Impact of new measured Mediterranean

mineralization rates on the fate of simulated

aquaculture wastes

Patrizia De Gaetano1, Paolo Vassallo2, Marco Bartoli3, Daniele Nizzoli3,

Andrea M. Doglioli4, Marcello G. Magaldi5, Mauro Fabiano2

1 DIFI, Dipartimento di Fisica, Università di Genova, Genoa, Italy.

2 DIPTERIS, Dipartimento per lo Studio del Territorio e delle sue Risorse, Università di Genova, Genoa, Italy.

3 DSA, Dipartimento di Scienze Ambientali, Università di Parma, Parma, Italy.

4 Laboratoire d'Oceanographie Physique et Biogeochimie LOPB-UMR 6535, CNRS, OSU/Centre d'Oceanologie de

Marseille, Aix-Marseille Universite, Marseille, France.

5 Division of Meteorology and Physical Oceanography (MPO), Rosenstiel School of Marine and Atmospheric

Science (RSMAS), University of Miami, FL, USA.

Corresponding author address:

Patrizia De Gaetano, DIFI, Dipartimento di Fisica, Università di Genova, Via Dodecaneso 33,

16146 Genova, Italy.

Phone: +39 010 353 6478 Fax: +39 010 353 6354 E-mail: degaetano@fisica.unige.it

Running title:new Mediterranean mineralization rates for FOAM

Keywords: Mediterranean Sea, mineralization rate, biodegradation modeling, aquaculture

impact, Sparus aurata, Dicentrarchus labrax.

Preprint submitted to Aquaculture Research February 16, 2009

1

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

1

Abstract

The results of two field campaigns performed in July and October 2006 in a typical

Mediterranean fish farm (Adriatic Sea) are introduced. The measured values are used to

simulate the fate of aquaculture wastes through FOAM module.

The campaigns allow accurate measurements of particulate carbon input and benthic

respiration rates. The experimental results indicate that carbon oxidation follows

different pathways which are temperature dependent.

FOAM is setup to simulate two different scenarios, cold and warm season. The results

are compared with the previous version of the module which used values measured

around salmon farms in the Atlantic coast. The new results show a smaller extension of

impacts compared to the Atlantic ones, due to a minor carbon input and to higher

mineralization rates. Results remain consistent with the previous version of the model

on two points: a) the negligible benthic impact of faeces with respect to uneaten feed;

and b) the impacts due to different feed release modes. They also indicate that the

extension of the impacted area and the organic carbon concentration on the seabed vary

seasonally.

The new setup based on Mediterranean methabolic values better reproduces the

nutrient load on the seabed and improves the overall FOAM performances.

2

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

2

1. Introduction

World aquaculture has been quickly growing during the last fifty years. It has passed

from a production of less than a million tonnes in the early 1950s to 59.4 million tonnes

by 2004. The average annual increasing rate is 8.8 %. Marine aquaculture represents

nowadays the 50.9 % of the total aquaculture yield (FAO Fisheries Department, State of

world aquaculture 2006 http://www.fao.org/ docrep/009/a0874e/a0874e00 ). Its

continuous expansion has been generating interest on predictive tools able to assess the

possible impacts for coastal ecosystems.

Indeed, several experimental studies have highlighted that particulate wastes

originated by marine fish farms are the main cause of environmental impact (Hall et al.,

1990; Holmer & Kristensen, 1992; Karakassis et al., 2000). Particulate products

increase the organic load on benthic environment and might result in changes in the

structure and functions of benthic communities (Tsutsumi et al., 1991; Wu et al., 1994;

Vezzulli et al., 2002, 2003; Pergent-Martini et al., 2006; Holmer et al., 2007; Hargrave

et al., 2008). However, dispersion pathways, settling and mineralization rates, burial

of organic particles for both uneaten feed and faeces, are all difficult processes to

follow. Even if the classical sampling approach remains essential for monitoring

purposes, a complete picture of the waste distribution in time and space can be achieved

with this method only with a huge and costly effort.

