Nofima Atlantic Salmon in Closed-Containment Systems Research Update

28.10.2014 1AIW#6

Bendik Fyhn TerjesenNofima

Aquaculture Innovation Workshop #6Vancouver, October 27th-28th 2014

�Current Atlantic salmon production, challenges, and trends in Norway

�Research on postsmolt production in RAS

�Effects of tank scale on postsmolt production (in flow-through)

�Research on postsmolt production in floatingsemi-closed containment systems in sea

Contents

AIW#6

Norwegian salmon farming

� Norways 2nd largest export commodity ~ 7 billion US$ annual value

� Norway exports salmon to approximately 100 countries� Produced around 1.3 million ton salmonids in 2013,

14 million meals every day� The industry employ > 20 000 persons in Norway

28.10.2014 3AIW#6

Norwegian salmon farming in 2050

28.10.2014 4

Envisioned value generation provided limiting factors for growth in the aquaculture value chain are addressed. Value estimates from Olafsen et al (2012)*. 1 US$ ~ 6 NOK

� 5 million ton salmonproduced annually i 2050?

� It has been projected thatNorway will produce 5 mill. ton salmon annually in 2050 (Olafsen et al 2012)*

� 5x increase in productionvolume, 8x in total valuecreation

� Large increases in relatedindustries, such as water treatment technologies

� This projection assumes thatsustainability issues areadressed and solved

AIW#6

* Olafsen, T., et al., 2012. Value creation from productive seas in 2050. Report from The Royal Norwegian Society of Sciences and the Norwegian Council of Academies of Engineering and Technological Sciences, Trondheim, Norway, 79 p.

Challenges for Norwegian salmon farming

• Sea lice is a serious challenge, for wild and farmed salmon• Escapes of farmed salmon may affect wild salmon populations• New sustainable feed resources must be found• Loss of fish: Mortality during production is 16.4% on average• Most of this loss, and lost production capacity, occur soon after

stocking at sea, of small smolts (<100 g)

• Huge efforts and R&D are being done to combat lice; everythingfrom laser canons against lice, cleaner fish, breeding lice-resistantsalmon, to submerged cages, tarpaulins, and closed-containmentsystems

10/28/2014 5

Photo: Stingray

AIW#6

Research on postsmolt production in RAS

28.10.2014 test 6

Smolt prod in closed RAS land-based

Postsmolts in semi-closedcontainment

in sea

Shorter periodin cages

before harvest

Semi-closedcontainmentin sea

0-80 g 80 g – 1 kg 1 kg – 5 kg

Smolt prod in closed RAS land-based

Postsmolt in closed RAS land-based

Shorter periodin cages

before harvest

0-80 g 80 g – 1 kg 1 kg – 5 kg

Closed-containmenton land

Today: 22 000 tons in closedsystems in Norway, as smolt

If all salmon in Norway to 1 kg in closed systems: 290 000, 13x increase!

Future production strategies in Norway?

AIW#6

Experimental series to increaseknowledge about postsmolt production in RAS

Bio-reaktor CO2

Filter

To tanks flow-through, full SW

Experiment 1

RAS medium-scale, varying salinity, varying water velocity

Bio-reaktor CO2

Filter

To cages in sea

Experiment 2

RAS industrial scale, varying salinity, fixed water velocity

Bio-reaktor CO2

Filter

To cages in sea

Experiment 3

RAS medium-scale, fixed salinity & water velocity

AIW#6

Exp 1. Salinity & water velocity in RAS for postsmolts

• SW-RAS may have higher running costs than FW-RAS due to CO2 and TAN removal efficiencies are lower than in FW

• Results in need for largerinstallations and/or higher flow

• Or can postsmolts be kept at lower salinity in RAS, and still handle full-strength SW at stocking in sea?

