Microbiological characterization of winery effluents: an

Water Science and Technology Vol 51 No 1 pp 19–26 ª IWA Publishing 2005

inventory of the sites for different treatment systems
F. Jourjon*, S. Khaldi*, M. Reveillere*, C. Thibault*, A. Poulard**, P. Chretien*** and J. Bednar****
* Laboratoire GRAPPE, Ecole Supérieure d’Agriculture d’Angers, 55, rue Rabelais, B.P. 748,
F-49007 Angers Cedex, France (E-mail: f.jourjon@groupe-esa.com)
** ITV, Château de la Frémoire, F-44120 Vertou, France
*** ITV, 42, rue Georges Morel, F-49070 Beaucouzé, France
**** Department of Languages, Clemson University, Clemson, SC 29634-0535, USA

Abstract In a more and more regulated and socially pressured environment, the durable management of
winery effluents must take into account their characteristics and their potential impact on their natural
setting. The object of this exploratory study is to establish an inventory of the microbiological composition
of winery effluents coming from different treatment systems. We have observed that winery effluents
are charged with micro-organisms, by a factor that ranges from 105 to 108 UFC/ml, and that the level
of “microbiological pollution” is independent of the type of system. The composition of the flora is closely
tied to the time of year and therefore to winery activities, so certain micro-organisms will be favoured in
certain periods and others will have a tendency to decrease. We have seen that from one year to
another our observations remain identical; the flora equilibrium therefore occurs systematically and
naturally. Faecal germs are found in very small quantities in winery effluent treatment systems. They
represent minor sanitary risks. Good correlations were observed between some micro-organisms and
some physical-chemical parameters (COD). It is, however, difficult to use these “easy-to-measure”
parameters as reliable markers of certain microbial populations.
Keywords Microbial pollution; winery effluents; ventilated storage; ventilated lagoon

Introduction
Current legislation requires wineries to treat their effluents from now on. The choice of
a treatment system is not simple and must integrate a number of elements.
The work accomplished has allowed us to characterize the effluents produced by wineries
by volume and polluting content (Racault and Lenoir, 1994; Jourjon et al., 1998); and also to
highlight impact risks of spreading effluents on plants and vines (Jourjon et al., 1999). Other
studies have discussed the performance of treatment systems (Racault et al., 1998; Rochard
et al., 1998).
All of these studies have contributed to our knowledge of the physical-chemical
characteristics of the effluents (parameters used in the legislation), but none of the studies
seems to have addressed the microbiological composition of the effluents during the treat-
ment put in place. However, taking into account the growing pressures tied to sanitary risks
and consumer demands, the microbiological criterion could become a supplementary
evaluation criterion for the performance of a treatment system. In order to respond to this
objective, the GRAPPE laboratory, in collaboration with the ITV (Centre Technique et
Interprofessionnel de la Vigne et du Vin), has become interested in this approach.
The objectives of the study are multiple. First and foremost, it is necessary to establish an
inventory of the microbial composition present in winery effluents. In order to accomplish
that, and establishing the most comprehensive point of view, different types of systems
were examined. The levels of microbial populations at different times of the year and the 19
 Table 1 Description of the monitored sites

Name of site: Lagoon Cascade CP8 Ventilated Cascade LS

(Vaslin Bucher) storage (Vaslin Bucher)

Type of Storage according Storage with Ventilated Storage with

treatment to lagoon type controlled ventilation, storage then controlled ventilation,
then spreading tangential filtration, poured through decanting, then
then poured through a sand filter pouring through
F. Jourjon et al.

a sand filter a sand filter

Region Anjou Anjou Pays Nantais Pays Nantais
Storage size 2,200 m3 30 m3 5 m3 3,000 m3
Type of Co-op cellar Private cellar Experimental Wine merchant’s
activity cellar harvesting facility
Monitoring 2001 March to July 2001 Sept-Dec 2000 Sept-Dec 2000
date and 2001 and 2001
Effluent In one part of the After stirring of Sampling at Sampling at
samplings lagoon the tank irregular intervals irregular intervals
1 sample per 1 sample
month on the per month
surface

variability of the flora composition from one year to the next for a single treatment system,
or between two different systems were studied. This work also focuses on a search for
eventual, easy-to-analyse indicators that will reveal a better understanding or predict the
microbiological characteristics of the effluents in question.

