Bioresource Technology 93 (2004) 155167

Mesophilic anaerobic treatment of sludge from saline sh

farm euents with biogas production
Ruth Gebauer

Department of Aquaculture and Natural Sciences, Finnmark University College, Follums vei 31, N-9509 Alta, Norway
Received 27 June 2003; received in revised form 2 September 2003; accepted 28 October 2003

Abstract
The mesophilic anaerobic treatment of sludge from saline sh farm euents (total solids (TS): 8.210.2 wt%, chemical oxygen
demand (COD): 6074 g/l, sodium (Na): 1010.5 g/l) was carried out in continuously stirred tank reactors (CSTRs) at 35
C. COD
stabilization between 36% and 55% and methane yields between 0.114 and 0.184 l/g COD added were achieved. However, the
process was strongly inhibited, presumably by sodium, and unstable, with propionic acid being the main compound of the volatile
fatty acids (VFA). When diluting the sludge 1:1 with tap water (Na: 5.3 g/l), the inhibition could be overcome and a stable process
with low VFA concentrations was achieved. The results of the study are used to make recommendations for the conguration of
full-scale treatment plants for the collected sludge from one salmon farming licence and to estimate the energy production from
these plants.
 2003 Elsevier Ltd. All rights reserved.
Keywords: Anaerobic treatment; Fish farming sludge; Salinity; Energy production; Biogas; Sludge treatment plant

1. Introduction
Due to signicant problems with diseases in Norwegian salmon farming, in the late 1980s some onshore
facilities for salmon grow-out in seawater were built, to
enable better control of rearing conditions. At these
onshore plants it is easier to control the discharge of
organic matter and nutrients (corresponding to that of
10,000 people per salmon licence) to the sea by purifying
the euents at the end of the outlet pipe, for example by
using micro sieves. But the purication produces sludge,
consisting of faeces and excess feed, that must be disposed of. If possible it should be reused as a fertilizer
which according to the regulations of the Norwegian
Ministry of Agriculture requires sucient stabilization
of the organic matter to avoid bothersome odours, and
hygienization to avoid spreading sh pathogens (Norwegian Ministry of Agriculture, 1998).
Stabilization and hygienization of sludge may be
achieved through several methods, among them anaerobic treatment, which, because of its energy pro-

Tel.: +47-7765-6377, +47-7845-0475.

E-mail address: ruth@hifm.no (R. Gebauer).

0960-8524/$ - see front matter  2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2003.10.024

duction, may be preferable to others. The energy may,

among other purposes, be used for pasteurisation of the
sludge, in case hygienization cannot be achieved through
the anaerobic treatment alone.
Because onshore sh farming in seawater is rarely
used, anaerobic treatment of these euents has not been
investigated before. The purpose of the present study is
therefore to nd out how the process works with saline
sh farming sludge, and what energy production may be
expected. The process may be problematic because the
sludge contains several substances at inhibitory concentrations: sodium (10.2 g Na/l) and sulphate (1.2 g
SO4 -S/l) from the seawater, which in the anaerobic
process will be reduced to sulphide, and ammonia from
the degradation of the protein in the faeces and the excess feed (29% VS, 2.43.0 g Tot-N/l). Additionally,
about 70% of the organic matter exists as particles and
the sludge arises at low temperatures (110
C). The
latter may require operating the process with a high
solid content, in order to minimize the energy demand
for heating of the sludge to process temperature.
When treating sludge collected from euents from
fresh-water trout farming, Kugelman and Van Gorder
(1991) found strong inhibition, attributed to ammonia,
when treating a concentrated sludge (46 wt% of TS,

156

R. Gebauer / Bioresource Technology 93 (2004) 155167

Nomenclature
Ca
CH4
Cl
CO2
COD
CSTR
HRT
H2 S
H2 S-S
K
Kj-N
MWh
Mg
Na
NH3 -N
NH4 -N
NO3 -N
NOK
OLR

calcium ion, Ca2 (mg/l)

methane
chloride ion, Cl
(mg/l)
carbon dioxide
chemical oxygen demand (g/l)
continuously stirred tank reactor
hydraulic retention time (days)
hydrogen sulphide (mg/l)
hydrogen sulphide sulphur
potassium ion, K (mg/l)
Kjeldahl nitrogen (mg/l)
mega watt-hour
magnesium ion, Mg2 (mg/l)
sodium ion, Na (g/l or mg/l)
unionized ammonia nitrogen (mg/l)
ammonium nitrogen (g/l or mg/l)
nitrate nitrogen (mg/l)
Norwegian crowns
organic loading rate (g COD/g VSS day
1 )

2.53.5 g Tot-N/l) in CSTRs. While Lanari and Franci

(1998) were able to successfully treat a diluted sludge
(1.32.4 wt% of TS, 0.160.24 g Tot-N/l) at only 25
C,
in an anaerobic lter lled with polyurethane foam, in
accordance with the reported threshold concentrations
for initial inhibition by ammonia of 15001900 mg TotNH4 -N/l (McCarty, 1964; Melbinger and Donellon,
1971; Koster and Lettinga, 1988).
Initial inhibition of anaerobic treatment of complex
waste by sodium has been reported from 34 g Na/l, and
total inhibition has been reported from 1011 g Na/l
(Georgacakis and Sievers, 1979: dairy waste; Toldra
et al., 1984: tomato waste). Serious disturbance by sulphide has been reported from 100 to 200 mg H2 S/l
(Lawrence and McCarty, 1966; Rinzema and Lettinga,
1988). However, Soto et al. (1991), Mendez et al. (1995),
Omil et al. (1996), and Punal and Lema (1999) reported
successful treatment of saline seafood processing euents with concentrations of sodium (512 g/l), sulphate
(0.62.7 g SO4 -S/l) and protein (14 g Tot-N/l) similar to
those in the sh farm sludge of the present investigation.
However, more than 80% of the organic matter in their
wastes was soluble, and the concentration of organic
matter was only 2550% of that in the sh farm sludge
of the present investigation.
In the present study anaerobic treatment of sludge
(8.210.2 wt% of TS) collected from saline sh farm
euents was investigated at mesophilic temperature (35

C) and in the easiest process conguration, the CSTR.
The results were used as the basis for recommendations
for process congurations for full-scale treatment plants

sulphate ion, SO2

4 (mg/l)
sulphate sulphur (mg/l)
suspended solids (g/l)
standard temperature and pressure, 0
C,
1 atm
Tot-H2 S total suphide, S2
+HS
+H2 S (mg/l)
Tot-H2 S-S total sulphide sulphur (mg/l)
Tot-N total nitrogen (g/l)
Tot-NH4 -N total ammonium nitrogen
Tot-P total phosphor (mg/l)
Tot-S total sulphur (mg/l)
TS
total solids (% of weight)
UASB upow anaerobic sludge blanket digester
VFA volatile fatty acids
vol.% % of volume
VS
volatile (organic) solids (% of weight)
VSS
volatile (organic) suspended solids (g/l)
wt%
% of weight

SO4
SO4 -S
SS
STP

and for estimating the potential energy production from

these plants.

