rotated freely without trapping air bubbles in supernatant water when caps were replaced over cores. The cores were arranged around a motor-driven rotating magnet (30 rpm) - a speed sufficient to keep the supernatant water mixed without disturbing the sediment surface. Cores and the magnetic stirrer were placed in a dark refrigerator set to approximate ambient temperature ( +2°C) at the collection site. At the end of the incubation period, caps were carefully removed and duplicate samples were taken by syringe from each core. One was used for oxygen and one for C02 determinations. The length of the incubation was chosen to result in a 5 to 50% change in gas concentrations. For most cores containing organically rich fine-grained sediments collected under salmon net-pens, this was from 90 to 120 min., depending on temperature. Longer incubation times were required at lower temperatures. Cores with lower amounts of organic matter collected at control sites away from net-pens required longer incubation periods (4 to 6 h). After the initial incubation, the volume of supernatant water in each core was measured using a 50 m1 syringe to remove the water without disturbing the sediment surface. Volume was measured in a graduated cylinder to the closest 0.5 ml. The supernatant water was replaced with seawater containing 1% (w/v) HgC12 using the spatula-deflecting delivery system described above. Small amounts of sediment were sometimes resuspended, but when this occurred sediment settled rapidly if cores were left undisturbed for a few minutes. The headspace in the cores above the sediment was completely filled to the top of the core tube to allow removal of initial samples for dissolved oxygen and C02 as described above. Rates of changes in dissolved oxygen and C02 before and after addition of seawater containing HgC12 were determined by calculating changes in the absolute amounts of the gases in water over a core (mg I-1 at the start and end of each incubation period multiplied by the supernatant water volume in fractions of a litre), corrected for the incubation time (in hours) and cross-sectional area of the sediment core (25.6 cm2 for cores 5. 7 em inside dia.). Fluxes before and after addition of HgC12 were reported as mg 0 2 and C02 m-2 h- 1•

13

Macrofauna Biomass Sediment from grab samples was quantitatively sieved through 0.5 mm mesh screen by gentle washing using cold running seawater. Macrofauna were removed and placed in 5% buffered formalin solution. Fauna were grouped into trophic types {D=deposit feeders, S=suspension feeders, P=predators, O=omnivores, and H=herbivores) as described in Wildish and Peer 1983. Cumcumariafrondosa (Holothuroidea) was considered to be a suspension feeder in our study. Dentalium entale was inadvertently weighed as suspension feeders. Molluscs were enumerated and dissected to remove tissue from shells for weight determination. Wet weights of specimens were determined within 1 h of sorting by blotting individuals or tissues on absorbent paper (15 sec.) and rapid weighing (15 to 30 sec.) to reduce weight loss through desiccation. Specimens were placed in a 5% buffered formalin solution for 3 to 4 wk and transferred to 90% ethanol prior to analysis of species composition.

Water Movement Measurements A commercially available wall patch compound was used to cast cylinders of - 15 em long x 4 em wide and of - 150 g dry weight. They were deployed in a metal frame which acted as a swivelling vane to face the current. After field exposure for a known time and drying in a 60°C oven, the change in mass was used to calculate the gypsum dissolution index (DI) due to water movement: [W0- W 1] flow DI = -----------------[Wo-Wt1 still

from the known weight loss (W0-W1) measured in still seawater at the same temperature and salinity (Wildish et al., unpubl.) and where water movement was absent.

DATA PRESENTATION Data for different variables measured in this study are presented by station number in four appendices. Sediment profiles of water content, redox potentials, total sulfides, ammonium, organic carbon, and nitrogen appear in Appendix A; pore water profiles of salinity, ammonium and sulfate are presented in Appendix B; benthic oxygen uptake and carbon dioxide release data appear in Appendix C; and biomass of macrofauna in different trophic groups are tabulated in Appendix D. Sediment inorganic grain size spectra are shown graphically in Appendix E. Summaries of means, maxima, minima, standard deviations, and standard errors for each variable for grouped cage and reference site stations for all variables are presented in Table 2. Values for geotechnical and geochemical variables are summarized for the surface (0 to 2 em) sediment layer. For stations where it was possible to calculate significant (p 0. 76) with depth over the number of sampling intervals indicated in parentheses.