• This exp ran from 70 g til 850 g at different salinities (12-22-32 pptS) or exersize levels (0.3 og 1 BL/s), at12L:12D photoperiod

10/28/2014 9

Nofima Centre for Recirculation in Aquaculture (NCRA), Sunndalsøra, Norway

AIW#6

Salinity, ppt S

12 22 32

The

rmal

gro

wth

coe

ffici

ent

0.0

1.6

1.8

2.0

2.2

Factor salinity, p<0.01

Factor exercise: p<0.01

Exercise (1.0 Bl/s)

No exercise (0.3 Bl/s)

Exp. 1: Also salmon benefits from less salt and more exercise: Improved growth

12 ‰ salinity and 1 Bl/s exercisemade this fish reach 850 g/indfirst

Terjesen et al., upublisert, 2013AIW#6

Exp. 1: Improved feed utilization at lower salinities(FCR, feed:gain)

28.10.2014 11

Salinity (ppt)

12 22 32

FC

R (

fee

d/g

ain

)

0.00

0.80

0.85

0.90

0.95

1.00

1.05

1.10ExerciseNo exercise Salinity:p=0.071

Exercise: NS

Salinity (ppt)

12 22 32

FC

R (

fee

d/g

ain

)0.000.80

0.85

0.90

0.95

1.00

1.05

1.10

1.15

1.20ExerciseNo exercise

FCR entire trial (70-800 g) FCR 250-450 g

Salinity:p<0.005Exercise:NS

AIW#6Terjesen et al., upublisert, 2013

Exp.1: Improved survival of low salinities, when kept in similar conditions throughout

Terjesen et al., unpublished, 2013

AIW#6

Fish weight (g)

0 200 400 600 800

% C

um

mu

lati

ve s

urv

ival

0

50

60

70

80

90

100

12‰ + Exercise22‰ + Exercise32‰ + Exercise12‰ - Exercise22‰ - Exercise32‰ - Exercise

12‰ salt, 99%22‰ salt, 97%

32‰ salt, 71%

Mean survival in cages

Sensitive period

28.10.2014 Havbruk 2014, Tromsø

Exp.2: Industrial scale RAS; effects of salinity at fixed water velocity, with stocking in cages

400

m3 700

m3

Drum-

filterMBBR

FW RAS

>400’

smolt

12ppt RAS

51’ post-

smolt, 81g

12ppt RAS

61’ post-

smolt, 92 g

12ppt RAS

50’ post-

smolt

12ppt RAS

50’ post-

smolt

12ppt RAS

50’ post-

smolt

12ppt RAS

50’ post-

smolt

12ppt RAS

98’ post-

smolt

22ppt RAS

49’ post-

smolt, 78g

22ppt RAS

62’ post-

smolt, 100g

22ppt RAS

50’ post-

smolt

22ppt RAS

50’ post-

smolt

22ppt RAS

50’ post-

smolt

22ppt RAS

50’ post-

smolt

22ppt RAS

103’ post-

smolt

SW

116’ post-

smolt, 110g

SW

114’ post-

smolt

SW

99’ post-

smolt

SW

116’ post-

smolt

Industrial scale RAS: GSF Adamselv

Control cages in sea: GSF Loppa (90 m circumsphere)

Cages in sea:

GSF Loppa

(140m

circum-

sphere)

SW

101’ post-

smolt

SW

101’ post-

smolt, 75g

80-110g ~200 g ~300 g

Exp. 2 design: effects of salinity in industrial scale RAS, Grieg SeaFood

Frode Mathisen, unpub., 2013

AIW#6

FHF Postsmolt E project (Medhus, Norheim, Norwegian Veterinary Institute):� Detected nephrocalcinosis and red blood cells in kidney tubuli, increasingly with

duration of experiment

12ppt 22ppt

Exp.2: Industrial scale RAS; effects of salinity at fixed water velocity

Kolarevic, Terjesen et al., unpubl., 2013

Dato

12.a

ug.1

3

26.a

ug.1

3

09.se

p.13

23

.sep.

13

07.o

kt.13

21

.okt.

13

04.n

ov.1

3

18.n

ov.1

3

[CO

2] (

mg/

l)

0

5

10

15

20

25

30

35

40

Dato

12.au

g.13

26

.aug.