2. Methods
2.1. Inoculum
The inoculum was taken from an experimental
anaerobic digester that was originally inoculated with a
mixture of digested municipal sewage sludge and cow


manure at the Agricultural University of Norway at As,
Southern Norway. Examination of the biomass by antiagent tests (Ahring and Nrgaard, 1994) showed the
occurrence of Methanosarcina barkeri and Methanococcus varnietii, but the absence of Methanogenium sp.
UCLA. Thus the methanogenic biomass was predominated by species that are normally found in low-salinity
digesters with suspended cultures.
2.2. Fish farming sludge (substrate)
The sludge (substrate) was collected with an air-ushed ribbon strainer in a pilot plant at the onshore sh
farm for Atlantic salmon grown out at Hemnskjel in
Middle Norway (Ulgenes et al., 1994), with the farm
operated at a feed coecient of 1.06. The composition is
provided in Table 1.
The sludge (substrate) was collected during two
periods totalling 1012 h in April 1992, at a surrounding

R. Gebauer / Bioresource Technology 93 (2004) 155167

Table 1
Composition of sludge from sieving of saline sh farm euents with an
air-ushed ribbon strainer
Component

Content

TS (wt%)
VS (% of TS)

8.210.2a
49.854.1a

Protein (% of VS)
Fat (% of VS)
Carbohydrates (% of VS)

29b
15b
56b

COD (g/l)

60.374.1a

Kj-N (mg/l)
NH4 -N (mg/l)
NO3 -N (mg/l)
Tot-P (mg/l)
Tot-S (mg/l)
SO4 -S (mg/l)

24403040a
430530a
2.22.7b
13501683b
9901230b
9201150b

Na (mg/l)
K (mg/l)
Ca (mg/l)d
Mg (mg/l)
Cl (mg/l)

10 200c
476c
4640 (1670)c
1759c
23 600c

10 Samples.
4 Samples.
c
2 Samples.
d
Concentration of dissolved Ca in parentheses.
b

temperature of 1012
C, and prior to the experiments

stored frozen at )28
C for up to two and a half years.
2.3. Experimental set up
The experiments were carried out in two 15-l digesters
of metal that were placed in a common water bath. The

157

experimental set up for one of the digesters is shown in

detail in Fig. 1. The digesters were operated semi-continuously at 35
C with 46 l of sludge volume and
stirred continuously at 200 rpm. The digesters were
sampled and fed manually, through, respectively, a
double siphon and a tube through the digester lid. The
biogas was collected in 15-l aluminium bags. For analysis of the gas composition, gas samples were taken
with a gas-tight syringe through a gas sampling point
placed on the tube leading to the gas-sampling bag. The
gas production was measured manually by emptying of
the gasbags by suction of acid sodium sulphate solution
(DIN 38 414-S8, 1985) from a tight demijohn of 25 l,
and following weighing of the collected liquid. The
weighed amount was corrected for the pressure dierence caused by the height dierence between the uid
levels in the demijohn and the collection can.
2.4. Experimental design
At rst, one of the digesters was started up in order
to get the process into operation with (undiluted) sh
farming sludge at a HRT of about 30 days (cf. Section
2.5.1), because this HRT traditionally has been recommended for the starting of anaerobic CSTRs. However,
due to malfunction of the equipment, the real HRT was
somewhat lower (cf. Table 2). Because the inoculum was
taken from a low salinity environment, the salinity in the
digester was increased gradually over a period of 266
days in order to enable probable adaptation of the
biomass. This was achieved by successive feeding with
dierent types of sh farming sludge with increasing

Fig. 1. Experimental set up for the experiments with sludge from saline sh farm euents, drawing for one of the two digesters.

158

R. Gebauer / Bioresource Technology 93 (2004) 155167

Table 2
Operating conditions during the mesophilic treatment of undiluted sludge from saline sh farm euents (Salinity: 35; Na: 10.2 g/l; TS: 8.210.2
wt%; VS: 4.85.5 wt%; COD: 60.374.1 g/l)
Period

Days

Operating condition

SLV
(ml)

OLRa
(g COD/l day)

HRTa
(days)

1a

157

First start up increase of sludge volume

37506100

50.9

1b

5895

36506000

61.5

1c

96124

Semi-continuous operation, withdrawal of sludge

for serum bottle experiments on day 78 and
following increase of sludge volume
Semi-continuous operation, withdrawal of sludge
for serum bottle experiments on day 118 and
following increase of sludge volume

1.0 until day 40,

1.55 from day 41
1.38

51506000

1.56

41.2

2a
2b

125191
192207
208232

Stop of operation
Increase of sludge volume
Semi-continuous operation

35004200
42004000

2.12
2.55

32.7
27.5

233252
253260

Stop of operation
Semi-continuous operation

4000

1.9

65

4a

261266
267282

40006100

2.42

26.2

4b

283337

Stop of operation
Second start up with sludge from the reactor,
increase of sludge volume
Semi-continuous operation

58405240

2.52

27.9

5
6a
6b
6c

338359
360363
364383
384402

Stop of feeding
Some feeding
Semi-continuous operation
Semi-continuous operation

51004600
47004900
49004800
4800

0
1.44
2.85
3.12

1
48.1
24.3
24.0

7a

402404
405423

49006000

1.15

54.4

7b

424440

Stop of operation
Third start up with sludge from the reactor,
increase of sludge volume
Semi-continuous operation

6000

1.24

60

Average for the period.

salinity (cf. Section 2.5.1). Because the process still was

unstable with high VFA concentrations after 400 days of
operation (cf. Section 3.1), the digester was operated at
60 days HRT for 40 more days. However, the increase in
HRT did not lead to a reduction to the VFA concentrations in the digester (cf. Section 3). Therefore, a second digester was started with diluted sludge (1:1 with
tap water) with 30 days HRT (cf. Section 2.5.2) in order
to establish a well functioning process with low concentrations of VFA. Because time was short this process
could not be further optimised.
2.5. Running of experiments
During start up, the digesters were feed irregularly
(cf. Sections 2.5.1 and 2.5.2) and only sampled occasionally. During periods when the sludge volume had to
be increased, at start up or after removal of sludge for
serum bottle experiments (results shown in Gebauer,
1998), the digesters were feed regularly, daily or every
second day, but not sampled, until the working volume
was reached. During semi-continuous operation, the
digesters were rst sampled and then fed once a day,
every day at the same time. However, feeding was

postponed when the pH had decreased substantially.