Methods
Monitoring was done on four different treatment systems in Anjou and the Pays Nantais
(Table 1). Each sample was gathered in a sterile flask and the quantity was one litre. The
analyses of the Pays Nantais sites were done at the IDAC and the analyses in the Anjou
region were done by the GRAPPE laboratory. Occasionally, the analysis methods differ for
certain micro-organisms, but this did not hamper the interpretation of the results since only
identical periods of monitoring were compared.
Physical-chemical and microbiological analyses were done on each sample. The
chemical demand for oxygen analyses (COD) were done using the rapid Hach method and
the pH measurements with a pH meter from the Inforlab laboratory, equipped with a Xérolyt
Mettler Toledo probe. The suspended solids were accomplished by filtration through a GFC
(Whatman) filter on a vacuum spray boom and drying by an infra-red Sartorius weighing
device. The turbidity was measured with an Hach turbidity meter.
For the microbiological approach, on the one hand we used the various classical analyses
done in urban water treatment facilities, accounting for the global contamination level
of the effluents, that is to say, the aerobic flora at 22 and 37 C (AF), and the 22 C anaerobic
flora (ANAF). On the other hand, we counted the lactic bacteria, the acetic, and the yeasts.
Since these micro-organisms are found in wine (in varying quantities according to the time of
year), it can be assumed that they are also found in winery effluents.
Finally, we looked for indicators of faecal contamination: Escherichia coli, enterococci,
and sulfur-reducing anaerobic spore bacteria (SRB), in order to evaluate the eventual
sanitary risks associated with winery wastewaters, occasionally mixed with domestic
effluents.
The methods used adhere to NF norms:
– For aerobic flora at 22 and 37 C, the norm was NF EN ISO 6122 (T90-401).
– For anaerobic flora at 22 C, NF 90-417 (GRAPPE laboratory) and a Shaedler medium
20 (IDAC).
– For the lactic bacteria, the MRS medium was used.
– For the acetic bacteria, we used the Carr (GRAPPE) medium or a gelatinized beer wort
medium (IDAC).
– The yeasts were counted using the NF VO8-059 norm.
– For the faecal contamination indicators:
– Escherichia coli: at IDAC the method chosen was the NF EN ISO 9308-3 T90-433;
in the GRAPPE laboratory, the V08 053 method was used.

F. Jourjon et al.
– Enterococci: the same method was used in both laboratories: NF EN ISO 7899-1.
– Sulfur-reducing anaerobic bacterial spores (SRB): the NF T 90-415 method was
used.
The results were interpreted using principal component analysis (PCA). The software used
was Statgraphics Plus 5.0 and Univwin Plus 4.1 (Sigma Plus).

Results and discussion

Global floral level of winery effluents
All of the winery effluents analyzed present a high level of micro-organisms. Adding up the
different types of micro-organisms counted, and excluding aerobic flora at 37 C (because it
was probably already taken into account in aerobic flora at 22 C), we obtain a variable that
we will call “inventoried biomass”. It is not the total flora present in the medium but rather
the total of the different flora inventoried by our methods.
Figure 1 shows us that the global level of the flora varies in time but that the total count is
between 105 and 108 UFC/ml. This figure is comparable to the quantity of micro-organisms
contained in wine, which is 106 UFC/ml.
This observation is valid, no matter what the time period is or what type of treatment
is used. The oxygenation of a system is not a determining factor for the level of bacterial
pollution, since a non-ventilated system (lagooning) can have a higher number of micro-
organisms than in a ventilated system (cascade CP8).
The global number of micro-organisms is independent from the time period, and therefore
from the winery activity and the treatment system, but it seems to have a connection with
the winery origin of the effluents.

1,00E+09

1,00E+08 cascade cascade

LS 2000 LS 2001

Lagooning
1,00E+07
UFC/ml

1,00E+06
cascade CP8
Ventilated
Ventilated
stock 2000
stock 2001
1,00E+05
Harvesting - Harvesting-
Filtering and winemaking
winemaking Bottling

1,00E+04
12/08/00 01/10/00 20/11/00 09/01/01 28/02/01 19/04/01 08/06/01 28/07/01 16/09/01 05/11/01 25/12/01
date

Figure 1 Inventoried biomass in the different systems studied 21

 1,00E+09 10000
aerobic
9000 flora 37C
1,00E+08
8000 aerobic
flora 22C
1,00E+07 7000
ANA Flora
6000

COD mg/l
UFC/ml 1,00E+06
5000 Acetic
F. Jourjon et al.