Therefore, the interest in tracking aquaculture wastes with mathematical models has

been rapidly increasing in time (Gowen et al., 1989; Gillibrand & Turrell, 1997;

Henderson et al., 2001). Cromey et al. (2002) developed a particle tracking model

including hydrographic data for modelling resuspension and changes in the benthic

faunal community. Hydrodynamic models of settling, resuspension and decay of net-

pen wastes coupled with transport models were also used for assessing the

3

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

3

environmental impacts of marine aquaculture (Panchang et al., 1997; Dudley et al.,

2000). Doglioli et al. (2004) took into account the three-dimensional ocean circulation

and its variability in tracking different aquaculture wastes developing the advection-

dispersion model POM-LAMP3D (Princeton Ocean Model – Lagrangian Assessment

for Marine Pollution Model). However, they did not consider the environmental

response to the organic load from the cages. De Gaetano et al. (2008) (hereinafter

referred to as DDMVF08) recently improved the predictive capability of POM-

LAMP3D coupling its outputs with the numerical benthic degradative module FOAM

(Finite Organic Accumulation Module). Following the work of Findlay & Watling

(1997; hereinafter referred to as FW97), FOAM utilizes the near bottom flow velocity

and the organic carbon flux to the sediment to estimate the degradation activity of the

benthic communities and the sediment state. With respect to the sediment state, FOAM

assigns a specific mineralization rate value and computes the net organic load

accumulating on the seabed.

The mineralization rates and the benthic metabolism activity represent key parameters

for the accuracy of the model prediction. Nevertheless, the lack in literature of values

specifically targeting Mediterranean conditions, obliged DDMVF08 to use the only

values available, i.e. the ones measured under Salmon rearing cages along the Maine

coast (FW97).

In this study, we want to fill this gap and compare the new prediction of the

environmental impact with the previous study results (DDMVF08). Therefore, the

original intent of this work is represented by coupling in situ measurements and

numerical experiments. Namely, in this study we measured the mineralization rates and

the benthic activity in Mediterranean conditions. We then used these values in new

FOAM simulations.

4

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

4

The experimental part of this study considers two sampling campaigns. They were

performed during the year 2006 in order to measure the biodepositon rates, background

sedimentation, benthic O2 respiration and CO2 production in a Mediterranean marine

fish farm. In particular i) we measured the carbon flux to the sediment and the surface

sediment respiration rate under floating cages for Sea Bass and Sea Bream rearing; ii)

we differentiated the sea floor into a three-state categorization according to measured

variations in the respiratory quotient and iii) we estimated per each sediment state, the

different rates of organic matter degradation through CO2 production.

In the modeling part of this study, we applied FOAM using the same advection-

dispersion simulations presented in a previous study (DDMVF08; the reader is referred

to the first part of Tab. 1 of DDMVF08 for the detail on the POM-LAMP3D

parameters). This decision was supported by two main reasons: a) the model setup

utilized in DDMVF08 reproduced very satisfactory both the hydrodynamics and the

dispersion in the investigated area; b) the primary focus of this study is the application

of specific Mediterranean degradative parameters in order to move toward a more

realistic scenario.

We present the results of the new implementation with Mediterranean benthic

parameters comparing them with the previous ones. Finally, we discuss our results in

the framework of the prediction of potential impacts associated with marine aquaculture

activities.

The rest of the paper is organized as follows. In Section 2 a description of the sampling

activities, core incubation, analytical methods and model setup are provided. The results

of the flux measurements and numerical experiments are presented in Section 3. A

discussion is given in Section 4 and finally, the conclusions are presented in Section 5.

5

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

5

2. Material and methods

2.1. Field experiment

Water and sediment samples were collected from the “Tortuga srl.” off-shore fish farm

located in the Manfredonia Gulf in the South Adriatic Sea (41°37’12 N, 15°56’55 E).

The sea cages are located at about 3 km from the coast on a water column ranging from

8 to 12 m. The farm is composed of 16 floating cages for the rearing of Gilthead Sea

Bream (Sparus aurata) and Sea Bass (Dicentrarchus labrax) and has a production of

about 650 ton year-1. For its features (reared species, farm dimensions and productive

target), this farm can be considered a typical Mediterranean off-shore fish farm. Indeed,

this site allows to measure, in a relatively small area, sediments characterized by quite

different loads of organic carbon and associated microbial activities. The expected large

range of C input, benthic O2 and CO2 fluxes allows in turn to calibrate the model on a

reliable data set including most of the typical situations of the Mediterranean area.