The pH value in the digester was measured daily, before
feeding. The gas production was measured two to three
times per week, and the gas composition was analysed
twice per week. The TS, VS and COD of the raw sludge
were analysed in every new batch of sludge. The TS, VS,
COD, VFA and ammonia in the digested sludge were
analysed twice per week. The alkalinity and the Na, K,
Ca, Mg and Cl were analysed occasionally.
2.5.1. Digester with undiluted sludge
This digester was operated for a total of 440 days. The
operating conditions during the whole period are presented in Fig. 2 and summarized in Table 2. The process
was started up with 0.8 l of inoculum, 2 l of a freshwater
sh farming sludge (TS: 3.6 wt%; VS: 2.9 wt%), 0.95 l of
tap water and 4 g of sodium bicarbonate (Na2 CO3 ) as a
buer. During the following days, the pH value was
maintained at around 7.0 by the almost daily addition of
a few grams of sodium bicarbonate. Thus in total about
60 g of Na2 CO3 were added during the start up period.
When the methane content in the biogas was signicant,
having exceeded 15 vol%, the digester was fed sporadically to avoid acidication due to overloading. First, on

R. Gebauer / Bioresource Technology 93 (2004) 155167

159

Fig. 2. Operating conditions, performance and pH-value and VFA-concentrations in the digester, during the mesophilic treatment of undiluted
sludge from saline sh farm euents (salinity: 35; Na: 10.2 g/l; TS: 8.210.2 wt%; VS: 4.85.5 wt%; COD: 60.374.1 g/l).

day 27 and 30 respectively, 500 ml of a diluted saline sh

farming sludge with low salinity (7 after dilution,
Gebauer and Hanssen, 1992) was added. Then, on day
41, 43 and 44 respectively, 50 ml of the same saline sh
farming sludge (salinity of 14) was added undiluted.
From day 45, the sludge of the present investigation

(salinity of 35, cf. Table 1) was used as substrate. First,

the digester was fed every second day with 200 ml of this
sludge, until the working volume of 6 l was reached on
day 57 (cf. Fig. 2). Further on the digester was operated
as summarized in Table 2, according to the procedures
that were described in the previous section.

160

R. Gebauer / Bioresource Technology 93 (2004) 155167

The sodium concentration in the digester was increased gradually from 1400 mg/l at start up to 5500,
6200, and 7700 mg/l on day 30, 40 and 57 respectively.
9100 mg/l were reached at the end of Period 1 (day 124),
and the sodium concentration of the raw sludge (cf.
Table 1) at the end of Period 3 (day 266).
Roughly three dierent operating regimes may be
distinguished: uneven OLR around 1.5 g COD/l day
1
and HRT around 40 days (Period 1 in Table 2 and Fig.
2), OLRs around 2.53.1 g COD/l day
1 and HRT
around 2428 days (Periods 26 in Table 2 and Fig. 2),
and OLR around 1.2 g COD/l day
1 and HRT around
60 days (Period 7 Table 2 and Fig. 2). The operation had
to be stopped four times for practical reasons (day 125
191, day 233252, day 261266, day 402404). During
these periods the digester content was kept at room
temperature without stirring. Twice, (day 267 and day
405) the digester was emptied and cleaned and started
up again with the former digester content.
2.5.2. Digester with diluted sludge
This digester was operated for a total of 74 days
under the conditions presented in Fig. 3. It was started
with 2 l of digested undiluted sludge as inoculum and 2 l
of tap water. At start up the pH value in the digester was
almost 8 and was adjusted several times to pH levels of
7.27.5 through addition of hydrochloric acid during the
rst 7 days of operation. Feeding started on day 7, when
there was measured about 10 vol.% of methane in the
biogas. The digester content was adjusted to 3900 ml,
and 300 ml of the digester content was replaced with
diluted sh farming sludge once a day. When the pH
value decreased substantially, feeding was stopped for
some days. Therefore, the loading was somewhat uneven
until day 28, when daily feeding was started at an OLR
of 1.1 g COD/l day
1 and 30 days HRT.
2.6. Analytical methods
The Kj-N was analysed according to standard
methods (APHA, 1989). The fat content was analysed
according to Folchs method (Folch et al., 1957).
The TS, VS, COD, Tot-NH4 -N, NO3 -N, Tot-P, TotS, SO4 -S and Cl were determined according to Norwegian standards (status of 1994). Before analyses of
the COD, the samples were diluted to less than 700 mg
COD/l and less than 200 mg Cl/l and homogenized. The
concentrations of the cations: Na , K , Ca2 and Mg2
were analysed by inductive couplet plasma analyses
(ICP). The composition of the biogas with respect to
CH4 , CO2 and H2 S was analysed on a gas chromatograph equipped with a packed column and a thermic
conductivity dectector (TCD) and with helium as carrier
gas. The concentrations of VFA were analysed by a gas
chromatograph equipped with a megabore capillary
column and a ame ionization detector (FID), with

Fig. 3. Operating conditions, performance and pH-value and VFAconcentrations in the digester, during the mesophilic treatment of diluted sludge from saline sh farm euents (salinity: 17.5; Na: 5.3 g/l;
TS: 4.5 wt%; VS: 2.3 wt%; COD: 33.7 g/l).

nitrogen as carrier gas. The alkalinity was determined

from unltered samples, according to the method by
Hill (1990) for sludge with a high concentration of VFA.
All analyses were carried out with two replicates, and
the COD analyses were carried out with two samples
each from two replicates (four analyses) usually with
dierences of less than 5% between the replicates. The
presented values are the average values of the replicates.

R. Gebauer / Bioresource Technology 93 (2004) 155167

More information about the analytical methods is provided in Gebauer (1998).