Site

n

Mean Salinity (u) (%o)

11 13 3 15 3 15 15 13 10 5 7 8

30.0 26.6 24.5 30.0 30.2 29.5 32.2 28.9 26.0 27.6 30.5 29.9

Ammonium Gradient

Carbon Burial Rate

(mM m·1)

(g C m·2 d"1)

Cage Sites: 1 2 3 4 5 61 62 7 8 9 10 11

(1.3) (3.3) (5.2) (1.5) (1.6) (2.1) (0.4) (2.5) (3.3) (4.2) (0.7) (1.8)

4.30 (11) 5.10 (6)

0.099 0.120

2.00 (15) 0.96 (15) 2.10 (13)

0.046 0.022 0.048

9.50 (4) 4.50 (7) 3.20 (8)

0.220 0.100 0.074

0.83 1.70 0.83 1.00 1.30

(16) (5) (6) (8) (14)

0.019 0.039 0.019 0.023 0.030

0.54 (9) 0.89 (12)

0.012 0.020

Reference Sites: 6A2 12 13 14 15 16 17 18 19 20 21 22

16 5 9 11 14 4 10 13 15 3 2 3

1 June 29, 1994 2 September 27, 1994

32.1 (0.4) 26.0 (4.7) 27.9 (5.2) 29.8 (2.1) 27.3 (4.3) 24.8 (2.6) 27.8 (3.7) 27.3 (5~7) 27.5 (2.2) 30.1 (1.1) 31.5(-) 29.7 (1.1)

18

Figure 2. Summary of variables measured at 22 stations in the Western Isles region of the Bay of Fundy during 1994. Horizontal lines indicate mean values calculated in Tables 2 and 3 for cage (Stations 1 to 11) and reference (Stations 6A and 12 to 22) sites described in Table 1. Shaded bars in Panels B and F represent the range of values for bottom water salinity and sulfate concentration in the region.

''

19

~

1 indicate that variance at cage sites was greater than that at the reference sites.

Variable

Benthic Enrichment Index Redox Potential (Eh) Chemical C02 Release Macrofauna Biomass (Suspension Feeders) Ammonium Gradient Ammonium (Pore Water) Carbon Burial Rate Total C02 Release Total Oxygen Uptake Pore Water Sulfate Total Sulfide Macrofauna Biomass (Deposit Feeders) Macrofauna Biomass (Total) Pore Water Salinity Chemical Oxygen Uptake Nitrogen Grain Size Organic Carbon Deposit Feeders as Percentage of Total Percent Water Ammonium (Electrode)

Number

Ratio

17 4 14 21 15 8 16 13 11

8.51 3.95 2.21 1.83 1.78 1.60 1.74 1.26 1.24 1.16 1.14 1.11 1.02 0.90 0.85 0.81 0.74 0.74 0.72 0.69 0.50

6 5 19 18 2 12 10

3 9

20 1 7

32

Figure 3. Mean values for different variables measured at cage and reference sites presented by number in Table 2. Line indicates a slope (b) of 1 (equal values for means at cage and reference sites). Values below the line indicate that the mean value for that variable at all cage sites was higher than the same variable at the reference sites. Points on the line indicate that mean values were similar at cage and reference sites. Points above the line indicate that mean value for reference sites was greater than that at cage sites.

33

17

• •

10 3

18

UJ

w

UJ

~

w

0

zw

a: w LL w a: a:

4

• 10 2

7 .3

1

s•

21

• • 5

2

11 .13 • • 20 19.14

6

10 1

.12

0

LL

9



z < w 10°

e1s

~

w

_J

m < 10- 1 a:

-

< >

16



1 o- 2 10- 2 10- 1 10° 10 1

10 2

10 3

VARIABLE MEAN FOR CAGE SITES

34

Table 8. Ranked probability values (p) from one-way ANOVA comparisons of variables measured at cage and reference sites (Table 2). p values 100 mv (Fig. 2D). The group mean + u for Eh for reference sites was +217 + 113 mV (Table 2). Redox potentials were also usually positive in subsurface sediment layers at these sites. In contrast, redox potentials at all cage sites, except Sites 2, 10, and 11 were < + 100 mV and in most cases < 0 mV. The group mean + u for these sites was +31 + 75 mV (Table 2). Redox potentials > + 100 mV indicating an oxidized sediment-water interface were observed at Site 6A during 1990 and 1991 similar to the mean value of + 117 mV in the present study. Values between + 100 and -100 mV, characteristic of

36 anoxic sediments, were measured under cages at Site 6 in this earlier study (Hargrave et al. 1993a). Eh potentials at all cages sites other than 2, 10 and 11, with potential < + 100 mV, show that organic matter loading occurs at these stations in excess of those occurring at reference sites. Site 4 is the most heavily impacted, followed by Sites 5, 9, and 1.