1,00E+05 bacteria
4000

3000 Lactic
1,00E+04 bacteria
2000
1,00E+03 yeast
1000

1,00E+02 0 COD
04/10/0011/10/00 18/10/00 25/10/00 01/11/0008/11/0015/11/00 22/11/00 29/11/00 06/12/00
Sample date

Figure 2 Evolution of the cascade LS biomass during the “harvesting-winemaking” period in 2000

The evolution of various micro-organisms in connection with the period of winery activity
Activities in a cellar vary all year long. During the harvesting and winemaking period, the
effluents are much more charged with organic matter than during the bottling period (Racault
et al., 1994; Jourjon et al., 2000).
The flows of incoming effluents are also quite varied all during the year (Arcanger et al.,
1999). Figures 2 and 3 show the flora present in two almost identical systems: the Vaslin
Bucher system at different periods of activity: harvesting–winemaking on the one hand,
and filtering-bottling on the other.
In Figure 2, we see that the different types of micro-organisms evolve in September and
December. Right after the harvest, the lactic and acetic bacteria are prevalent and remain
strongly present until December, even if, at the end of year, the 22 C aerobic flora becomes
the principal flora. In Figure 3, we observe that the evolution in time of the different
micro-organisms is almost identical and that the 22  C aerobic flora remains prevalent during
the filtering–bottling period.
The nature of the micro-organisms present in the different stocking systems is in close
connection with the activity of the cellar. During the winemaking period, a wide variety

1,00E+09 2500
Aerobic flora 22C

1,00E+08
2000
Aerobic flora 37C
1,00E+07
COD mg/ml

1500
1,00E+06
UFC/ml

ANA Flora

1,00E+05
1000
Lactic bacteria
1,00E+04
500
1,00E+03 Acetic bacteria

1,00E+02 0
COD
01
16 1
01

30 1
01

14 1
01

28 1
01

11 1

18 1
01

02 1
01
0

7/
4/

4/

4/

4/

5/

5/

5/

5/

6/

6/

6/

6/

7/
/0
/0

/0

/0

/0

/0

/0

/0

/0

/0

/0

/0

/0

/0
09
09

23

07

21

04

25

Sample date

22 Figure 3 Evolution of the cascade CP8 biomass during the “filtering and bottling” period in 2001
Table 2 Results of the counts (UFC/ml) for the ventilated stocking system

Mid-October End November

Micro-organisms 2000 2001 2000 2001

Aerobic flora 37 C 1,19.105 4,5.105 4,14.105 1,71.105
Aerobic flora 22 C 2,75.105 1,64.106 1,77.105 1,82.105
Anaerobic flora 5,4.105 1,31.105 7,8.104 2,04.105
Lactic bacteria 5,8.107 9,8.106 4,1.104 6,31.104

F. Jourjon et al.
Acetic bacteria 1,62.107 5,4.106 7,00.102 6,31.103
Yeasts 1,8.106 7.106 2,3.103 1,6.102

exists in the microbiological composition of the winery effluents within a single site. On the
contrary, when the influx of winery effluents is weak, the various micro-organisms follow
the same evolutionary profile.

Comparison of microbiological composition of winery effluents over two consecutive years

Two years of monitoring were accomplished during the harvesting season in the Pays
Nantais (Ventilated storage and cascade LS). In Table 2 we see the results of the “ventilated
storage” system for two dates of sampling and for two different years.
No matter what the year, at the beginning of the harvest-winemaking (mid-October), we
observe a high quantity of lactic bacteria, acetic and yeasts, and a negligible number of
aerobic 22 and 37 C bacteria; and the anaerobic bacteria are dominant, whereas the number
of acetic and lactic bacteria has diminished. These observations hold true for both years.
From one year to the other, we therefore note that the same phenomena occur with, at
the beginning of the harvest, an important quantity of lactic bacteria, acetic and yeasts,
whereas at the end of the winemaking period, the 22 C and 37 C aerobic flora become
dominant.