2.7. Calculations
The protein content was calculated from the content
of Kj-N and NH4 -N according to the formula: g protein 6.25  (g Kj-N-g NH4 -N). The carbohydrate content was calculated as the dierence between the VS and
the protein and fat content. The concentration of H2 S
in the sludge was calculated from the percentage of H2 S
in the biogas by means of Henrys law: [H2 S]sludge
a [H2 S]gas , with the absorption coecient, a, set to
24.661 mg H2 Sliquid /vol% H2 Sgas (calculated from Lawrence and McCarty, 1966). The concentration of TotH2 S in the sludge was calculated from equilibrium at
the pH value in the sample: [Tot-H2 S]sludge [H2 S]sludge
(1+10pH
pK1 ), with a pK1 value of 6.83 at 35
C
(Lawrence and McCarty, 1966). The concentrations of
NH3 -N was calculated from the NH4 -N concentration

161

and the pH value in the sample according to: [NH3 N] 1/(1+10pKa
pH ) [NH4 -N], with pKa values of 9.03
(Whiteld, 1974) and 8.95 (Perrin, 1982) respectively for
undiluted and diluted sludge at 35
C.
The COD methanised was calculated as the ratio of
the methane production per g COD added and the
stochiometric methane production of 0.350 l/g COD at
STP. The COD converted to VFA (CODVFA ) was calculated from the concentrations and COD values of the
dierent VFA. The COD anaerobically degradable was
calculated as the sum of the COD removed COD and
the CODVFA .

3. Results
The operating conditions, the performance and the
control parameters pH and concentration of VFA during the operating periods are presented in Figs. 2 and 3
for the digesters with undiluted and diluted sludge,

Table 3
Operating conditions and performance, during characteristic operating periods of the mesophilic treatment of undiluted sludge (salinity: 35; Na:
10.2 g/l) and diluted sludge (salinity: 17.5; Na: 5.3 g/l) from saline sh farm euents, ordered with respect to increasing HRT
Undiluted sludge

Diluted sludge

Operating period

6c

2b

4b

1c

7b

Parameter
HRT (days)
OLR (g COD/l day
1 )
Length of period (days)

24.0
3.12
19

27.5
2.55
25

27.9
2.51
55

41.2
1.56
29

60
1.24
17

30
1.10
34

Feed
CODin (g/l)
TSin (wt%)
VSin (wt%)
% VS of TSin

74.9
9.36
4.90
52.3

70.1
9.43
4.92
52.2

70.0
9.19
4.81
52.3

64.2
9.4
4.76
50.6

74.2
10.18
5.51
54.1

33.7
4.51
2.30
51.0

Sludge in digester
CODout (g/l)
TSout (wt%)
VSout (wt%)
% VS of TSout

39.5
6.51
2.40
36.9

29.7
5.79
1.97a
34.0

40.6
6.67
2.51
37.6

29.2
5.09
1.75
34.4

37.5
6.21
2.18
35.1

13.6
2.88
0.93
32.3

Stabilization
% COD removed
% VS removed

43.3
48.2

55.2
48.2

36.7
47.4

53.6
59.0

53.7
61.9

60
58

Gas composition
vol.% methane
vol.% H2 S

50.9
2.33.5

51.7
2.52.8

48.9
2.83.6

50.0
2.43.3

54.1
2.22.5

57.6
11.6

Meth. Product. (STP)

1 methane/g COD added
1 methane/g VS added
1 methane/g COD removed
1 methane/l sludge added

0.136
0.201
0.314
10.1

0.161
0.221
0.291
11.3

0.114
0.160
0.309
8.0

0.165
0.215
0.306
10.6

0.184
0.241
0.343
13.7

Vol. Meth. Prod. Rate (STP)

1 methane/l day
1

0.414

0.409

0.286

0.256

0.228

0.174b

Spec. Meth. Prod. Rate (STP)

1 methane/g VS in dig. day
1

0.017

0.020

0.011

0.014

0.010

0.019b

a
b

2b

0.154b
0.220b
0.257b
5.2 (diluted)b

Value caused by low VS concentration at the start of the operating period. VS accumulated in the digester during the operating period.
Uncertain value, see Section 4 in the text.

162

R. Gebauer / Bioresource Technology 93 (2004) 155167

Table 4
Conditions in the digester during characteristic operating periods of the mesophilic treatment of undiluted sludge (salinity: 35; Na: 10.5 g/l) and
diluted sludge (salinity: 17.5; Na: 5.3 g/l) from saline sh farm euents, ordered with respect to increasing HRT (nd not determined)
Undiluted sludge

Diluted sludge

Operating period

6c

2b

4b

1c

7b

2b

Condition
pH-value

6.957.0

7.057.1

6.857.0

6.766.95

7.05

6.77.0

Alkalinity (mg CaCO3 /l)

6200

nd

nd

nd

6500

>3000

VFA: acetic acid (mg/l)

16602000
stable
36704380
stable

7911102
stable
16802660
increasing

9801480
stable
34605430
increasing

nd

11701590
stable
38404650
stable

282466
stable
40159
stable

nd

9310
445
741
1300
18 800

10 520
494
1050
1610
19 720

nd

nd

5320
240
500
770
9500

Sulphide
H2 S (mg/l)
Tot-sulphide (HS
+H2 S)(mg/l)

6286
137200

6269
169202

6989
160202

5981
124155

5462
144164

2540
6293

Ammonia
NH3 -N (mg/l)
Tot-NH4 -N (mg/l)

11
1289

nd

11
1443

nd

1519
15071694

<20
<1000

Propionic acid (mg/l)

IONS from the salts
Na (mg/l)
K (mg/l)
Ca2 (mg/l)
Mg2 (mg/l)
Cl
(mg/l)

Table 5
Degradation of COD during characteristic operating periods of the mesophilic treatment of undiluted sludge (salinity: 35; Na: 10.2 g/l) and diluted
sludge (salinity: 17.5; Na: 5.3 g/l) from saline sh farm euents, ordered with respect to increasing HRT
Undiluted sludge

Diluted sludge

Operating period

6c

2b

4b

1c

7b

2b

HRT (days)
% COD methanised
% COD removed
% CODVFA
% COD anaerob. degradable

24.0
38.9
43.3
13.8
57.1

27.5
46.0
55.2
6.4
61.6

27.9
32.6
36.7
15.3
52.0

41.2
47.1
53.6
ND
>53.6

60
52.6
53.7
12.8
66.5

30
44.0a
60.0
2.1
62.1

Uncertain value, see Sections 3.2 and 4.1 in the text.