TOTAL SULFIDES Totals= concentrations in surface sediment layers at all cage sites were 10 to 1000 times greater than values measured at reference sites (Table 2; Fig. 2E). At most sites, totals= levels increased between 4 and 18 em in subsurface layers. At some sites, however, there was no consistent increase ins= with sediment depth, or the subsurface maxima extended over a broad depth range. While at some sites there was considerable variation between the two cores collected at one site (for example, Sites 6 [June 24], 10, and 12), most proflles in replicate cores were similar. Sites 1, 4, 5, 8, and 9 had mean totals= levels > 1000 JLM in the surface sediment layer indicating that high rates of organic matter loading had led to s= accumulation. Maximums= levels (6 to 7 mM) occurred in subsurface (2 to 8 em) sediment layers at Site 1 (Appendix A). S = concentrations in surface sediments at Site 6 have fallen dramatically from high levels(> 100 mM) reached in 1990-1991 (Hargrave et al. 1993a) to < 1 mM in the present study (Appendix A). Fish pens at Site 6 were moved to a new location (approximately 500 m north of their original location) in July-August 1992. Site 6 sampled in the present study was therefore the old farm site which had not received direct input of particulate wastes from fish pens for 2 yr. During this time, total S = levels in sediments at this site have been reduced by up to two orders of magnitude.

DISSOLVED SULFATE AND AMMONIUM IN PORE WATER AND DERIVED AMMONIUM GRADIENTS AND CARBON BURIAL RATES Attempts to normalize pore water sulfate and ammonium results to a constant salinity did not remove excessive variation in the data for these variables. Within the variation of sulfate determinations ( + 1 mM) and considering the problems due to sample extraction indicated by salinity measurements, with the exception of cores from Stations 9 and 10, the sulfate concentrations in pore water did not decrease with depth as would be expected with removal due to sulfate reduction. In contrast, while dissolved ammonium concentrations were as variable as those for sulfate, perhaps due to problems of sample preparation mentioned above, significant gradients of increasing ammonium with sediment depth (correlation coefficients > 0.95) were measured in 15 of the 24 cores (Table 3). These gradients (units mM m-1) were used to calculate carbon burial rate (CBR) (units g C m-2 d- 1) using the equation: CBR = 0.023 x (Ammonium Gradient) derived in Cranston 1994.

37 This analysis shows that Site 9 is the most impacted of the farm sites examined with organic carbon burial rates of 0.22 g C m·2 d-1 • Sites 2, 1, and 10 follow with values for CBR of approximately 0.1 g C m·2 d-1 • A relative 95% confidence interval for these estimates (due to variance in sample handling, storage, and analyses) calculated from data summarized by Cranston (1994) to derive the empirical equation between ammonium gradients and carbon burial rates is +25%. In the present study, measurements of CBR were based on the analysis of ammonium gradients in a single core from each site. Heterogeneity in sedimentary properties under salmon net-pens would result in a higher confidence interval for CBR at cage sites.

AMMONIUM (ELECTRODE) Ammonia concentrations measured with the Orion® electrode were > 200 JLM at all cage sites with the exception of Site 3 and < 200 JLM at all reference sites (Table 2; Fig. 2G). At most sites, ammonium concentrations increased with sediment depth, but there were some sites (Sites 4 and 10) where a reverse trend occurred with maximum values in surface sediment layers. The lack of data from the ammonium electrode at Cage Sites 5 and 6 was caused by contamination of the electrode with an organic film after analysis of highly organic sediment from Site 4. There was a significant positive linear correlation between the measurements of ammonium in surface sediments from different sites by the two methods (r-2=0.80, n=22), but ammonium concentrations in pore water were lower than values measured by the electrode (b=0.38). Both techniques indicated that ammonium concentrations were highest at Cage Site 4 (1.2 and 3 mM, respectively).