During the pre-survey activity carried out in May 2006, six sampling stations were

identified covering a broad range of conditions (i.e. heavily to moderately organic

loaded). In the same area, in July and October 2006, intensive sampling activities were

carried out where sedimentation and benthic respiration rates were measured. In

particular at the six sampling sites, three sediment traps (diameter 0.125 m,

diameter/height ratio > 3) along a single mooring block were deployed at 0.80 m from

the bottom. These traps were left in situ for 48 hours. After recovery, trap sample

portions were filtered on fiberglass membranes (Whatman GF/F diameter 25 mm,

nominal porosity 0.45 μm) by a vacuum pump, with pressure not exceeding 25 kPa to

avoid particles breaking on the filter and matter loss in the dissolved phase. In order to

determine the carbon flux to the sediment C , the recovered material was then

6

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

6

analyzed for particulate carbon by a CHN elemental analyzer (mod. CHNS-O EA 1108,

Carlo Erba).

Respiration rates and inorganic carbon production were measured by means of intact

cores incubation. To this purpose, sediment cores (i.d. 0.08, length 0.40 m, n = 4) were

collected in all the sampling stations by scuba divers. All cores were brought to the

laboratory within a few hours from sampling for further processing and incubation

procedures, described in detail in Dalsgaard et al. (2000). The day after the sampling,

the water in the tank was exchanged and the cores were incubated in the dark for

measuring dissolved oxygen flux O2 and total inorganic carbon flux CO2 .

Water samples were collected at regular time intervals with plastic syringes. Samples

for O2 determinations were transferred to glass vials (Exetainers, Labco, High

Wycombe, UK) and Winkler reagents were added immediately (Strickland & Parsons,

1972). Samples for CO2 were also transferred to glass vials and immediately titrated

with 0.1 N HCl following the Gran procedure (Anderson et al., 1986).

2.2. FOAM module setup

As in DDMVF08, the computation of oxygen demand is based on the relationship

between the measures of oxygen consumption and organic carbon sedimentation.

Therefore, FOAM computes the sediment oxygen demand (O2dem) as a function of the

organic carbon flux toward the sea bottom (FlxBot) according to the relation:

O2dem = C FlxBot D (1)

7

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

7

The values of the C and D constants are here obtained by linear regression of the

measured Mediterranean relationship between dissolved oxygen and the carbon flux

toward the sediment.

On the basis of the ratio between the oxygen supply (the reader is referred to

DDMVF08 for details) and the oxygen demand (O2dem), FOAM assigns three different

states to the sediment. At each sediment state corresponds a different mineralization

rate. In the present work, the mineralization rates are based on experimentally calculated

respiratory quotients. These quotients are the ratio between measured total inorganic

carbon flux and dissolved oxygen flux, as suggested by Dilly (2003) and Hargrave

(2008):

RQ =CO2

O2(2)

Calculated RQs are grouped in the following categories of sediment state:

if RQ < 0.8 corresponding to an elevated re-oxidation of end-products

coming from anaerobic metabolism. This is generally coupled to limited

organic load;

if 0.8 < RQ < 1.2 corresponding to an intermediate condition, which is

generally coupled to moderate organic load;

if RQ > 1.2 corresponding to an accumulation of end-products coming

from anaerobic metabolism. This is generally coupled to elevated organic

load.

The mineralization rates associated with the three categories are then calculated as the

average of measured CO2 within each category. For the sake of clarity in the

8

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

8

comparison with previous paper (DDMVF08), in the following we adopt the same

notation for the mineralization rates: N s for , I s for and H s for

category. Finally, FOAM computes the net organic carbon concentration remaining on

the seabed, subtracting these three mineralization rates from the organic carbon

concentration accumulated on the sea bottom.

Summarizing, the new in situ measurements provide to FOAM the values of

parameters C and D and the estimate of the three mineralization rates N s , I s and

H s .

All parameters employed in the benthic module are reported in Table 1.

3. Results

3.1. Experimental field results

Carbon flux to the sediment C , dissolved oxygen flux O2 and total inorganic

carbon flux CO2 range from 1.3 to 107.0, 5.8 to 194.2 and 0.3 to

242.3 mmol m−2 d−1 respectively.

In Figure 1, we show the relationship between O2 and C measured in July

(empty dots) and in October (full dots). Both sampling campaigns show a positive

relationship between O2 and C . This relationship has been found to be dependent

on water temperature. In July (October) experiment, when the water temperature is

around 27°C (18°C), the molar ratio is 1.4 (0.4) and the y-intercept is 54.5 (31.8). These

values are adopted for the parameters C and D in Eq. (1).