13

09.se

p.13

23

.sep.

13

07.o

kt.13

21

.okt.

13

04.n

ov.1

3

18.n

ov.1

3

[CO

2] (

mg/

l)

0

5

10

15

20

25

30

35

40Avg. [CO2] 13 ± 5 mg/L Avg. [CO2] 18 ± 7 mg/L

Yersinia

introduced by

accident to

RAS

Yersinia

introduced by

accident to RAS

� Detected Yersinia ruckeri at last sampling point� Detected Tenacibaculum sp.

AIW#6

Period

Aug-O

kt

Okt-Nov

Aug N

ov

Th

erm

al g

row

thco

eff

icie

nt

(TG

C)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.012 ppt RAS22 ppt RASControl cages, in sea

Exp. 2: Industrial scale RAS, growth and survival p=0.09

� Good growth rate in RAS, but yersinosisgave negative effect

� After stocking at sea, higher mortality in RAS groups, likely due to sub-optimal water quality and yersinosis

Preliminary conclusions:� In RAS with no apparent disease issues

with, 12 ppt S should be used due to better growth, skin health, and water treatment, as found in Exp. 1

� In RAS with typical FW-pathogen challenges, a higher salinity should be used, provided that water treatment capacity is sufficient

� After this exp., Grieg SeaFood has changed the RAS water treatment and biosecurity, and now have good experiences with post-smolts in RAS

AIW#6 Date

25.no

v.13

9.des

.13

23.d

es.13

6.

jan.14

20

.jan.

14

3.feb

.14

17.fe

b.14

3.

mar

.14

Cum

mul

ativ

esu

rviv

al (

%)

0

70

75

80

85

90

95

100

Control cages, in sea 12 ppt RAS, after sea transfer22 ppt RAS, after sea transfer

Average survival to harvest size(Skrudland & Bleie, 2014)

Kolarevic , Terjesen, et al., unpubl., 2013

Nofima

NCRA

FW-RAS 1000g12‰ -RAS+Exercise

Cage

12:

1224L

Cage

2000g

2000g

2000g

600g+ Exercise 12‰ -RAS

600g 1000g Cage

1000g

12‰ -RAS+Exercise9 tanks 6 tanks 3 tanks

3 cages 3 cages

3 cages

Exp. 3: medium-scale RAS with fixed salinityand water velocity, then stocking in cages

• Exp. 1: Potential for very high (99%) survial, but sensitive phase 300-700 g• Exp. 2: Importance of water quality and biosecurity, and again sensitive postsmolts• Exp. 3. goal: “test of concept” post-smolt RAS prod, with changes according to

Exp.1&3 results, now followed with stocking in cages

Ytrestøyl & Terjesen, unpublished, 2014

12L:12D

AIW#6

Exp.3: Changes to reduce mechanical contact and improve skin healthTemp in RAS keptrelatively low (12-13ºC), 0-3ºC differences to cages Feeding in exp. tanks

ended late, -2 daysAdded product w/iso-eugenol

sedation to exp.

tanks (2.7 mg/l)

Added product w/

polyvinylpyrrolidone

(PVP, 60 mg/l)

and EDTA to nets,

weighing tanks, holding tanks

Used long time & careful netting and

loading (1 h/tank)

(w/12 ppt S, iso-eugenol & PVP)

Careful pumping from tanks on truck, to exp. cages

Ytrestøyl and Terjesen, unpubl. 2014AIW#6

�Transfer @600 g: Appetite right aftertransfer, and gradual increase to July, from then on above tables for same size/temp

�Transfer @1000 g: Higher condition factor at transfer. Long time without detectable feedintake. Gradual improvement from August, then abovetables for same size/temp

Exp.3: Feed intake in postsmolts produced in RAS, after stocking in cages, at two size-classes

Ytrestøyl and Terjesen, unpub, 2014

AIW#6

Date

mai jun jul aug sep okt nov

Fee

d in

take

(%

BW

day

-1)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8Transfer 1 @600 gTransfer 2 @1000 g