respectively. More values for the operating conditions

and the performance during characteristic operating
periods of both processes are presented in Table 3.
Values for the conditions in the digesters during the
same periods are presented in Table 4, and values for the
degradation of COD during the same periods are presented in Table 5.
3.1. Digester with undiluted sludge
The digester with undiluted sludge was operated for a
total of 440 days under the conditions described in Table
2 in Section 2.5.1 and in Fig. 2.
During Period 1 (cf. Table 2 and Fig. 2) the digester
was operated with an uneven OLR of around 1.5 g
COD/l day
1 , HRTs around 40 days and increasing
sodium concentration (cf. Section 2.5.1 of the Methods).
Gas and methane production started around the 15th

day of operation. After 7080 days of operation the

percentage of methane in the biogas had reached the
stable level of 50 vol.%. During the rst 95 days of
operation the gas production was unstable, as a result of
the unstable loading. When the loading was stabilized
the gas production also stabilized at around 0.34 l/g
COD added (at STP) corresponding to 0.46 l/l day
1 (cf.
Fig. 2 and Table 3). From day 65 the stabilization of
COD stabilized slightly above 50% and the pH was
stable around pH 7. The volatile fatty acids were not
measured during the rst 124 days of operation.
During the Periods 24 the digester was operated
with an OLR between 2.1 and 2.5 g COD/l day
1 and
HRT around 28 days, after 67 days of rest (cf. Table 2
and Fig. 2). This time the stabilization of COD decreased steadily from 60% on day 208 to 30% on day 334
(cf. Fig. 2 and Table 3). However, the gas production
was stable both throughout Period 2b (day 208232)

R. Gebauer / Bioresource Technology 93 (2004) 155167

and throughout Period 4b (day 283337), at the values

given in Table 3, before it suddenly stopped on day 337
(cf. Fig. 2). The pH-value was rather stable, rst slightly
above pH 7 during Period 2, and then slightly below pH
7 during Period 4. But the concentrations of VFA increased strongly, from zero at day 192 to about 2000
and 5000 mg/l for acetic and propionic acid, respectively, on day 337, when the gas production stopped.
It was assumed that the cessation of the gas production was caused by inhibition due to high concentrations
of volatile fatty acids as a result of overloading. Therefore, the feeding was stopped, and no feeding took place
during the following 22 days (Period 5). Gas production
began again the day after feeding was stopped and
continued during the whole period, decreasing from 0.33
to 0.19 l/l day
1 , while the percentage of methane in the
biogas increased to 60 vol.%. The concentration of all
VFA other than propionic acid decreased right after stop
of feeding, to below 400 mg/l after 3 days without feeding. However, the concentration of propionic acid increased for 10 more days after feeding was stopped, to
6000 mg/l, before decreasing slightly to 5000 mg/l at the
end of Period 5. When the pH value had increased to 7.1,
feeding was started again, on day 360 (Period 6) at a
somewhat higher OLR, 3.1 g COD/l day
1 , and a shorter
HRT, 24 days, than before. The change in the operating
conditions was caused by changing to a batch of sludge
with a higher VS concentration, and by a reduction in the
sludge volume in the digester due to evaporation in the
previous period. During Period 6 stable operation was
achieved 15 days after the restart of feeding, under the
conditions presented in Fig. 2 and Table 3, but the
concentrations of VFA were high.
Finally, during Period 7 (day 405440) the digester
was operated with lower OLRs around 1.2 g COD/l
day
1 and higher HRTs of 5560 days (cf. Table 2 and
Fig. 2), in order to achieve a reduction of the VFA
concentrations. The average stabilization of COD, the
gas production, the percentage of methane in the biogas
and the pH-value all increased, to the values presented
in Table 3. The concentration of propionic acid increased slightly, to 4500 mg/l. Due to the new start up
the level of acetic acid rst decreased and then increased
to about 2000 mg/l and was still increasing at the end of
the experimental period. The concentrations of the
longer VFA decreased to below 500 mg/l.
3.2. Digester with diluted sludge
This digester was operated for a total of 74 days under
the conditions described in Section 2.5.2 and Fig. 3.
The digester was mainly operated at 30 days HRT
and the corresponding OLR of 1.1 g COD/l day
1 .
Feeding was started on day 7, but until day 28 the
loading was somewhat uneven (cf. Section 2.5.2). Gas
production started immediately after feeding was star-

163

ted, at 0.18 l/l day
1 , corresponding to 0.16 l/g COD

added, and increased steadily to 0.30 l/l day
1 and 0.32
l/g COD added. However, the great dierence between
the percentage of COD removed and the percentage of
COD methanised (cf. Table 5) indicates that the gas
production was measured 2530% too low due to measurement errors that not could be localized. Thus the
values provided for the gas and methane production
from the diluted sludge must be considered as uncertain.
The stabilization of COD varied rst around 60% and
from day 39 stabilized at 60% (cf. Fig. 3), and the percentage of methane in the biogas stabilized at 58 vol.%
after 34 days of operation. The pH-value stabilized
slightly below pH 7. The concentration of propionic acid
was above 1000 mg/l during the rst 25 days of operation, due to the propionic acid content of the inoculum
(digested undiluted sludge). But, on day 30 it had decreased to about 100 mg/l, while the concentration of
acetic acid had increased to about 450 mg/l. Both concentrations stabilized at these levels throughout the
remainder of the operating period.

4. Discussion
4.1. Performance
The digesters were operated as low-loaded sewage
digesters, with HRTs in the upper range, and performed
correspondingly with respect to stabilization of organic
matter, methane yield, volumetric methane production
rate and specic methane production rate (Metcalf &
Eddy, 1991; ATV, 1996). However, the digested undiluted sludge was not stabilized according to VFAcriteria (demanded: VFA < 1000 mg/l; Loll and M
oller,
1984). The percentage of methane in the biogas was
lower than the usual 60% in biogas from sewage digestion, which indicates inhibition of lipid degradation. The
methane yield and, as far as provided, the removal of
organic matter were in accordance with the values from
the trout farming sludge measured by Kugelman and
Van Gorder (1991) for fresh water sh farming sludge,
provided that the values for the yields are corrected for
the dierent COD/VS ratios of the two types of sludge.
As expected, both the methane yield and the removal
and degradation of organic matter (cf. Tables 5 and 3)
increased with increasing HRT, apart from the period
with the strong increase of the VFA, Period 4b in Fig. 2.
Anaerobic degradability (cf. Table 5) was, at comparable HRTs, similar during the digestion of undiluted
sludge and diluted sludge. This indicates that it primarily depended on the composition of the sludge, and
that the inhibition presumably not aected the hydrolytic and fermentative bacteria had.
The dilution of the sludge increased the COD removal by at least 5%, corresponding to the lower VFA