In Figure 2A (B), the data of CO2 measured in July (October) are reported in

relation to O2 . These data are grouped according to the three categories of sediment

9

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

9

state, based on the RQ. In July, the measured CO2 ranges from 3.6 to 87.2, 56.2 to

80.2 and 140.9 to 242.3 mmol m-2 d-1 for , and categories of sediment state,

respectively. In October, the measured CO2 ranges from 0.3 to 94.1, 20.7 to 66.2 and

44.2 to 141.8 mmol m-2 d-1. In July, the measured O2 ranges from 15.3 and 194.2,

50.3 and 93.0, 27.5 and 146.6 mmol m-2 d-1 for , and categories of sediment

state, respectively. In October, the measured O2 ranges from 25.3 and 141.8, 20.7

and 74.3, 33.9 and 96.7 mmol m-2 d-1.

The FOAM parameters N s , I s and H s are calculated as the mean values of

CO2 for each sediment state and for each season (big crosses in Figure 2).

The values obtained are reported in Table 1.

3.2. Numerical results

In the light of the obtained experimental results, we decide to simulate two macro-

periods: the warm and the cold season using parameter values obtained from July and

October in situ experiment, respectively (Table 1).

Then, we perform the same numerical experiments as in DDMVF08, namely:

A1) slowly sinking feed in continuous release;

A2) quickly sinking feed in continuous release;

B1) slowly sinking feed in periodical release;

B2) quickly sinking feed in periodical release;

C1) slowly sinking faeces (continuous release);

C2) quickly sinking faeces (continuous release).

Periodical release means that the feed is supplied twice a day. Slowly and quickly

sinking are referred to the minimum and maximum values of settling velocity for the

10

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

10

two different waste typologies: uneaten feed and faeces (the reader is referred to

DDMVF08 for details).

For the two macro-periods, the simulation results are presented in relation to i) the

extension of the impacted area, defined as the whole area where particles are still

present even after the benthic degradation activity; ii) the organic carbon concentration,

defined as the carbon quantity per impacted square meter of sediment and iii) the

occurrence of the different categories of sediment states.

For a direct comparison, the tables (Table 1, Table 2 and Table 3) show our results

together with the DDMVF08 ones.

In general, the impacted area is always smaller than in DDMVF08. In the warm season

smaller impacted areas are registered with respect to the cold one (Table 2). A

negligible impact is always associated to fish faeces compared to uneaten feed. For the

slowly sinking feed particles both in periodical and in continuous release (Exp A1 and

B1), the impacted area is always larger than for the quickly sinking ones. The slowly

sinking feed particles released in the periodical mode (Exp. B1) results in the widest

impacted area.

In all scenarios, the faeces result completely degraded (Table 3). The slowly sinking

feed particles released in periodical mode (Exp. B1) results in the smallest organic

carbon concentration. The quickly sinking feed particles release in the periodical mode

(Exp. B2) results in the greatest organic carbon accumulation. The predicted organic

carbon concentration due to the uneaten feed in the warm season is always smaller than

in the cold season. Moreover, with the exception of the quickly sinking feed particles

released in the periodical mode (Exp. B2), the organic carbon concentration in the cold

season is greater than in DDMVF08.

11

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

11

As expected in the above-mentioned results, for C1 and C2 experiments the sediment

is practically always in the category (Table 4). Considering the uneaten feed, in the

cold season, the day occurrence of and sediment states is lower than in the

other scenarios. While in the warm season, the and states are higher. In all

simulated scenarios, the maximum occurrence of the sediment state is due to the

feed released in periodical mode (Exp. B1 and B2).

4. Discussion

The values of carbon flux to the sediment measured using sediment traps in our study

range between 1.3 and 107.0 mmol m-2 d-1 and include the range reported for other

Mediterranean areas with cage fish farming (1-45 mmol m-2 d-1 Pusceddu et al., 2007).

Despite significantly higher than background sedimentation, such organic C flux to the

sediment is anyway moderate and does not result in excess oxygen demand at the

benthic level. Respiratory quotients calculated from O2 and CO2 fluxes are in fact quite

below those reported for more sheltered and organic rich Mediterranean coastal lagoons

where anaerobic mineralization rates play a major role in the overall microbial

respiration activity (Bartoli et al., 2005).