ORGANIC CARBON AND NITROGEN The average value for organic carbon at cage sites was increased by 40% over the value for reference sites (Table 2); however, the range of values for organic carbon (C) under cages (C=0.8 to 3.8%) was similar to that at reference sites (C=0.4 to 3.2%). A similar percentage increment occurred for nitrogen (N) between cage and reference sites, and the ranges of values at these two groups of stations overlapped (0.08 to 0.6) (Table 2). The absolute values for C and N, and ranges, are similar to those measured in other coastal areas of the Bay of Fundy such as the Annapolis Basin (Hargrave et al. 1993b). This comparison excludes the surface 0- to 2- and 2- to 4-cm layer at Station 4 where fish food pellets formed a visible upper layer on the sediment. The high values of organic carbon (22.7 to 27.3%) and nitrogen (3.02 to 3.83%) (C:N atomic ratios 8.3 to 9.2) in this layer can be compared to values in dry food pellets fed to salmon in August 1990 at Site 6 (50.4% C and 8.5% N by weight; C:N atomic ratio of 6.92) (Hargrave et al. 1993a). Decomposition of organic matter in pellets plus mixing with surface sediments would decease the percentage content of organic carbon and nitrogen and increase the C:N atomic ratio. As expected from sediment grain size distributions, fine-grained deposits contained more organic carbon and nitrogen than coarser deposits. C:N atomic ratios (7 to 9)

38 indicate marine-derived freshly produced biogenic organic matter at all stations with no large contribution of terrigenous or marine macrophyte material which typically have C:N ratios > 10. The variability inC and N content, along with inorganic grain size distributions, indicates that organic matter accumulates with fine sediments at sites where hydrodynamic conditions allow sediment to deposit. The increment in mean values for C and N in surface sediments at cage sites over values at reference sites (+40%) reflects residual organic matter accumulated after decomposition. It represents organic matter remaining from sedimentation of waste food and feces and enhanced sedimentation of fine particulate matter from the water column through flocculation processes as discussed above.

BENTHIC ENRICHMENT INDEX The average BEl values for both cage and reference sites were positive (Fig. 2Q). Negative values appeared at four cage sites (Sites 1, 4, 5, and 9). Site 17 was the only reference site with a negative BEl value. Negative values of BEl are associated with areas receiving organic carbon sedimentation at rates > 1 g C m-2 d- 1 (Hargrave 1994b). BEl values calculated for Site 6 from measurements of Eh and organic carbon in surface (upper 1 em) sediments from under fish pens in 1990-1991 varied from -1750 to -2800, while positive values (84 and 268) were calculated from data collected in June and September, respectively, in this study. Values > + 1000 at the edge of the fish pen array and a proximate reference site in the earlier data set can be compared with the mean value ( +660) calculated for Reference Site 6A (Fig. 2Q).

BENTHIC OXYGEN UPTAKE AND CARBON DIOXIDE RELEASE Mean values for total benthic oxygen uptake and C02 release at cage sites were 175% and 200% higher than values at reference sites (Table 2; Fig. 2K and M). The range of values for oxygen uptake at cages sites (16 to 98 mmol m-2 d- 1) and at reference sites (10 to 42 mmol m-2 d-1) are similar to the range of rates measured under and adjacent to pens at Site 6 in 1990-1991 (Hargrave et al. 1993a). Between-site variation (coefficients of variation=olmean) (c.v.) for both oxygen and C02 flux was greater at cage sites (0.6 to 0.7) than at reference sites (0.5 to 0.6). Increased heterogeneity in benthic metabolism could reflect spatial patchiness of organic matter sedimentation of waste food pellets and feces under fish pens (Gowen et al. 1994; Hargrave 1994b; Silvert 1994). Measures of oxygen uptake and C02 release in the presence of HgC12, that represent chemical fluxes in the absence of biological activity, were higher at cage than at reference sites (Table 2; Fig. 2L and N) as observed for measures of total gas exchange; but the relative magnitude of enhancement was greater for chemical C02 release (+199%) than for chemical oxygen uptake (COD) (+64%). Variability in COD was lower (c.v. =0.33) than observed for total gas exchange, and values were similar at both cage and reference sites. COD is due in part to accumulated reduced inorganic end products such as sulfides produced during anaerobic metabolism