164

R. Gebauer / Bioresource Technology 93 (2004) 155167

concentrations in the digested sludge (cf. the periods 2b

in Tables 3 and 5). However, the methane yield did not
increase correspondingly, to values around 0.20 l/g COD
added. This raises doubts about whether the measured
value of 0.154 l/g COD added is correct (cf. comments in
Section 3.2 of the Results). From the presented values,
the methane yield and removal and degradation of organic matter for high rate CSTRs with 1020 days of
HRT may be (linearly) extrapolated to about 0.100.12
l/g COD added, 3540% and 5356% respectively. Wellfunctioning processes may achieve methane yields of
about 0.20 l/g COD added and COD removal and
degradation of 6070%. The consequences of the results
for the conguration of full-scale plants are further
discussed in Section 4.7.
4.2. Process stability
The treatment process with undiluted sludge was too
unstabledue to strong inhibitionto be scaled up.
The inhibition and instability were indicated by high
concentrations of VFA (4.27.2 g/l as acetic acid), with
high propionate/acetate ratios of 2.43.7, that were far
beyond the value of 1.4 that Hill et al. (1987) proposed
as the threshold value for a stable digestion process. The
concentrations of isobutyric and isovaleric acid were
also far beyond the threshold value of 6 mg/l that Hill
and Holmberg (1988) proposed as an indicator for
process stability. However, because of the high alkalinity in the digester sludge (cf. Table 4), the high concentrations of VFA did not cause process disturbance,
due to acidication of the digester content. It is more
likely that the propionic acid directly inhibited the
hydrogenotrophic methanogens (Hobson and Shaw,
1976) and, probably, other bacterial groups active in the
digestion process.
The inhibition was overcome by dilution of the sludge
1:1 with tap water, and a stable process (propionate/
acetate ratio 0.10.3) with satisfactory VFA concentrations (<600 mg/l) could be achieved after 30 days of
operation (cf. Fig. 3 and Table 4). However, dilution will
increase both the digester size and thus the capital costs of
the treatment plant, and the energy demand for heating
the sludge to process temperature (cf. Section 4.7).
4.3. Reasons for inhibition
Previously, Soto et al. (1991) suggested that either
ammonia (up to 4 g NH4 -N/l) or sulphide (beyond 40
133 mg H2 S/l) was the reason for the inhibition occurring during the anaerobic treatment of saline (Na: 9.9
and 8.4 g/l respectively) seafood-processing wastewaters.
The inhibitory eect of the sodium was assumed to have
been overcome through adaptation of the biomass and
by the antagonistic eects of other cations (Omil et al.,
1995). However, in the present study, the ammonia

concentration in the digester with undiluted sludge

(cf. Table 4) was only slightly above the reported
threshold levels of 15001900 mg/l for initial inhibition
of unadaptated suspended cultures (McCarty, 1964;
Melbinger and Donellon, 1971). In addition, the H2 S
concentration was lower than the inhibitory 133 mg
H2 S/l measured in the experiments by Soto et al. (1991)
(cf. Table 4), because of a higher COD:SO4 -ratio in the
sh farming sludge than in the seafood processing
wastewater, so that more sulphide was removed with the
biogas. Thus, in the present study, only the sodium, and
perhaps other salt-ions, could have been the main reason for the strong inhibition. This assumption is supported rst by the high concentration of propionic acid
in the digester, because the propionic acid-using bacteria
are more sensitive to sodium than the acetoclastic
methanogens (Liu and Boone, 1991; Feijoo et al., 1995),
and then by the fact that dilution to the moderately
inhibitory sodium concentration of 5.3 g/l (McCarty,
1964) enabled a stable digestion process.
4.4. Adaptation of the biomass
The process with undiluted sludge was strongly
inhibited even after more than 400 days of operation.
This indicates that only insignicant adaptation of the
biomass to the high sodium concentration can have
occurred, even if the sodium concentration was increased gradually during start up (cf. Section 2.5.1). This
was in contrast to the results obtained by Soto et al.
(1991) and Omil et al. (1995, 1996), who both reported
signicant adaptation, and indicates that the biomass
used in the present study did not contain species that
could tolerate high sodium concentrations. Thus, Shipin
et al. (1994) and Aspe et al. (1997) found little adaptation in methanogenic cultures from low salinity environments, but considerable adaptation in cultures from
saline environments. This indicates that adaptation to
high sodium concentrations is more likely to happen as
a result of selection of tolerant species than by adaptation of every single microorganism. These tolerant species were probably absent from the biomass used in
the present investigation, as the inoculum was taken
from low salinity environments (cf. Section 2.1 of the
Methods), the bacteria in the faeces in the sludge (substrate) had mainly been exposed to the low salinity
environments (1012) in the intestine, and the seawater for the sh farm was pumped from 70 m depth
and had low bacteria concentrations. Additionally, the
storing at )28
C may have been unfortunate for the
survival of salt tolerant methanogens.
4.5. Antagonistic eects
Because of the strong inhibition of the process, it is
unlikely that the other ions in the sh farming sludge

R. Gebauer / Bioresource Technology 93 (2004) 155167

acted antagonistically on the sodium inhibition. This

result also goes against the observations of Soto et al.
(1991) and Omil et al. (1995, 1996), who reported substantial antagonistic eects during the treatment of
seafood processing wastewaters with the ion composition of: 12 g Na/l, 0.17 g K/l, 0.51 g Ca/l, 0.38 g Mg/l,
19.2 g Cl/l and 0.26 g SO4 /l (Feijoo et al., 1995). However, this ion composition deviated signicantly from
that of the sh farming sludge used in the present
study (cf. Tables 1 and 4). In particular, the concentration of magnesiumwhich, according to Kugelman and
McCarty (1965), acts synergistically on sodium inhibition two-ion systemswas three times higher in the sh
farming sludge. This could explain the lack of antagonistic eects in the present investigation.
4.6. Development of active methanogenic biomass
The specic methane production rate of the biomass
was low throughout the whole operating period (cf.
Table 3). In contrast, Soto et al. (1991) and Omil et al.
(1996) could, during 100357 days of treatment of saline
seafood processing wastewaters, increase the methane
production rates of the biomass by three to four times to
0.18 and 0.16 g COD/g VSS day
1 (corresponding to
0.063 and 0.056 l methane/g VSS day
1 ) from the same
low methane production rates as measured in the present study (Omil et al., 1995). The dierent results may be
explained by the high particle content found in the
digesters with sh farming sludge. The particles probably reduced the contact between substrate and biomass
and thus prohibited biomass growth and development
of more active sludge. This indicates that low particle
content in the digester sludge may be a condition for the
development of active methanogenic biomass.
4.7. Consequences for full-scale treatment plants
The main goals of the treatment process are stabilization and hygienization of the sludge. Additionally,
the process may be interesting as a method for energy
production.
It was only possible to achieve stabilization, according to VFA-criteria, in the process with diluted sludge;
and the methane yield of this processabout 6.2 l/l
sludge estimated with the comment in Section 4.1
would be sucient to warm the sludge to process temperature (required: 4.5 l/l sludge at incoming sludge
temperature of 1
C). However, to reduce the reactor
size and the reactor costs treatment of diluted sludge
should be further investigated in a process conguration
with increased biomass retention as an anaerobic contact process, an AF or an UASB.
But, unless unexpensive external heat is available for
heating of the sludge to process temperature (see below),
treatment of undiluted sludge will give a greater net