One of the main assumptions behind the calculation, which is the linearity between

oxygen demand and carbon input, is thus likely to be satisfied. The previous version of

the module (DDMVF08) used Atlantic data, where organic carbon flux is significantly

higher and range between ~80 to ~550 mmol m-2 d-1 (FW97). This is not surprising as

Mediterranean fish farms are generally smaller in comparison with the Atlantic ones.

Moreover, as we have showed that fish feaces give a negligible contribution to the

overall sedimentary organic load, it is likely that salmon farming requires a greater

12

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

12

input of fish feed compared to Sea Bream and Sea Bass. Differences in the carbon flux

can be also due to the employed feed: the new generation feed has, in fact, increased

floating properties (Vassallo et al., 2006), allowing fish to better exploit the provided

pellets and therefore minimizing wastes. Despite the fluxes of organic matter to the

sediment tend to be lower than in FW97, experiments A1 and A2 indicate, for more

than 50% of the simulated period, a moderate to elevate level of stress of the benthic

compartment ( and sediment state categories coupled to moderate and elevate

organic load, respectively). This is possible because the study area, near to the coast and

relatively shallow, is characterized by weak currents regime which means for our model

a moderate supply of oxygen to counterbalance respiratory needs.

The relationship between the carbon flux to the sediment and the oxygen consumption

found by FW97 is here re-evaluated for the Mediterranean aquaculture average

conditions. We find that the temperature variations drive this relationship. Two different

linear trends are assessed during July and October sampling campaigns. The July trend

implies greater values of O2 than in FW97, while the October one shows lower. This

is not surprising and has been recently reviewed by Glud (2008), who shows how

measurements of oxygen demand coupled to carbon input to the sea floor are still very

limited. He also underlines the need for seasonal studies, as it is likely that carbon

sedimentation and oxygen respiration rates are probably partially uncoupled. Organic

carbon sedimentation, measured via traps, is in fact a sort of instantaneous measurement

whilst oxygen demand reflects and integrates the occurrence of processes with a longer

time scale (Glud, 2008). With equal carbon flux to the sediment and water circulation

regime, the higher oxygen demand of warm season forces the model towards a larger

occurrence of the intermediate conditions ( category of sediment state) compared to

13

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

13

DDMVF08. The opposite is observed in cold season when the occurrence of an elevated

re-oxidation of anaerobic metabolism end-products ( category) increases.

The mineralization rates result higher than what previously assessed by FW97 apart

from the cold season intermediate organic load state that shows lower ability to degrade

organic load. Moreover, the mineralization rates measured in July are the highest. This

gives rise to higher degradation and consequently lower organic accumulation on the

bottom and smaller impacted area than in the other scenarios.

As in DDMVF08 and Doglioli et al. (2004), we perform a qualitative comparison

between simulation results and in situ measurements in the four stations around the farm

(for details on the exact location of the sampling stations and sampling methodology,

the reader is referred to Doglioli et al., 2004). As showed in Fig. 3, the carbon flux

toward the sediment is highest in station S2 and lowest in S4 in agreement with the

westward transport characterizing the studied area. The higher degradation rates of the

warm season setup result in the lowest values of organic matter flux to the bottom

(continuous black line). On the other hand, the cold season setup (dashed black line)

shows highest carbon fluxes only in the stations with the greatest organic accumulation

(S1 and S2). These trends seem to be in a better agreement with the measured pattern of

accumulation with respect to DDMVF08 (dotted gray line).

Therefore, the new parameters allow a better discrimination between the two macro-

periods evidencing the temperature-dependence of the processes involved. This

dependence influences also the impact typologies: the warm season is characterized by

a more frequent occurrence of moderate and elevate organic load and by a higher

degradation causing lower organic carbon concentration on the sea bed and smaller

impacted area. Managers and policy makers may take care of these differences in

planning the installation of new fish-farms or the expansion of existing ones. A balance

14

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

14

among the organic matter spread or load and the occurrence of different sediment states

may be accurately evaluated with the application of the model.