Analysis of the amounts of faecal contamination indicators in the winery effluents

Table 3 presents the averages of the different faecal contamination indicators obtained in
the 4 monitored sites.
These different micro-organisms are small in quantity no matter what system is used or
what the monitoring period is. No norms exist concerning the presence of these bacteria
in residual industrial waters, but in alluding to studies already done (Moncault, 2003), we
estimate the quantity of enterococci at between 1 and 10 UFC/ml in a natural lagoon, and
between 10 and 100 UFC/ml of E.coli. In an urban water treatment facility (Faby, 1997),
the values of an untreated water are around 105 to 106 UFC/ml for enterococci and 106 to 107
UFC/ml for faecal coliforms.
Our results are a lot closer to the concentrations of a natural lagoon. We can therefore
conclude that these treatment systems carry a low risk of contamination by faecal germs. It is
difficult to interpret the results of the SRB spores, because these germs do not systematically

Table 3 Results of the analyses (in UFC/ml) of the faecal contamination indicators in different
treatment systems

E. coli Enterococci SRB spores

Ave Max Min Ave Max Min Ave Max Min

Ventilated storage 2.97 23 0 44.45 240 0 3328.64 20,000 1

Cascade LS 12.87 23 0 13.89 23 0 247.73 2,000 2
Lagoon 8.67 0 21 17.06 34 3 2533.33 51,000 0
CP8 cascade 57 114 0 51.67 116 17 898 2,500 0
23
 represent a pollution of faecal origin and are not necessarily pathogenic, because they are
also germs of a telluric origin.

Correlations between the results of microbiological and physical-chemical analyses

It appears interesting to research correlations between physical-chemical parameters and
microbiological ones in order to see if physical-chemical indicators of the presence of certain
micro-organisms in winery effluents exist.
F. Jourjon et al.

Taking the strong variations in the physical-chemical and microbiological composition of

the effluents into account, it seemed pertinent to us to make correlations for each separate
period using a principal component analysis (PCA).
The variables retained are COD, Escherichia coli, enterococci, 22 C aerobic flora (AF 22),
37 C aerobic flora (AF 37), anaerobic flora (ANA F), acetic bacteria (ACE) and lactic
bacteria (LAC), pH, and yeasts.

Harvesting–winemaking period
The correlations were established using the results obtained for the cascade LS and the
ventilated stocking systems (Figure 4). No matter what the treatment system, the results
coming from the PCAs are comparable. We will therefore represent a single PCA which will
take the two monitored sites into account and for both years.
Principally for 1–2 (which represent 65% of the information), we observe three groups
of variables. On the first axis, horizontal, the “COD”, “acetic bacteria”, and “lactic bacteria”
are positively correlated among themselves. A weaker correlation is also observed with the
“yeast” and “E. coli”variables.
These five variables (Group 1) are negatively correlated with pH (Group 2). On the
second axis, vertical, other positive correlations (forming Group 3) are obtained between
“anaerobic flora”, “enterococci”, “aerobic F 22 C”, and “aerobic F 37 C”. These last two
variables are connected during this period (R: 0.65). The variables of Groups 1 and 2 are
independent of those from Group 3. The correlations observed during this period between
the cited groups are all highly significant (a=1%).
Component 2 – Inertia : 29.6%

AF22 Entero
AF 37
ANA F

COD
pH
Yeasts
E.coli
ACE
LAC

Component 1 – Inertia: 35.9%

Figure 4 Circle of the correlations of the studied variables for the cascade LS and ventilated
24 storage systems
 Component 2 – Inertia : 25.4%
AF 22
Entero
COD
pH AF 37

F. Jourjon et al.
ACE
LAC

Yeasts
E.coli
ANA F

Component 1 – Inertia : 56.7%

Figure 5 Circle of the correlation of the studied variables for the cascade CP8 and lagoon systems

Filtration–bottling period
Given the small number of complete results per sample for the CP8 cascade system and for
the lagoon, the CPA for the period from March to July will be done with the results of the
two studied sites for this period (Figure 5).
Principally for 1–2 (representing 82% of the information), we observe four distinct groups
of variables. On the horizontal axis 1, highly significant positive correlations are obtained
between “lactic bacteria”, “COD”, “acetic bacteria”, and “aerobic F 37 C” (Group 1). The
coefficients of the correlations between these variables are always greater than 0.92. This
group significantly opposes (a=5%) to pH (Group 2).
On the vertical axis 2, the variables “anaerobic Flora” and “yeasts” are positively
correlated between themselves (a=10%). This third group of variables are opposed to a
lesser degree to “aerobic F 22 C”, (Group 4).
The aerobic flora 37 and 22 C are, in this case, independent variables (R:x0.20). No
matter what the period of the year, the variables “lactic bacteria”, “acetic bacteria” and the
“pH” are obtained. For the other bacteria listed, we do not notice as clear a tendency. We
only observe that flora aerobic 22 C can be connected or independent to the flora aerobic
37 C variable, according to the period.
The indicators of faecal contamination listed are not specifically connected to a group
of bacteria.

