The Babagulu Forest Reserve is located approximately 50 km northeast of Kisangani, within Tshopo province in the Democratic Republic of Congo. It covers an area of 1,693.42 hectares [19]. The reserve is geographically positioned between 25˚35' and 25˚37' East longitude, and between 0˚26' and 0˚47' North latitude. It is bisected by National Road No. 4, which divides the reserve into two distinct sections ().

Figure 1. Location of the Babagulu Forest Reserve. Legend = Babagulu Forest Region (Région forestière de Bagulu), DR Congo (RD Congo), Tshopo Province (Province de la Tshopo), National Road No. 4 East (Route Nationale N˚4 Est), Study Area (zone d’étude), Boundary of the Babagulu Forest Reserve (Limite de la reserve forestière de Bangulu), Northern Block and Southern Block (Bloc du Nord et Sud). (Source: QGIS software (version 3.12))

The climate of this reserve is classified as equatorial Af according to Köppen’s classification [20][21]. This is characterised by hot and humid conditions with dry seasons that are scarcely discernible. The region is subject to substantial rainfall and temperature variability, receiving between 1500 and 2000 mm of annual precipitation, with an average of approximately 1750 mm. The temperatures fluctuate between 20˚C and 30˚C, maintaining a mean annual temperature of around 25˚C [20][21]. The vegetation of this reserve is predominantly composed of monodominant primary forest, chiefly characterised by the presence of Gilbertiodendron dewevrei (De Wild.), (Fabaceae) [22]. In addition to this dominant forest type, the reserve encompasses a mosaic ecological zone, including fallow land, young and mature secondary forests, as well as mixed primary forest formations [23]. The hydrographic network of the Babagulu Forest Reserve comprises approximately ten watercourses. In the northern sector, the principal streams include Minamba, Makombo, Mangoingoi, and Bombe. The southern sector is traversed by Madje, Maseko, and Kinombe streams (). Other notable watercourses within the reserve include Makala, Mafwaga, Mboa (also referred to as Pembe), Ngayani, and Kokondodja.

Figure 2. Location of the main watercourses in the Babagulu Forest Reserve. Legend = Congo River and tributaries (Fleuve Congo et rivières), Streams (Ruisseaux), Streams within the reserve (Ruisseaux de la Réserve). (Source: QGIS software (version 3.12))

Shrimp specimens were collected from the Kinombe and Maseko rivers. These sites are described below.

2.1.1. Kinombe

Kinombe is located in the southern sector of the Bagulu Forest Reserve, positioned at 00˚27.032' North latitude and 025˚35.414' East longitude, at an elevation of 465 m. Due to its relative isolation, the site experiences minimal anthropogenic disturbance. To facilitate, four sampling stations were established two upstream and two downstream of the main access routes to the watercourse, each spaced 50 m. Vegetation cover in the area is estimated at 80%, while the riverbed is characterised by accumulations of branches and deadwood, assorted plant debris. Substrate composition varies across sample points and includes pebbles, silt, sand, gravel, and stones. The riverbanks are predominantly clay-based and support riparian vegetation. Average depth: 0.6 m, Average riverbed width: 7.4 m; Average dissolved oxygen: 5.68 mg/l; Average pH: 7.81; Average electrical conductivity: 49.31 µS/cm; Average water temperature: 23.30˚C; Average velocity: 0.29 m/s and the water is clear.

2.1.2. Maseko

This watercourse is situated within the southern sector of the Babagulu Forest Reserve, positioned at 00˚27.282' North latitude and 025˚35.489' East longitude, at an elevation of 453 m. It lies in a densely forested area that remains undisturbed by agricultural practices or logging. Four sampling stations were established two upstream and two downstream of the primary access routes to the watercourse, with each station spaced 50 m. The vegetation cover is notably dense, estimated at approximately 85%, and the riverbed contains a substantial accumulation of deadwood, branches, and plant debris. The substrate predominantly consists of silt and mud, although patches of sand and gravel are also present. The banks are generally composed of clay and are lined with riparian vegetation. Hydrological and physicochemical parameters recorded at the site include Average depth: 0.4 m; Average riverbed width: 5.2 m; Average dissolved oxygen: 5.31 mg/l; Average pH: 7.43; Average electrical conductivity: 32.33 µS/cm; Average water temperature: 23.36˚C; Average velocity: 0.26 m/s. The water is generally clear, though it exhibits a slight turbidity due to suspended fine sediments.

Specimens of M.sollaudii were collected from the Kinombe and Maseko rivers over a three-month period, spanning August to October 2021. Sampling was carried out once each month, and at each sampling station, three distinct fishing techniques were employed: Funnel trap fishing, hand net fishing (targeting turbid trickles) and manual scooping was performed using a 5-litre basin to capture shrimp specimens.