5. Conclusions

The new FOAM simulations are shown to better reproduce the nutrient load on the

seabed in the investigated area than in DDMVF08. The new setup which is based on the

Mediterranean benthic metabolism, leads to more realistic results. The obtained results

confirm what already assessed in DDMVF08: the feed particles are the primary cause of

ecological impact to the benthos community and the more frequent conditions of high

sediment stress level and the higher values of organic carbon concentration on the

seabed are due to the feed particles released in periodical mode (twice in a day).

Furthermore, the present study highlights that water temperature is a key parameter for

the simulation of the responses of the marine environment to organic load. In fact,

during warm season the sediment is more likely to show the occurrence of disturbed

states. However the impact, in terms of concentration and extension, is reduced thanks

to the increased ability to mineralize organic loads (resilient ecological system). On the

contrary, in the cold season, undisturbed sediment states and a lower ability to

mineralize organic matter are registered. This leads to higher concentrations and higher

impacted areas on the benthos (resistant ecological system).

Further investigations are necessary to systematically characterize the Mediterranean

fish farms. This can be achieved applying the entire model POM-LAMP3D and FOAM

to several sites. Moreover, several sampling campaigns should be carried out in order to

obtain a whole range of variations of FOAM parameters with temperature and improve

the prediction capability of the model.

15

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

15

Acknowledgements

The authors thank Roberto Festa for his useful advices and Corrado Ratto for his

assistance and his interest in our research. More thanks to the Staff of “Tortuga srl.” for

their collaboration, Federcoopesca and Cecilia Silvestri (APAT) for the environmental

data and their support. Moreover we warmly thank the EcoTechSystem srl, Spin-off of

the Polytechnic University of Marches, for the sediment traps.

CINFAI (National Consortium of Italian Universities for Physics of Atmospheres and

Hydrosphere) is gratefully acknowledged to support the activities of the Physics of the

Atmosphere and Ocean Group - Department of Physics - University of Genoa (Italy).

References

Anderson L.G., Hall P.O.J., Iverfeldt A., Van Der Loeff M.M.R., Sundby B. &

Westerlund S.F.G. (1986). Benthic respiration measured by total carbonate production.

Limnology and Oceanography, 31, 319-329.

Bartoli M., Nizzoli D., Naldi M., Vezzulli L., Porrello S., Lenzi M. & Viaroli P. (2005).

Inorganic nitrogen control in wastewater treatment ponds from a fish farm (Orbetello,

Italy): denitrification versus Ulva uptake. Marine Pollution Bulletin, 50, 1386-1397.

Cromey C., Nickell T., & Black K. (2002). DEPOMOD-modelling the deposition and

the biological effects of wastes solids from marine cage farms. Aquaculture, 214(1-4),

211-239.

16

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

16

Dalsgaard T., Nielsen L., Brotas V.P., Underwood G., Nedwell D., Sundbck K.,

Rysgaard S., Miles A., Bartoli M., Dong L., Thornton D., Ottosen L., Castaldelli G. &

Risgaard-Petersen N. (2000). Protocol handbook for NICE-Nitrogen cycling in

estuaries: a project under the EU research programme. Marine Science and Technology

(MAST III), 62National Environmental Research Institute, Silkeborg, Denmark.

De Gaetano P., Doglioli A.M., Magaldi M.G., Vassallo P. & Fabiano M. (2008).

FOAM, a new simple benthic degradative module for the LAMP3D model: an

application to a Mediterranean fish farm. Aquac. Res., 39, 1229-1242.

Dilly O. (2003). Regulation of the respiratory quotient of soil microbiota by availability

of nutrients. FEMS Microbiology Ecology, 43, 375-381.

Doglioli A.M., Magaldi M.G., Vezzulli L. & Tucci S. (2004). Development of a

numerical model to study the dispersion of wastes coming from a marine fish farm in

the Ligurian Sea (Western Mediterranean). Aquaculture, 231(1-4), 215-235.

Dudley R., Panchang V., & Newell C. (2000). Application of a comprehensive

modeling strategy for the management of net-pen aquaculture waste transport.

Aquaculture, 187, 319-349.

Findlay R. & Watling L. (1997). Prediction of benthic impact for salmon netpens based

on the balance of benthic oxygen supply and demand. Mar. Ecol. Prog. Ser., 155, 147-

157.

17

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

17

Gillibrand P. & Turrell W. (1997). The use of simple models in the regulation of the

impact of fish farms on water quality in Scottish sea lochs. Aquaculture, 159, 33-46.