Conclusions
Winery effluents are therefore rich in micro-organisms, in the range from 105 to 108 UFC/ml.
The type of system has little influence on the “microbiological pollution” level, but we only
studied four systems, three of which were ventilated.
The flora composition is very dependent on the period of the year and therefore on the
activities of the winery, so certain micro-organisms will be found in certain periods, and on
the contrary, others will have a tendency to diminish and inversely. Thus, the acetic and
lactic bacteria and the yeasts are dominant at the beginning of the harvest and progressively
diminish during the year. Whereas a specific flora, the aerobic flora, becomes dominant
during the second period of the year. 25
 We see that from one year to the next observations remain identical; therefore, the
equilibrium of the flora occurs systematically and naturally.
Faecal germs are present in very small quantities in the treatment systems of winery
effluents, but occasionally the installation of a tangential filter (cascade CP8 system)
filtering 0.1 mm, can also provide a supplementary security.
In spite of the fact that physical-chemical parameters are correlated to micro-organisms,
the amount of data is insufficient to create viable and models and predictions. Thus, the COD
F. Jourjon et al.

is correlated with acetic and lactic bacteria and yeasts, but is it really a “marker” of these
micro-organisms?
Just as for the harvesting period, the pH is negatively correlated to the yeasts (R=x0.77).
This study adds elements of security to the eventual sanitary risks of winery effluents
and in particular for treatment installations situated near dwellings, but it should be
completed by the analysis of other treatment sites.

References
Arcanger, P., Jourjon, F. and Réveillere, M. (1999). Analyses des flux d’eau dans les caves angevines pour une
meilleure gestion des effluents vinicoles. Revue Française d’œnologie, N 178, pp. 27–30.
Baradeau, E., Jourjon, F. and Brosseau, J.L. (1999). Impact de l’épandage des effluents vinicoles sur vigne.
Revue des œnologues, N 90, pp. 27–30.
Faby, J.A. (1997). Etudes de l’utilisation des eaux épurées en irrigation. Office international de l’eau. Etude
réalisée pour le compte du ministère de l’agriculture, de la pêche et de l’alimentation. FNDAE.
Montpellier, France.
Jourjon, F. and Arcanger, P. (1998). Caractérisation des flux d’effluents vinicoles, application aux caves
angevines. CEMAGREF (Ed.), 2ème congrès international sur le traitement des effluents vinicoles.
Bordeaux, France, pp. 299–305.
Jourjon, F., Pidoux, M. and Alonso, S. (1998). Epandage des effluents vinicoles: impact et principaux critères
technico-économiques et environnementaux d’évaluation des pratiques. CEMAGREF (Ed.), 2ème congrès
international sur le traitement des effluents vinicoles. Bordeaux, France, pp. 105–112.
Jourjon, F., Racault, Y. and Rochard, J. (2001). Effluents Vinicoles: Gestion et Traitements. Editions Feret,
Bordeaux (ISBN 2902416725).
Moncault, E. (2003). Communications personnelles sur le suivi de station d’épuration de la ville de Paris.
Syndicat Interdépartemental pour l’Assainissement de l’Agglomération Parisienne (SIAAP), Colombes,
France.
Racault, Y. and Lenoir, A. (1994). Evolution des charges polluantes de deux caves vinicoles du Sud-Ouest de
la France sur un cycle annuel. CEMAGREF (Ed.) Congrès International Effluents vinicoles. Narbonne,
France.
Racault, Y., Cornet, D. and Vedrenne, J. (1998). Application du traitement biologique aérobie double étage
aux effluents vinicoles: évaluation de deux procédés lors des vendanges en bordelais. CEMAGREF (Ed.),
2ème congrès international sur le traitement des effluents vinicoles. Bordeaux, France, pp. 197–206.
Rochard, J., Desautels, F., Viaud, M.N. and Pluchard, D. (1998). Traitement des effluents par stockage aéré:
mise en œuvre et optimisation. CEMAGREF (Ed.), 2ème congrès international sur le traitement des
effluents vinicoles. Bordeaux, France, pp. 171–177.

26