39 (Hargrave et al. 1993a). Anaerobic respiration would increase dissolved C02 in sediment pore water above concentrations in overlying seawater. Positive concentration gradients across the sediment-water interface would increase diffusion and result in higher C02 release rates. Benthic COD can be calculated as a percentage of total oxygen uptake to examine changes in the proportion of uptake sensitive to poisoning with HgC~. In an earlier study at Site 6 during 1990-1991, when totals= under fish pens reached maximum levels of 500 mM in August, COD accounted for a large proportion (40 to 100%) of total oxygen uptake (Hargrave et al. 1993a). In the present study, COD represented a smaller proportion of total oxygen demand at both cage (12 to 25%) and reference (19%) sites (Table 2). The lower value at Site 6 is consistent with the lower amounts of s= in sediments following the movement of fish pens in 1992 as discussed above.

MACROFAUNA BIOMASS The range of wet tissue biomass in benthic macrofauna observed in our study (6 to 2600 g m-2) spans previous observations at other locations in the Bay of Fundy (Hargrave et al. 1993a; Wildish et al. 1986; Wildish and Peer 1983). Total macrofauna biomass was approximately 63% higher at reference sites (mean of 473 g wet weight m-~ than at cage sites (mean of 291 g m-2) (Fig. 2R), but the difference was not significant (one-way ANOVA, F=0.373, p=0.548) due to the variance in the data (c.v. of 1.8 at cage and reference sites). In contrast, the average biomass of deposit feeders at cage sites (22 g m-2) was significantly higher (F=2.808, p=0.109) than the value for reference sites (13 g m-~ (Fig. 2S), indicating that high rates of organic matter loading associated with net-pens enhances food supply for this macrofauna trophic group. The biomass of suspension feeders was not significantly different at cage and reference sites (F=0.023, p=0.882) (Fig. 2R and U). Generally, reference sites had a greater diversity of major taxa (Table 4). Excluding polychaetes, 42 species were present at all reference sites, while 32 species occurred at cage sites. Reference sites averaged 4.36 major taxonomic groupings per station, compared to 4.00 for cage sites. There was considerable variation in macrofauna biomass between samples collected within one site due to the presence of large individuals including the echinoderms Cwncumaria frondosa and Strongylocentrus droebachiensis and the mollusc Modiolus modiolus in some samples. Variation between sites was highest for suspension feeders, reflecting the clumped distribution common for many of these species. The coefficient of variation was twice as high for suspension feeders at the cage sites (2.64 vs. 1.44 at reference sites), perhaps due to patchiness in sedimentary geochemical conditions and spatially non-homogeneous deposition of food pellets and fish fecal matter under net pens (Hargrave et al. 1993a). Values for c.v. for deposit feeder biomass between cage (0.70) and reference (0.64) sites were lower. Differences in physical factors such as current speed, modal grain size, oxygen availability, and organic loading result in aggregations on different scales in suspension and deposit-feeding benthic macrofauna.

40 It is therefore imperative that replicate sampling be carried out for accurate assessment of macrofauna biomass distribution.

