Figure 1. Location of study areas.

This study was carried out in the Boussouma and Nabadogo reservoirs (Figure 1). Nabadogo is located in the Nando region, at 11˚58'0" north latitude and 1˚28'0" west longitude, about 70 km from the capital, Ouagadougou. It was created in 1986 with a capacity of 562,602 m³ [14]. As for Boussouma (12˚54'35" N and 1˚4'58" W), it is located in the rural commune of Koubri in the Kadiogo region. It was created in 1987 with a capacity of 4,426,780 m³ [14]. Both reservoirs benefit from a Sudano-Sahelian climate, determined by a tropical climate with two main seasons: i) a rainy season from May to October marked by humid monsoon winds and ii) a longer dry season from November to April dominated by harmattan winds. The average annual temperatures are between 17˚C and 36˚C, i.e., a temperature range of 19˚C. Water levels are rarely higher than 700 mm per year ([15]: pp. 164, 188).

Data collection was performed during the dry season. The fish were sampled from November 22 to 28 in Boussouma and from December 4 to 10, 2023, in Nabadogo, i.e., 7 days of sampling per reservoir. The fish were sampled during experimental fishing using cast nets with mesh sizes of 5, 10, and 15 mm and gill nets with mesh sizes of 5, 10, 15, and 20 mm. We also used longlines for fish sampling. On average, 230 cast net throws were made per day in each reservoir during the sampling period. The gillnets, measuring 100 m in length and 1.5 m in depth, used in this study are multi-filament nets. In the water, the fishermen were moving from downstream to upstream generally by canoe and occasionally on foot (Figure 2). The gillnets were set between 4 pm and 5 pm and surveyed the next day between 8 am and 9 am. More than ten nets were used per sampling day and per reservoir. Additionally, we sampled in small water pools upstream and downstream of each reservoir.

All fish caught were sorted and then identified in the field using the West African Freshwater and Brackish Fish Identification Keys of ([16]: pp. 457, 815). Total length and body weight were measured using a millimeter-sized ichthyometer and an electronic scale with a range of 3 kg ± 0.1 kg, respectively.

Figure 2. Fishing techniques used to sample fish: throwing cast net (a), gillnet survey (b).

The fish population sampled was characterized by various parameters:

1) Species richness (Taxa_S) is the number of species recorded in the catches;

2) The Shannon-Weaver Diversity Index (H’) [17]: it allows diversity to be expressed by taking into account the number of species and the abundance of individuals within each of these species. Thus, a community dominated by a single species will have a lower coefficient than a community in which all species are codominant. The values of this index vary between 0.5 (very low diversity) and 4.5 (in the case of large samples of complex communities) [18]. It is calculated according to the following formula:

where i varies from 1 to s (total number of species in the community), p i is the proportion of individuals of each species, p i = n i / N , n i is the number of individuals representing species i, and N is the total number of individuals.

3) We calculated the index of evenness of Pielou: Equitability_J = H'/log₂(S), where H' is the value of the Shannon index and S is the taxonomic richness (Taxa_S). This index denotes a balance relationship between species and the individual richness of a sample. It ranges between 0 and 1, which expresses the absolute distribution of relative abundance of species [19].

4) The value for Simpson’s Diversity, sometimes called a Dominance Index, is calculated as 1 – D, with D = ∑ n i ( n i − 1 ) / N ( N − 1 ) , where n i is the total number of individuals of species i and N is the total number of all individuals in the community. The higher the value for this index, the higher the diversity of species.

5) We also plotted the rarefaction curve using the method from [20].

6) The length-weight relationship is a parameter that allows us to check the growth of the fish population. It is established using formula [21]:

with Lt = total length in cm; P = body weight in g; A = constant; and b = allometry coefficient.

The parameters a and b are estimated after the logarithmic transformation of the previous power function into a linear function with the formula lnP = lna + blnLt.

The value of b indicates the type of growth of the species under consideration [22].

If b = 3, then there is isometric growth (the growth in weight is equal to that in height);If b > 3, then the allometry is major or positive. The fish grows faster than it grows;If b < 3, then the allometry is minor or negative. The fish grows slower than it grows.

