Lake Sonfonia is located near the suburbs of Foulamadina and Sonfonia, the landmarks general university lansana county of Conakry, the Blue zone of sonfonia. The length of Lake Sonfonia is about 2.5 km and 100 to 200 m wide in places. The depth varies between 6 to 7 meters in the rainy season and between 3 to 4 meters in the dry season. The lake is fed by a natural source of groundwater, it houses the drinking water treatment point of the SEG (Figure 1 and Figure 2).

The sampling was carried out at six stations according to human activities and pressure from local populations. Each station has been geo-located as shown in Table 1.

At each station, two sampling campaigns were carried out during low water periods corresponding to the dry season March 2021 and two others during flood periods corresponding to the rainy season August 2021. The collection, transport and storage of samples were in accordance with the protocol defined by [7] and [8] (Figure 3).

The enumeration of the germs sought was done by the membrane filter technique and the culture of the germs was carried out on cardboard culture media (NKS) (see Figure 4: filtration device with aspirator and culture media on cardboard). The following culture media were used for sprout incubation: Azide for faecal streptococcal (SF) (Figure 5 and Figure 6); isolation Chromocult for the isolation of Total Coliforms (TC), Fecal Coliforms (CF) and Escherichia coli (E. coli) (Figure 7 and Figure 8) and finally Bismuth Sulfite for the isolation of Salmonella.

Before use, the SCN are moistened with 3.5 ml of distilled water using a pasteur pipette.

The 250 ml water sample to be analyzed was diluted successively: (dilution factor 1/10, 1/100 and 1/1000). 100 ml of water was filtered through a membrane filter (0.45 micron; dark green gridded green) or (0.45 micron; black grid white).

All bacteria present in the sample are retained on the surface of the membrane, which is then deposited on the sterile NKS using forceps. The NKS incubated in the Petri dishes with the lid are deposited in the incubator. The reading is made in 24 hours or 48 hours at a temperature varying between 22˚C - 44˚C.

Faecal streptococci have been counted by counting typical colonies of red to red-brown and smooth-edged (Figure 6), faecal coliforms form pink-red colonies (Figure 8). Escherichia coli grew as dark blue colonies at violettes.la confirmation of Escherichia coli was made by the addition of a drop of Erlich kovacs reagent. The principle of this reaction is based on the ability of Escherichia coli to split tryptophan into an organic compound indole thanks to the enzyme tryptophanase. This compound reacts with the reagent dimethylaminobenzaldehyde, to form a dark red coloration, indicating the presence of indole and thus confirms the Indole/Tryptophanase positive character of the Escherichia coli colony.

Salmonella will form lightly coloured colonies with a brown to black centre surrounded by a black halo (fish’s eye). All these operations take place under the hood in the presence of busen burner flames.

The number of colonies counted is expressed in colony-forming units per 100 ml of water (UCF/100 ml). The analyses of the water samples were carried out at the Guineo-German Medical Laboratory BP: 4529 Conakry E-mail: Info@lga-guinee.com.

The detection of faecal pollution in a natural environment is usually done by looking for germs indicating faecal contamination, that is to say specific bacteria of the intestinal flora, which are not necessarily themselves pathogenic, but whose presence indicates the existence of faecal contamination, therefore an epidemiological risk. Total coliforms (TC), faecal coliforms (CF) and faecal streptococci (FS) are the main indicators of faecal contamination. It should be noted that the use of total coliforms as an indicator of faecal contamination is increasingly contested because of the heterogeneity of this group of bacteria, some species of which may be of waterborne or telluric origin. On the other hand, the species Escherichia coli, which belongs to the group of faecal coliforms, seems to be one of the best indicators of a potential health risk [9] .

The analysis in Table 2 and Table 3 shows: The levels of bacteria indicative of faecal contamination differ from one station to another and according to the sampling period.

During the dry season with the low water flow high loads of bacteria were observed: 3000 CFU/100/ml of SF at the station (YAT); 450 CF CF/100 ml (USF); 50,000 CFU/100 ml CT (YAT) and 1200 CFU/100 ml E. coli (SOC). On the other hand in the rainy season the maximum values observed are respectively: 2600 CFU/100 ml of SF at the station (YAT); 78 CFU/100 ml CF (SOC); 9600 CFU/100 ml CT (SEG) and 802 CFU/100 ml E. coli (SOC).

The ratio R = CF/SF (the number of faecal coliforms and the number of faecal streptococci) in the different stations does not clearly indicate the origin of faecal pollution. Sometimes faecal pollution is strictly animal due to the presence of animal breeding such as oxen near the lake or even other animals such as (margouillats, migratory birds, cat, dog etc.), are permanent there. Their excreta will contaminate the water body with runoff, rainwater and leaching from agricultural land loaded with manure. Sometimes strictly human because of domestic wastewater, but also the transformation of the lake shore into a defecation site for the local population. Some reports indicate a mixed origin predominantly animal, others mixed predominantly human origin at the end of other uncertain origin [10] .

The presence of these indicators in water indicates faecal contamination [11] . The presence of E. coli in the various stations confirms this contamination by feces but also the possible presence of other pathogenic germs [12] . The stations (YAT) and (SOC) close to residential areas receive discharges of sewage, domestic wastewater has recorded peaks in bacterial load values.

The comparison of the levels of faecal contamination bacteria of the present study with a previous one carried out in 2018 on the waters of the same lake, indicates an increase in the content of fecal coliforms and faecal streptococci that exceed the maximum concentration allowed by the WHO and MDDEFP (62 CFU/100 ml in SF and from 200 to 1000 CFU/100 ml in CF), for surface waters, classifying the waters of Lake Sonfonia as of poor bacteriological quality.

The search for pathogenic bacteria such as salmonella in water usually requires a concentration step. Since Salmonella may be present in small numbers and have been altered in the aqueous environment, its search in water requires pre-enrich- ment [13] . Analysis of the results indicates a notable absence of salmonella. The lack of reagent for pre-enrichment could probably explain in part the notorious absence of salmonella at the various stations.

Stata 15 software was used for statistical analysis of the data. Because the sample size is small, the Shapiro-Wilk normality test indicated that the variables studied are not normally distributed. It was preferable to use a non-parametric test. For this we used the Mann-Whitney U-test to compare the value of the means of each of the variables observed during the two seasons as shown in Table 4.

The test results reveal that, on average, there are statistically significant differences between the values of the two seasons (March and August) of the following

parameters: SF (22˚C in 24 h), CT (22˚C in 24 h) and CF (44˚C in 24 h) in their respective incubation conditions. On the other hand, it appears from our study that there are on average no statistically significant differences between the values of these two seasons (March and August) for the parameters SF (37˚C in 48 h), CT (37˚C in 48 h) and E. coli (22˚C in 24 h and 37˚C in 48 h) with indicated incubation conditions.