The largest functional group was deposit feeders (27 species), followed by suspension feeders (18 species), predators (14 species), and omnivores (9 species). Herbivores (4 species) were poorly represented. Polychaetes, which in this case were mostly deposit feeders (Table 4), were the most abundant (13 species) and were present at all sites. The next most numerous and widely distributed group was bivalve molluscs (absent only at Sites 12 and 14), which are mostly suspension feeders. This distribution (deposit feeders> suspension feeders in numbers and distribution) corresponds with observations by Wildish and Peer (1983) found in the lower Bay of Fundy. Generally, biomass of deposit feeders was reduced at cage and reference sites (Sites 8, 9, 20, 21, and 22) where large modal grain sizes indicated coarse sediment (Fig. 2C). Sanders (1956) found that the biomass of suspension-feeding animals declined while that of deposit feeders increased, as the silt-clay content in the sediment increased. Deposit feeders were found to be associated with a fine-grained, net dispositional sediment, and suspension feeders with sandy sediments. However, recent evidence (Snelgrove and Butman 1994) suggests that sedimentary grain size is not the only or even primary determinant of infaunal species distribution. Grain size co-varies with sedimentary organic matter content, pore water composition, bacterial abundance, and species composition, all of which are influenced by near-bed flow regime. Bagarino (1992) contends that benthic macro-invertebrates inhabiting mud bottoms are more tolerant to high sulphide concentrations than those of hard or sand bottoms. For example, Shick (1976) found that Ctenodiscus crispatus from muddy bottoms survived hypoxia when exposed to 1.5 mM sulphide for 10 d, while Asterias vulgaris and A. forbesi from the rocky intertidal survived only 4 to 5 d. In our study, C. crispatus was present only at Station 4, a cage site with high sulphide concentrations (1.4 mM) and a small modal grain size (20 JLm). Asterias sp. was only present at Station 21, the reference site with the lowest sulphide concentrations (0.01 mM) and the largest modal grain size (450 JLm) measured in our study. Wildish and Peer (1983) and Wildish (1985) suggested that mutual exclusivity of deposit and suspension feeders arises because each trophic group tolerates a different range of tidal velocities and hence sediment-water interface dynamics. Deposit feeders are favoured at low current velocities and suspension feeders at higher current velocities where sediment deposition is low or absent. Accordingly, in our study, Site 2 which had the lowest current speed (gypsum dissolution index= 1. 00) had the highest biomass of deposit feeders (36.5 g m-2). Sites 9, 14, 20, and 21 had relatively fast current speeds (DI=2.96, 4.76, 5.49, and 3.94) which correlated with a higher biomass of suspension feeders (1008.1, 102.3, 296.7, and 433.2 g m-2 , respectively). It is possible that water movement due to wind, tide, and wave activity directly inhibit feeding by creating current velocity above a threshold speed (Wildish and Kristmanson 1979; Wildish and Kristmanson, unpubl.). Also, excessive water motion can cause resuspension of sediments which inhibit feeding (Winter 1973). This may explain the

41 low biomass of suspension feeders (1. 7 g m-2) at Site 22 despite the high current speed (DI=4.67). The absence of Capitella sp. at all cage sites sampled in 1994 contrasts past studies in the Western Isles region of the Bay of Fundy and other coastal areas (Hargrave et al. 1993a; Dubilier 1988; Grant et al. 1995). Cuomo (1985) showed that sulphide concentrations between 0.1 mM and 1 mM elicited optimal settlement, metamorphosis, and survival of Capitella sp. Du~ilier (1988) noted that no acute toxic effects, such as arrest of ciliary movement, occurred when Capitella was exposed to sulphide concentrations up to 2 mM. With sulphide concentrations > 2 mM, a high percentage of larva were adversely affected and prevented from settling quickly into the sediment. Generally, the disappearance of Capitella sp. indicates that sulphide levels are increasing. Although mean sulphide concentrations found at cage sites ranged from 0.4 mM to 3.4 mM throughout June and July (Fig. 2E), macrofauna samples were collected during the first 2 wk of August in 1994. Hargrave et al. (1993a) found that numbers of Capitella sp. under pens at Site 6 declined rapidly after early July 1990 when s= concentrations increased to > 10 mM. Thus, Capitella sp. would not be expected to be present in sediment with high sulphide levels. Pearson et al. (1983) proposed that the bivalve Nuculana tenuisulata could be used as an indicator species for organic matter loading beneath aquaculture sites. This deposit-feeding bivalve responds to an increased supply of sedimentary organic material. Results of our study indicate that N. tenuisulata was abundant at all cage sites, in response to increased organic loading. N. tenuisulata occurred at only four reference sites (Sites 17, 18, 19, and 20). In a recent study, Pocklington et al. (1994) found the small polychaete Nephtys neotena (formerly Aglaophamus neotenus) to be a similar indicator of organically enriched sediments under mussel lines in two embayments in eastern Canada.

WATER MOVEMENT MEASUREMENTS DI values for stations where gypsum blocks were exposed varied from 1.00 (Site 1) (dissolution rate equivalent to that expected in still seawater) to a maximum of 5.49 at Site 20 (Table 5). The average ( +SD) value for the index for all (n=9) cage sites (1.89+0.63) was lower than the corresponding value (n=8) for reference sites (3.32+ 1.66). DI is a dimensionless index requiring that the exposure temperature for gypsum cylinders be known. Seawater temperature at each site during exposure was known (or estimated) so that a blank weight loss could be determined for each site. Calibrations of DI values with actual current speed at different temperatures will be reported in Wildish et al., unpubl.). When these data are available, trophic group biomass response to different current velocities at different sites will be determined.