165

energy production due to reduced energy demand for

heating. Therefore, it should be further investigated
whether stabilization of undiluted sludge may be
achieved through the use of an inoculum from a saline
environment, through a better adaptation regime during
start up, and/or by increasing biomass retention through
an anaerobic contact process. However, because of the
high particle content of the sh farming sludge it may be
most appropriate, to divide the treatment process in two
steps, as earlier proposed by Kugelman and Van Gorder
(1991): a rst step in a CSTR for solubilisation and
acidication of the particulate organic matter, possibly
after it has been removed from the main stream, and a
second step in an AF or UASB for methanogenesis of
the dissolved organic substrate in the overow from the
rst step. In such a two-step process, both steps may be
optimised separately, both increasing the stabilization
and methane production and reducing the treatment
time and thus the reactor volumes and the costs for the
treatment plant.
It has to be further investigated whether sucient
hygienization may be achieved by mesophilic anaerobic
treatment. So far, only thermophilic (55
C) anaerobic
treatment has been approved for hygienization. Thus, if
hygienization by pasteurisation were to be required,
only the methane production from digestion of undiluted sludge (10 l/l sludge, cf. Table 3) would be sucient to warm the sludge to 60
C for pasteurisation
(requiring 8l methane/l sludge).
Today most Norwegian salmon farms operate at least
two to four salmon licences, but often one per single
location. Therefore a farm size of one salmon farming licence, i.e. presently the licence to use 700 tons of
salmon dry feed per year (Norwegian Directorate of
Fisheries, 1996), is used to evaluate the consequences
of the results of the present investigation for full-scale
plants. As 1520% of this feed will be recovered as
sludge dry matter, a salmon licence will discharge about
100 tons of VS/year. With a performance as in the
digesters of this investigation, the gross energy production from the collected sludge from one salmon farming
licence will be 180250 MW h per year, while warming
of the sludge to process temperature will reduce the
energy production to 80 or 165 MW h/year. The net
energy production corresponds to the energy demand of
two to ve Norwegian family households per year and
may be a signicant contribution to the energy supply of
the small settlements often situated in the neighbourhood of sh farms. Given an energy price of 0.40.5
NOK/kW h, the value of the net energy production
would be 32 00082 500 NOK/year. However, with the
process conguration examined in this study, the
investment costs for the digestion plant (digester size
170350 m3 ) would be around 1.52 million NOK
(Knap, 2002). Comparison of these numbers suggests
that the economic sustainability of the process requires

166

R. Gebauer / Bioresource Technology 93 (2004) 155167

lower investment costs for the treatment plant, and thus

optimalisation of the process towards smaller digester
sizes, as suggested above. Additionally, economic sustainability may be achieved if inexpensive external heat
is available to warm the sludge to process temperature,
as in cases where the sh farm is situated close to a
cooling or freezing facility.

Rolf Erik Olsen for the fat analyses. I would also like to
thank Gunnar Hartvigsen at The University of Troms
for a good course in Introduction to research and my
fellow students Therese With-Berge, Marianne Stensrd, Jrgen Mllmann and Sveinn-Are Hanssen for
their valuable comments on the paper. Finally, I would
like to thank Ian Harkness for proofreading of the paper
and language corrections.

5. Conclusions
References
Anaerobic treatment of saline sh farming sludge
could reduce the COD content by between 36% and
60% and yielded a methane production of 0.114
0.184 l methane/g COD added.
The gross and net energy production of the biogas
from one salmon farming licence would be 180250
MW h/year and 80165 MW h/year, respectively.
The treatment process with undiluted sludge was
unstable due to strong inhibition and did not result
in stabilized sludge.
The inhibition was most probably caused by sodium
and there were no signs of adaptation of the biomass
or antagonistic eects of other ions.
The high particle concentration in the digester most
probably prohibited the development of active methanogenic biomass.
The treatment process with diluted sludge (1:1 with
tap water) was stable, the sludge was stabilized and
the inhibition was overcome.
For full-scale treatment of undiluted sludge a twostep process with inoculum from a saline environment should be investigated.
For full-scale treatment of diluted sludge processes
with increased biomass retention should be investigated.
Treatment of both undiluted and diluted sludge may
be economically sustainable if inexpensive external
heat is available for warming of the sludge to process
temperature, for example from the condenser of a
cooling plant.

Acknowledgements
I would like to thank The Research Council of
Norway, The Technical University of Trondheim and
Finnmark University College in Alta for the nancial
support of this investigation, and the Technical University of Norway for my working place. I would like to
thank Jon Fredrik Hanssen at The Agricultural Uni
for kindly providing the inocversity of Norway at As
ulum and Birgitte Ahring and Claus Nrgaard at The
Technical University of Denmark for conducting the
bacterial analyses of the inoculum. I would like to thank
my former colleague at Finnmark University College