�Good growth rate during time in RAS (TGC=2.8)

�After stocking, comparable weightdevelopment betweencages and RAS (avg. temp. 12.8ºC og 12.3ºC)

�Growth in the two sizeclasses reflecteddifferences in feed intake

�At ~1 kg, relatively lowmaturation, this is monitored closely

-RAS: 3% (1 kg)-Transfer 1, 7% (1.4 kg)-Transfer 2, 0.7% (1 kg)

Exp.3: Growth after transfer from RAS to cages in sea, at two size classes

Ytrestøyl and Terjesen, unpubl. 2014

AIW#6

Date

1.fe

b.14

1.ap

r.14

1.jun

.14

1.au

g.14

1.ok

t.14

1.de

s.14

Bod

y w

eigh

t (g

ind-1

)

0

500

1000

1500

2000

2500

Transfer 1, in cagesTransfer 2, in cages12 ppt S RAS

Transfer 1 Transfer 2Estimate

TGC RAS: 2.80

TGC RAS: 1.6TGC Transfer 1: 2.0

TGC Transfer 2: 0.04

�Very high survival, ~99%, calculated from smolt to 2.5 kg i cages, transfer 1

�Despite transfer intoRAS, out of RAS, on to truck, and into cages

�Shows that postsmolt production in closed-containment has potential for virtually nomortality

�However, focus must be on limited, and careful, handling of thefish

Exp. 3: Survival

Ytrestøyl og Terjesen, upublisert, 2014AIW#6

Date1.

feb.

14

1.ap

r.14

1.jun

.14

1.au

g.14

1.ok

t.14

Cum

mul

ativ

e su

rviv

al (

%)

0

580828486889092949698

100

Utsett 1, notUtsett 2, not12 ppt S RAS

Transfer 1Transfer 2

RAS: 98.0±1.1%

Transfer 1: 99.0±0.3%Transfer 2: 98.4±1.3%

Average survival in Norway to harvest size(Skrudland & Bleie, 2014)

Effects of tank scale on postsmolt production

28.10.2014 22

23

EU AquaExcel WP8: Up-scaling and validity of research results

• Scale is the size of fish containment units, i.e. tanks or cages, and biofilters

• There has been a tremendous increase in size of industrial tanks, cages, and biofilters

• Does the increasing gap in scale affect the relevance of scientific data?

• Information about scaling effects is also useful for industrial tank, bioreactor, unit design

• In AquaExcel we have modelled scale effects, and validated by experiments on sea bass, salmon, and bioreactors

• Effects on the fish (Nofima, SINTEF, NTNU, HCMR) and bioreactor nitrification rate (IMARES, Nofima, WUR)

Photo: Erik Røed

Photo: Nofima

AIW#6

Experimental design: scale effects in salmon- all scales in tank diameter (m)

Espmark, Terjesen et al, unpubl.

Phase I

AIW#6

Triplicate 40 m cages, 16 000 m3

200 000 smoltseach

Material and MethodsVariables that were standardized –between tank scales

Variables that were not standardized – but measured/monitored -tanks

Tank geometry Feed distribution angles Fish genetic & batch origin

Fish genetic & batchorigin

Tank water turnover Feeding time regime adjusted in tanks

Initial fish density Water velocity Feed batch

Feed batch + feedingregime

Water quality parameters (except DO, controlled to 85% sat.)

Temperature adjusted in tanks, according to cage (5to 14ºC)

Light (300-400 lux), and photoperiod (natural)

Photoperiod (natural)

Temperature (5 to 14ºC), oxygen saturation (85% sat.)