42 RELATIVE SENSITIVITY OF VARIABLES FOR DETECTING DIFFERENCES BETWEEN CAGE AND REFERENCE SITES Variable means at cage and reference sites (from Table 2) were compared (as maxima/minimum ratios) to determine their usefulness for detecting benthic enrichment at aquaculture sites (Fig. 3). Ratios for several variables at cage sites were more than three times greater than those at reference sites. Ranking of variables (Table 6) indicated that total sulfide was the most sensitive indicator for organic enrichment at cage sites, followed by Eh potentials and total C02 release. Mean values at cages sites for several other geochemical variables (ammonium measured by electrode in pore water, carbon burial rate, ammonium gradients and benthic enrichment index) were more than three times values at reference sites. These would also be sensitive indicators for detecting organic enrichment. With the exception of deposit feeders as a percent of total biomass, measures of biomass of trophic groups of macrofauna were less sensitive (cage: reference ratios < 2). Ratios close to 1 associated with sediment pore water (water content, salinity and sulfate) were the least sensitive indicators of differences between cage and reference sites. These variables would not be useful for monitoring benthic enrichment effects. Ratios of c.v. (cage:reference) for different variables were compared to determine if variance was homogeneous prior to using one-way ANOVA (Table 7). Ratios > 1 indicate that variance at cage sites was greater than that at reference sites, while ratios < 1 arise from lower relative variability at cage sites. The ratio was highest (8.51) for the benthic enrichment index, reflecting more a heterogeneous distribution of Eh potentials (ratio 3.95) in surface sediments. On the other hand, although average amounts of organic carbon were higher in surface sediments at cage sites (Table 6), concentrations were less variable than they were at reference sites (c.v. ratio 0.74). The high variability of values for BEl at cage sites probably reflects patchiness in sedimentation of uneaten food pellets and fish fecal waste products mentioned above. Combining measurements of Eh and organic carbon to calculate BEl provides a variable that is sensitive to heterogeneous sedimentary geochemical conditions at cage sites. Suspension feeder biomass was also more aggregated at cage sites than at references sites (c.v. ratio 1.83) (Table 7). This contrasts ratios for biomass of deposit feeders and total macrofauna (ratios 1.11 and 1.02, respectively). Values close to 1 indicate a similar degree of aggregation for these faunal groups at cage and reference sites. Ranking of probability values from one-way ANOVA tests for differences between variables at cage and reference sites showed highly significant differences (p::::J

14-16

C)

3: ....

-CG

C"'

w

...

16 -18

0

C)

0 ..J

18-20 20-22 22-24

:/\ 1

10

100

Diameter, IJm

24-26 26-28

147

Core 2-6

Core 2-5

~

~

Depth, em 0-2 2-4 4-6 6-8

_/\ 0~

.....c..

C)

·3: ....c (1)

~

(1)

C'"

w

0 'I""

C)

0 ..J

10 -12 12-14 14-16

-cu .:::::::s

8-10

/

16-18 18-20

~

~ __/\

20-22 22-24 24-26 26-28 28-30

1

:8:

10

100

Diameter, 1-1m

J\ ~

~

~ 1

10

100

Diameter, IJm

148

Core 4-5

Core 3-4

Depth, em 0-2 2-4 4-6 6-8 0~

.... ·~ .... c:

8 -10

""

.s::.

C)

Q)

10 -12 12-14

Q)

C'O

·-::l>

14-16

C"'

w

16 -18

0 or-

C)

0 ..J

18-20 20-22 22-24 24-26 26-28

1

10

100

Diameter, IJm

1

10

100

Diameter, 1Jm

149

Core 5-3

Core 6-9 Depth, em 0-2 2-4 4-6 6-8 8-10

0~

... ..... .s::

10 -12

C)

·-

12-14

..... s::::

14-16

(1)

~

-

(1)

C'CS

.~

16-18

:::l

CT

w

18-20

0 or-

C)

0 ..J

20-22 22-24

~

~

~

24-26

~

26-28

~

28-30 1

10

100

Diameter, J,Jm

1

10

100

Diameter, J,Jm

150

Core 7-5 Depth, em 0-2 2-4 4-6 6-8 8-10 0~

10 -12 12-14

...... c:

14-16

Cl)

co

·->::::l

16-18

cr

w

18-20 20-22 22-24 24-26 26-28 28-30

1

10

100

Diameter, IJm



1

10

100

Diameter, 1Jm

151

Core 9-8

Core 10-2 Depth, em 0-2 2-4 4-6 6-8

~ 0

.......