Gowen R., Bradbury N., & Brown J. (1989). The use of simple models in assessing two

of the interactions between fish farming and marine environment. In: Aquaculture – A

Biotechnology in Progress (ed. by N. DePauw, E. Jaspers, H. Ackefors &N.Wilkins),

pp.1071-1080. European Aquaculture Society, Bredene, Belgium.

Guld R.N. (2008). Oxygen dynamics of marine sediments. Marine Biology Research,

4, 243-289 .

Hall P., Anderson L., Holby O., Kollberg S. & Samuelsson M. (1990). Chemical fluxes

and mass balances in a marine fish cage farm. i. carbon. Mar. Ecol. Prog. Ser., 61, 61-

73.

Henderson A., Gamito S., Karakassis I., Pederson P.& Smaal A. (2001). Use of

hydrodynamic and benthic models for managing environmental impacts of marine

aquaculture. J. Appl. Ichthyol., 17, 163-172.

Hargrave B.T., Holmer M. & Newcombe C.P. (2008). Towards a classification of

organic enrichment in marine sediments based on biogeochemical indicators. Marine

Pollution Bulletin, 56(5), 810-824.

18

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

18

Holmer M. & Kristensen E. (1992). Impact of marine fish cage farming on sediment

metabolism and sulfate reduction of underlying sediments. Mar. Ecol. Prog. Ser., 80,

191-201.

Holmer M., Marbà N., Diaz-Almela E., Duarte C.M., Tsapakis M. & Danovaro R.

(2007). Sedimentation of organic matter from fish farms in oligotrophic Mediterranean

assessed through bulk and stable isotope (δ13C and δ15N) analyses. Aquaculture, 262,

268-280

Karakassis I., Tsapakis M., Hatziyanni E. Papadopoulou K. & Plaiti W., 2000. Impact

of cage farming of fish on the seabed in three Mediterranean coastal areas. ICES J.

Mar. Sci, 57(5), 1462-1471.

Panchang V., Cheng G., & Newell C. (1997). Modeling hydrodynamics and aquaculture

waste transport in Coastal Maine. Estuaries, 20, 14-41.

Pergent-Martini C., Boudouresque C.F., Pasqualini V. & Pergent G. (2006). Impact of

fish farming facilities on Posidonia oceanica meadows: a review. Marine Ecology, 27,

310-319.

Pusceddu A., Fraschetti S., Mirto S., Holmer M. & Danovaro R. (2007). Effects of

intensive mariculture on sediment biochemistry. Ecol. Appl., 17, 1366-1378.

Strickland J.D. & Parsons T. (1972). A practical handbook of seawater analysis.

Bulletin of Fisheries Research Board of Canada, 1672nd ed.

19

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

19

Tsutsumi H., Kikuchi T., Tanaka M., Higashi T., Imasaka K & Miyazaki M. (1991).

Benthic faunal succession in a cove organically polluted by fish farming. Marine

Pollution Bulletin, 23, 233-238.

Vassallo P., Doglioli A., Rinaldi F. & Beiso I. (2006). Determination of physical

behaviour of feed pellets in Mediterranean water. Aquac. Res., 37(2):119-126.

Vezzulli L., Chelossi E., Riccardi G. & Fabiano M. (2002.) Bacterial community

structure and activity in fish farm sediment of the Ligurian Sea (Western

Mediterranean). Aquacult. Int., 10(2), 123-141.

Vezzulli L., Marrale D., Moreno M. & Fabiano M. (2003). Sediment organic matter and

meiofauna community response to long-term fish-farm impact in the Ligurian Sea

(Western Mediterranean). Chem. Ecol., 19(6), 431-440.

Wu R., Lam K., MacKay D.W., Lau T.C. & Yam V. (1994). Impact of marine fish

farming on water quality and bottom sediment: a case study in the sub tropical

environment. Marine Environmental Research, 38, 115-145.

20

417

418

419

420

421

422

423

424

425

426

427

428

429

430

20

List of Figures

1. Relationships between the carbon flux to the sediment C and the benthic

metabolism as measured by O2 consumption O2 . Solid black line identifies

the general trend of October samplings (●); dashed black line identifies the

general trend of July samplings (○) and dotted gray line follows the trend

assessed by Findlay & Watling (1997)

2. Relationship between measured dissolved oxygen flux O2 and total

inorganic carbon flux CO2 . Solid and dashed lines identify the threshold of

limited and elevated organic load to the sediment (RQ < 0.8 and RQ > 1.2),

respectively. A) July data; B) October data.