Variables that were standardized between cage and tank

AIW#6

Growth rate in salmon postsmolts increases with scale (m diameter tank) and is affected by previous scale history

Design:• Reared from 70g

smolts to 1000g post-smolts

• Flow-through, temp as in sea

• Same fish and feed as in controltriplicate, 40 m diameter cages

40 m diam

AIW#6

Espmark, Terjesenet al, unpubl. 2014

Research on postsmolt production in floatingsemi-closed containment systems in sea

28.10.2014 27AIW#6

Semi-closed containment systems for postsmolt production, taking water from below the sea lice belt

28.10.2014 28

Photo: Aquafarm Equipment, 21 000 m3, 450 m3/min flow

Preline, 2 000 m3

Photo: Lerøy

AquaDome. Photo: Cermaq, 5 700 m3

HDN Flexibag: 1 600 m3. Photo Smøla KS

AIW#6

OPP exp 5a: Industrial scaleexperiment, semi-closedsystem to 1 kg (Marine Harvest, UNI Research, Nofima)

Photo: Bendik Fyhn Terjesen

Photo: Bjarte Sævareid

Photo: Aquafarm Equipment

28.10.2014 30

Smolt producerSemi-closed tank

Sampling program to 1 kg:Weight, length, condition factor, TGCGill tissue, NKA, plasmaSamples for molecular and histopathology analyses

of skin healthExternal welfare indicators (e.g. fins, cataract, skin)Water quality (temp, CO2, TAN, O2, TSS, etc)Fish health screening (Postsmolt E, NVI)

24

8

Nov Apr June

Photoperiod

1 3 6 mnd0

OPP5a: Industrial scale experiment, postsmolt production in semi-closed system: Design

Reference cages

200 000, 0+ smolt transferred to each treatment 13/10 & 17/11

AIW#6

Calabrese, Handeland, et al, unpubl.

28.10.2014 31

OPP5a: Status experimental fish beforetransfer to semi-closed or reference cages

Date

2/9/

13

16/9

/13

30

/9/1

3

14/1

0/13

28

/10/

13

11/1

1/13

25

/11/

13

Na

+, K

+-A

TP

as

e

acti

vit

y

0

5

10

15

20

0.9

1.0

1.1

1.2

1.3

1.4

NKA activity

Condition factor

transferred to semi-closed24L:0D

Co

nd

itio

n f

acto

r (W

/L3)

17 ppt S

Date

2/9/

13

16/9

/13

30

/9/1

3

14/1

0/13

28

/10/

13

11/1

1/13

25

/11/

13

Bo

dy w

eig

ht

(g in

d-1

)

0

20

40

60

80

100

120

Transferred to semi-closed24L:0D 17 ppt S

Calabrese, Handeland, et al, unpubl.

AIW#6

Date

01.o

kt.1

3

01.d

es.1

3

01.fe

b.14

01.a

pr.14

01.ju

n.14

Tem

pera

ture

(°C

)

0

2

4

6

8

10

12

14

16

18

Ref. cagesSemi-closed

Transfersemi-closed

Transfer reference cages

28.10.2014 32

OPP5a: Industrial scale experiment, postsmolt prod. in semi-closed system: Weight

Comparable weight developmentin semi-closed

Water temperature

Date

1.ok

t.13

1.n

ov.1

3

1.de

s.13

1.

jan.14

1.

feb.

14

1.m

ar.1

4

1.ap

r.14

1.

mai.

14

1.jun

.14

Bod

y w

eigh

t (g

ind-1

)

0

100

200

300

400

500

600

700

800

900

1000

1100

Weight based on feed intake ref. cagesWeight based on feed intake semi-closedWeight sampling semi-closed

Stockingsemi-closedStocking

ref. cages

(Ref. cages based on fed amounts and FCR=1.0)

Calabrese, Handeland, et al, unpubl.

AIW#6

OPP5a: Industrial scaleexperiment, postsmolt prod. in semi-closed system: Growth to 1 kg

1200

1220

1240

1260

1280

1300

Semi-closed, Molnes Reference cages

Heat-sum (Σ°C, to 1 kg)

0

0.5

1

1.5

2

2.5

3

3.5

4

Semi-closed, Molnes Reference cages

Thermal growth coefficient

(TGC) to 1 kg

Calabrese, Handeland, et al, unpubl.