8-10

·-

10-12

.....r::::

12-14

-~ :::l

14-16

-'= C) (1)

~

-cu

(1)

C"

w

-

16-18

0

C)

0 ....1

18-20 20-22 22-24 24-26 26-28

1

10

100

Diameter, IJm

1

10

100

Diameter, IJm

152

Core 11-6

~

~

0~

...,;.t: C)

·3: ....c Q)

-cu

~ ~

C"

w

0

C)

0

..J

Depth, em 0-2 2-4 4-6 6-8 8-10

~

~

~

10-12 12-14 14-16

Q)

-~ ::J

Core 12-8

~

~ ~

~

~

1

10

100

Diameter, IJm

16 -18 18-20 20-22 22-24 24-26 26-28

..

28-30

1

10

100

Diameter, 1-1m

153

Core 13-6

Core 13-5

~

~ ~

...... ~ ·- ~ 3: ... ~

Depth, em 0-2 2-4

~

4-6

~

6-8 8-10

0~

.c:

10-12

C)

CD

12-14

c

14-16

-cu> CD

·-0'":::s w

~

16 -18 18-20

0 'I""

C)

0

...1

~

20-22

~

~ ~ ~

0

22-24 24-26 26-28 28-30 1

10

100

Diameter, 1Jm

1

10

100

Diameter, 1-1m

154

Core 14-8

~ 0~

....;-

.c

·-

C) Cl)

~ ...., c::

Cl)

C'CS

·-::s>

8 8 ~ ~

Depth, em 0-2 2-4 4-6 6-8 8-10 10 -12 12-14 14-16 16-18

C"

w

18-20

0 'r""

C)

0 ...1

20-22

~ ~

22-24 24-26 26-28 28-30

1

10

100

Diameter, IJm

155

Core 15-2

Core 15-3 Depth, em 0-2 2-4 4-6 6-8 8-10

0~

.....,...

10 -12

..c:

C)

·-

12-14

c:

Q)

14-16

-~ :::s

16-18

Q)

3: .....,

-co

C'"

w

-

18-20

0

C)

~

~

0

...J

20-22 22-24 24-26 26-28 28-30

1

10

100

Diameter, J.lm

1

10

100

Diameter, J.lm

156

Core 17-8

Core 16-2

Depth, em 0-2 2-4 4-6 6-8 0~

... .c

8-10

~

C)

·Cl)

10-12

3: ~

c:

-ns

12 -14

::l

14-16

Cl)

.~

C'"

w

16-18

0

""" C) 0

...J

18-20 20-22 22-24 24-26 26-28 1

10

100

Diameter, IJm

1

10

100

Diameter, IJm

157

Core 19-11

Core 18-8 Depth, em 0-2 2-4 4-6 6-8 8-10 0~

...,"'

10 -12

.r:.

C)

·-

12-14

Cl)

3: ...,

c: Cl) ca >

-

14-16

·-:::s

16-18

C"

w 0

'I""

C)"

0 ..J

~

18-20 20-22 22-24 24-26 26-28 28-30

1

10

100

Diameter, IJm

1

10

100

Diameter, J,Jm

158

Core 20-2

Core 21-5 Depth, em 0-2



.

2-4

..

4-6 6-8 0~

....."'

8-10

.s::

C)

·-

10 -12

s:::

12-14

Q)

3: .....

Q) C'CS

·->::::s

14-16

C'"

w

16 -18

0 'r'"

C)

0 ...1

18-20 20-22 22-24 t

24-26 26-28

1

10

100

Diameter, IJm

1

10

100

Diameter, 1Jm

159

Core 22-8

Depth, em 0-2 2-4 4-6 6-8 ~ 0

.:

8-10

J:

C)

·3: .....

10-12

(I)

c:

-cu

12-14

-~ :l

14-16

( I)

C"

w

16-18

0

""" C) 0 ..J

18-20 20-22 22-24 24-26 26-28

1

10

100

Diameter, 1Jm