3. Daily organic carbon loading rate in the four sampling stations around the fish

farm. Data from sampling observation are reported in gC kg-1 d-1 and are

referred to 2000-2005 (light gray bar). Results obtained by FOAM simulations

with the three applied settings are reported in gC m-2 d-1.

21

431

432

433

434

435

436

437

438

439

440

441

442

443

444

21

22

Fig 1

22

23

Fig 2

23

24

Fig 3

445

24

List of Tables

1. Input parameters for FOAM module.

2. Time-averaged impacted area at the bottom for the different experiments

considering warm season, cold season and DDMVF08 parameterization.

3. Time-averaged of the organic carbon concentrations remaining at the bottom

after degradation for the different experiments considering warm season, cold

season and DDMVF08 parameterization.

4. Time-occurrence of the benthic states for the different experiments considering

warm season, cold season and DDMVF08 parameterization.

25

446

447

448

449

450

451

452

453

454

25

FOAM parameters

value

Warm

season

Cold

season

Physical domain (km) (same as in DDMVF08) 8x4

Horizontal resolution (m) (same as in DDMVF08) 40x20

O2 supply parameter A (mmolO2 m-2 d-1) (same as in

DDMVF08)736.3

O2 supply parameter B (mmolO2 s m-3 d-1) (same as in

DDMVF08)672.5

O2 demand parameter C (mmolO2 mmolC-1) 1.4 0.4

O2 demand parameter D (mmolO2 m-2 d-1) 54.5 31.8

Carbon average mineralization rate in the

sediment state N s (mmolC m-2 d-1)45.1 ± 26.8 31.1 ± 27.7

Carbon average mineralization rate in the

sediment state I s (mmolC m-2 d-1)68.2 ± 17.0 47.8 ± 21.3

Carbon average mineralization rate in the

sediment state H s (mmolC m-2 d-1)188.6 ± 47.1 89.9 ± 34.9

I range amplitude parameter Δfw (same as in

DDMVF08)0.5

Table 1

2626

IMPACTED AREA

Exp. Simulation

Typology

Warm season Cold season DDMVF08

(release) mean ± std

(m2)

mean ± std

(m2)

mean ± std

(m2)

A1Slow feed

(continuous)3393 ± 419 3521 ± 540 3576 ± 582

A2Quick feed

(continuous)3200 ± 22 3202 ± 35 3202 ± 41

B1Slow feed

(periodical)4118 ± 605 4451 ± 542 4513 ± 563

B2Quick feed

(periodical)3245 ± 203 3268 ± 251 3277 ± 266

C1 Slow faeces <1 51± 200 377 ± 656

C2 Quick faeces 5 ± 59 342 ± 525 941 ± 962

Table 2

2727

ORGANIC CARBON CONCENTRATION

Exp. Simulation

Typology

Warm season Cold season DDMVF08

(release) mean ± std

(gC m-2)

mean ± std

(gC m-2)

mean ± std

(gC m-2)

A1Slow feed

(continuous)1301 ± 357 1464 ± 408 1450 ± 404

A2Quick feed

(continuous)1353 ± 417 1551 ± 464 1490 ± 453

B1Slow feed

(periodical)1075± 331 1226± 372 895 ± 380

B2Quick feed

(periodical)1405 ± 340 1560 ± 379 1590 ± 387

C1 Slow faeces < 1 < 1 < 1

C2 Quick faeces < 1 < 1 < 1

Table 3

2828

29

OCCURRENCE OF DIFFERENT SEDIMENT STATES

Exp. Simulation

Typology

Warm season Cold Season DDMVF08

(release) categories

(% days)

categories

(% days)

categories

(% days)

A1Slow feed

(continuous)49 44 7 94 5 1 74 22 4

A2Quick feed

(continuous)43 52 5 96 3 1 71 27 2

B1Slow feed

(periodical)85 4 11 89 8 3 87 4 9

B2Quick feed

(periodical)86 4 10 90 6 4 88 4 8

C1 Slow faeces 100 0 0 100 0 0 99 0 1

C2 Quick faeces 100 0 0 99 1 0 99 0 1

Table 4

29