AIW#6

28.10.2014 34

OPP5a: Industrial scale experiment, postsmolt prod. in semi-closedsystem: Survival to 1 kg

�Indicates that postsmolt production to 1 kg in semi-closed systems, has potential for very lowmortality

�However, after thisproject, the fish werekept, and in September experienced toxic algae, and became infected withAGD, amoebic gilldisease

�Marine Harvest thereforeharvested the tank thisSunday

Date

1.ok

t.13

1.

nov.1

3

1.de

s.13

1.

jan.1

4

1.fe

b.14

1.

mar

.14

1.

apr.1

4

1.m

ai.14

1.

jun.1

4

Cum

mul

ativ

e su

rviv

al (

%)

0

5

86

88

90

92

94

96

98

100

Ref. cagesSemi-closed

Transfer,semi-closedTransfer,

referencecages

Semi-closed, Molnes99.2%

Ref. cages, 89.1%

Calabrese, Handeland, et al, unpubl.

AIW#6

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Nu

mb

er

of

fem

ale

ma

ture

lic

e p

er

fish

28.10.2014 35

OPP5a: Industrial scale experiment, postsmolt prod. in semi-closed system: Low, but not zero sea lice

Mature sea lice per postsmolt

Regulated limit

Calabrese, Handeland, et al, unpubl.

AIW#6

OPP5a: Industrial scale experiment, postsmolt prod. in semi-closed system: External welfare scores to 1 kg

Foto: Bendik Fyhn Terjesen

Means ± SD

N=60-80 fish. Each indicator is given a score between 0-2 based oncondition (0=good, 2=bad). Values are given as means±SD

Kolarevic, Terjesen, et al, unpubl.

AIW#6

Cataracts Skin lesions Operculum damage fin damage

Before transfer 0,01±0,12 0,63±0,48 0,01±0,12 0,91±0,37

3 months 0,10±0,34 0,36±0,48 0,02±0,13 0,87±0,42

6 months 0,24±0,45 0,48±0,50 0,01±0,11 0,96±0,25

OPP5a: Industrial scale experiment, postsmolt prod. in semi-closed system:Water quality to 1 kg

Foto: Bendik Fyhn Terjesen

Water sampling Molnes 12.05.14

Sampling location Temp (oC) pH Conductivity (mS/cm) Salinity (ppt) Oxygen (%sat) Turbidity (NTU) CO2 (mg/l)

Inlet 7 8.08 50.7 101 0.39

2m 7.28 7.8 50.8 32.2 82 0.28 4

10m 7 7.89 50.8 32.2 82 0.42 3.5

15m 6.9 7.9 50.9 32.2 83 0.5 3

Kolarevic, Terjesen, et al, unpubl.

Thanks for your attention!Thanks to all contributors in Nofima:Jelena Kolarevic, Trine Ytrestøyl, Harald Takle, Torbjørn Åsgård, Roger Selset, Yuriy Marchenko, Britt Kristin Reiten, Per Brunsvik, and more

And to all in the OPP-consortium:

Ørjan Tveiten, Sara Calabrese, Olav Breck, Cato Lyngøy (formerly MH)Frode Mathisen, Per Magne EriksenSvein Martinsen, Rune Iversen +HDN-consortiumHarald Sveier, Gunnar Steinn GunnarssonSigurd Handeland, Lars Ebbesson, Tom Ole NilsenSigurd StefanssonBjørn Olav Rosseland, Hans Christian Teien Sveinung Fivelstad, Camilla Hosfeld, Kristin Kvamme Åse Åtland, Torstein KristensenAgathe Medhus, Øyvind Vågnes, Kari Norheim, Brit HjeltnesErlend Haugarvoll, Rolv HaugarvollCato Lyngøy (Hauge Akva)

Funded by an industry consortium (Marine Harvest, Grieg, Smøla KS m.fl.)Funded by the Research Council of Norway-Project 217502/E40, “Optimized postsmolt production”(OPP)Associated activities funded by FHF-Project #900816 Postsmolt A, D and EFunded by the European Union, FP7-INFRAAquaExcel: AQUAculture infrastructures for EXCELLence in European Fish Research