7) The condition factor makes it possible to assess the state of the fish’s overweight. It is useful for understanding the impacts of variations in environmental conditions on species. The condition factor is calculated from the following formula [23]:

where P is the total weight in grams, Lt is the total length in centimeters, and b is the allometry coefficient deduced from the weight-height relationship. The constant b is deduced from the height-weight relationship (the allometry constant).

The statistical analyses (Student’s t-test at the 95% threshold) were done in R and PAST software.

A total of 29 species of fish belonging to 12 families were recorded in all the experimental catches. Specifically, 24 species belonging to 11 families have been recorded in the Boussouma reservoir, while in Nabadogo there are 15 species belonging to 9 families (Table 1). The analysis of the fish populations of these reserves has highlighted the existence of 8 families and 10 species common to the two sites. On the other hand, 14 species and 3 families (Anabantidae, Danionidae and Polypteridae) were found only in the Boussouma reservoir, and vice versa for the Nabadogo reservoir, with 5 species and 1 family (Protopteridae). Two species of fish (Sarotherodon galilaeus, 51.59% and Oreochromis niloticus, 20.31%), all belonging to the Cichlidae family, dominate in Boussouma’s catches. As far as the Nabadogo reservoir is concerned, the most represented species are three: Coptodon zillii (45.21%) andOreochromis niloticus(17.99%) from the family Cichlidae and Enteromius macrops (22.92%) from the family Cyprinidae (Figure 3).

Table 1. Assemblage of fish species by study site.

Note: N: number of individuals; FR: relative frequency.

Figure 3.Relative frequency of fish species by reservoir.

Diversity indices such as the Shannon-Weaver index, Pielou’s evenness, and Simpson’s dominance, along with their confidence intervals, are summed up in Table2, and the rarefaction curve is illustrated by Figure 4. The diversity index such as Shannon-W The Shannon-Weaver diversity index was higher in the Boussouma reservoir (H' = 1.7) and lower in the Nabadogo reservoir (H' = 1.5). These values indicate that the different reservoirs are not very diverse.

Table 2. Estimate of diversity index by study site.

Note: CI: confidence intervals.

Figure 4. Rarefaction curve of fish species by reservoir.

Descriptive statistics of the seven (7) species common to both reservoirs are presented in Table 3. Indeed, three (3) species out of the ten (10) species common to the two reservoirs were excluded from the analyses because of very low numbers (<10 specimens).

Table 3.Descriptive statistics for 7 common fish species of the Boussouma and Nabadogo reservoirs.

Note: Min: minimum; Max: maximum; N: number of specimens.

Statistical analyses (Table 4) showed that the allometry coefficient varied between 2.61 (Enteromius macrops) and 3.25 (Brachyalestes nurse) at Boussouma, compared to 2.30 (Sarotherodon galilaeus) and 3.14 (Schilbe intermedius) at Nabadogo. In the Boussouma reservoir, one species (Brachyalestes nurse) out of the 9 common species showed positive allometric growth (b > 3; p < 0.05), i.e., they grow faster in weight than they grow in size. As far as the Nabadogo reservoir is concerned, one species, namely Schilbeintermedius, showed positive allometric growth (b > 3; p < 0.05). Negative allometric growth (b < 3; p < 0.05) was observed in two species in the Boussouma reservoir, while in Nabadogo only one species (Enteromiusmacrops) also showed negative allometry. All other species showed isometric growth (b ≠ 3; p > 0.05).

Table 4. Estimated length-weight relationship parameters for 9 common fish species of the Boussouma and Nabadogo reservoirs.

Note: CI: confidence intervals; A+: positive allometry; A−: negative allometry; I: isometry.

The average values of the condition factor are shown in Table 5. Analysis of the table shows that 4 fish species were the most overweight in the two reservoirs: Cichlidae (Sarotherodongalilaeus, Oreochromisniloticus, and Coptodonzillii)andEnteromiusmacrops. The best condition factors were noted in Sarotherodongalilaeus in both reservoirs (3.15 ± 2.38 at Boussouma and 7.60 ± 1.80 at Nabadogo).

Table 5. Condition factor coefficient obtained for 9 fish species common to both reservoirs.