Ahring, B.K., Nrgaard, C., 1994. Report from the bacterial analysis
of the digested saline sh farming sludge. Personal communication.
APHA, 1989. In: Clesceri, L.S., Greenberg, A.E., Trussell, R.R. (Eds.),
Standard Methods for Examination of Water and Wastewater,
17th ed. American Public Health Association, Port City Press,
Baltimore.
Aspe, E., Marti, M.C., Roeckel, M., 1997. Anaerobic treatment of
shery wastewater using a marine sediment inoculum. Water
research 31 (9), 21472160.
DIN 38 414, Teil 8, Juni 1985. Schlamm und Sedimente (Gruppe S)
Bestimmung des Faulverhaltens (S8). Deutsches Institut f
ur Normung (in German).
Feijoo, G., Soto, M., Mendez, R., Lema, J.M., 1995. Sodium
inhibition in the anaerobic digestion process: Antagonism and
adaption phenomena. Enzyme and Microbial Technology 17 (2),
180188.
Folch, J., Lees, N., Sloane, S.G.H., 1957. A simple method for the
isolation and purication of total lipids from animal tissue. Journal
Biological Chemistry 226 (5), 497509.
Gebauer, R., Hanssen, J.F., 1992. Anaerobic digestion of sludge from
aquaculture. In: Leinonen, S. (Ed.), Anaerobic treatment of
municipal solid waste and sludge for energy production and
recycling of nutrients. Proceedings of 2nd Nordic workshop on
anaerobic digestion, 12 October 1992 Joensuu, Finland, No. 105.
University of Joensuu Publications of Karelian Institute, pp. 43
50.
Gebauer, R., 1998. Anaerobic digestion of sh farming sludge. PhD
thesis 1998:123, Faculty of Civil and Environmental Engineering,
Norwegian Technical University, Trondheim.
Georgacakis, D., Sievers, D.M., 1979. Bacterial response to salts in
an anaerobic diary lagoon. Transactions of the ASAE 22 (1),
162168.
Hill, D.T., 1990. Alkalinity measurements in anaerobic digestion
systems as inuenced by organic acid level and endpoint pH.
Transactions of the ASEA 33 (5), 17171719.
Hill, D.T., Cobb, S.A., Bolte, J.P., 1987. Using volatile fatty acid
relationships to predict anaerobic digester failure. Transactions of
the ASAE 30 (2), 496501.
Hill, D.T., Holmberg, R.D., 1988. Long chain volatile fatty acids
relationships in anaerobic digestion of swine waste. Biological
Wastes 23 (2), 195214.
Hobson, P.N., Shaw, B.G., 1976. Inhibition of methane production
by Methanobacterium formicum. Water Research 10 (4), 849
852.
Koster, I.W., Lettinga, G., 1988. Anaerobic digestion at extreme
ammonia concentrations. Biological Wastes 25 (1), 5159.
Knap, A.H., 2002. Hifo Tech AS Porsgrunn, Norway. Personal
communication.
Kugelman, I.J., Van Gorder, S., 1991. Water and energy recycling in
closed aquaculture systems. In: Engineering Aspects of Intensive
Aquaculture. Proceedings of Aquaculture Symposium, Cornell
University, April 46, 1991. Northeast Regional Agricultural
Engineering Service (NRAES)-49, Ithaca, New York, pp. 8087.

R. Gebauer / Bioresource Technology 93 (2004) 155167

Kugelman, I.J., McCarty, P.L., 1965. Cation toxicity and stimulation
in anaerobic waste treatment. Journal Water Pollution Control
Federation 37 (1), 97116.
Lanari, D., Franci, C., 1998. Biogas production from solid wastes
removed from sh farm euents. Aquatic Living Resources 11 (4),
289295.
Lawrence, A.W., McCarty, P.L., 1966. The eects of suldes on
anaerobic treatment. Air & Water Pollution International Journal
10 (2), 207221.
Leschber, R., Loll, U. (Eds.), 1996. ATV-HandbuchKlarschlamm,
4th ed. Ernst & Sohn, Berlin, pp. 157198 (in German).
Liu, Y., Boone, D.R., 1991. Eects of salinity on methanogenic
decomposistion. Bioresource Technology 35 (3), 271273.
Loll, U., M
oller, U., 1984. Prozessziele der Klarschlammstabiliserung
und deren Zusammenhang Dimensionierungs-, Betriebs- und
Kontrollparametern. Korrespondenz Abwasser 31 (6), 361366
(in German).
McCarty, P.L., 1964. Anaerobic waste treatment fundamentals, part
III: Toxic materials and their control. Public Works 95 (11), 9194.
Melbinger, N.R., Donellon, J., 1971. Toxic eects of ammonianitrogen in high-rate digestion. Journal Water Pollution Federation
43 (6), 16581670.
Mendez, R., Lema, J.M., Soto, M., 1995. Treatment of seafoodprocessing wastewaters in mesophilic and thermophilic anaerobic
lters. Water Environment Research 67 (1), 3345.
Metcalf & Eddy, Inc., 1991. Wastewater engineering. Revised by
Tchobanoglous, G., Burton, F.L., 3rd ed. McGraw-Hill.
Norwegian Directorate of Fisheries, 1996. Forskrift om produksjonsregulerende tiltak for oppdrett av laks og rret med endringer,
status 2002 (Regulation, in Norwegian).
Norwegian Ministry of Agriculture, 1998. Forskrift om handel med
gjdsel og jordforbedringsmidler m.v. (Regulation, in Norwegian).
Norwegian Standards, elaborated by The Norwegian Standards
Association (NSF). Status of December 1994.

167

Omil, F., Mendez, R.J., Lema, J.M., 1995. Characterization of

biomass from a pilot plant digester treating saline wastewater.
Journal Chemical Technology Biotechnology 65 (4), 384392.
Omil, F., Mendez, R., Lema, J.M., 1996. Anaerobic treatment of
seafood processing waste waters in an industrial anaerobic pilot
plant. Water SA 22 (2), 173181.
Perrin, D.D. (Ed.), 1982. Ionisation Constants of Inorganic Acids
and Bases in Aqueous Solution, second ed. Pergamon Press,
pp. 38.
Punal, A., Lema, J.M., 1999. Anaerobic treatment of wastewater from a
sh-canning factory in a full-scale upow anaerobic sludge blanket
(UASB) reactor. Water Science and Technology 40 (8), 5762.
Rinzema, A., Lettinga, G., 1988. The eect of sulde on the anaerobic
degradation of propionate. Environmental Technology Letters 9
(1), 8388.
Shipin, O.V., Dunn, K.M., Shipin, V.Y., Rose, P.D., 1994. Saline
anaerobic digestion in advanced algal high rate oxidation ponding
for the treatment of organics in saline euents. In: Proceedings of
Seventh International Symposium on Anaerobic Digestion. 23
27th January, Cape Town, South Africa, International Association
on Water Quality, RSA Litho (Pty) Ltd, pp. 397404.
Soto, M., Mendez, R., Lema, J.M., 1991. Biodegradability and toxicity
in the anaerobic treatment of sh canning wastewaters. Environmental Technology 12, 669677.
Toldra, F., Flors, A., Lequerica, J.L., 1984. Revista de Agronomica y
Tecnologia de Alimentos, 400408 (in Spanish).
Ulgenes, Y., Eikebrokk, B., Gebauer, R., 1994. Oppsamling, Karakterisering og behandling av slam fra landbaserte oppdrettsanlegg.
SINTEF rapport STF60 A94065. SINTEF (Foundation for
Scientic and Industrial Research at the Norwegian Institute of
Technology) Trondheim (in Norwegian).
Whiteld, M., 1974. The hydrolysis of ammonium ions in sea watera
theoretical study. Journal Marine Biological Association UK 54
(6), 565580.