Prior to fishing activities or the employment of nets, a range of environmental parameters was recorded at each sampling station. Geographical coordinates (latitude, longitude, and altitude) were obtained using a GARMIN GPS device (GPSmap 62stc. Physicochemical measurements including water temperature, pH, electrical conductivity, and dissolved oxygen concentration were taken using a HACH multiparameter instrument. Additional site-specific recorded following: Riverbed width measured using a 50-metre measuring tape; Water depth determined using a measuring stick and tape; Substrate composition: comprised of sand, gravel, silt, deadwood, and herbaceous vegetation; Bank characteristics: predominantly Clay-based with riparian vegetation and Canopy cover (shading): assessed through direct visual observation.

below illustrates the sexes of M.sollaudii.

(a)

(b)

Figure 3. Males (a) and females (b) shrimp.

At each sampling station, specimens were placed into jars and grouped according to their size and morphological characteristics for subsequent identification. Taxonomic identification was conducted using established keys and reference materials, including those by [24]-[26].

Morphometric measurements were performed on 146 Macrobrachium sollaudii specimens using a precision calliper. The parameters recorded: Carapace Length (CL); Abdominal Length (AL); Rostrum Length (RL); Segment Length (SL); Total Length (TL) and Uropod Length (UL). In addition, the total weight (TW) of each specimen was measured using a calibrated balance Sartorius Universalis mark. All samples were preserved in jars containing 70% alcohol and appropriately labelled with the corresponding collection station and date of capture.

Measurement data were processed using Microsoft Excel (Windows version 2016), where values were summarised as means and standard deviations. To further analyse the data, the vegan package in R. (Core Team R) was employed to calculate multivariate dispersion (variance) and distance to the centroid. Prior to performing ANOVA and regression analyses, dataset was evaluated to ensure compliance with the requisite statistical assumptions. Residual normality was assessed using the Shapiro-Wilk test, and Levene’s test was employed to examine the homogeneity of variance (homoscedasticity) across groups. This analysis enabled the assessment of whether the measured variables within a given group differed significantly from those in other groups, and relative distance of each group from the centroid, representing the point of convergence. Subsequently, a multiple linear regression analysis was conducted in R to determine the extent to which each morphometric parameter influenced the others.

A total of 146 M.sollaudiispecimens were collected during the study, comprising 93 females and 53 males. As shown in , species abundance was markedly higher in Kinombe River, where 62.3% of the specimens were female and 27.3% male. In contrast, the Maseko River accounted for only 1.3% females and 8.9% males indicating a significantly lower population density in that locality.

Table 1. Number of specimens per river.

Table 2 clearly indicates that male specimens collected from the Maseko River exhibited the highest average total length (54.45 ± 13.82 mm), whereas those from the Kinombe River recorded the lowest average total length (35.6 ± 9.51 mm). This pattern was consistent across all measured morphometric parameters, with males from the Maseko River consistently displaying higher values, in contrast to their counterparts from the Kinombe River, which showed comparatively lower measurements.

Table 2. Measurements of M. sollaudiispecimens, considering sex.

*AL: Abdominal Length, CL: Carapace Length, RL: Rostrum Length, SL: Segment Length, TL: Total Length, UL: Uropod Length, TW: Total Weight.

illustrates the degree of similarity and divergence in the morphometric parameters of M.sollaudiispecimens collected from the Maseko and Kinombe rivers. While individuals from sires exhibit comparable morphometric traits, they differ markedly in terms of overall variance and group centroid positioning. Specimens from Masako River display greater uniformity, while those from Kinombe River show a broader range of variation. These findings suggest that environmental or ecological conditions may influence the morphometric expression of populations, even within the same species.

Figure 4. Multivariate Dispersion (Variance): Resemblance among group members. * Maseko in black and Kinombe in red.

It is noted that all morphological parameters showed a minor variation in the pairwise comparisons. The p-value presented below the diagonal, alongside the permuted p-value above the diagonal, indicates differences that were not statistically significant ().

Table 3. Pairwise comparisons.

Analysis of variance ANOVA was applied to assess whether the dispersions (variances) between the two groups (Maseko and Kinombe) differ significantly. The results showed a clear difference in group dispersion ( and ). further illustrates that the distance to the centroid between the two groups is substantial, which accounts for the observed variation in morphometric measurements among individuals from the Maseko and Kinombe rivers. This distinction is statistically supported by the p-value presented in .

Figure 5.Variance between groups (distance to centroid).

The distance between the centroids, representing the average position of each group within morphometric space, indicates a marked distinction between the two populations. This interpretation is supported by the results of the ANOVA (), which shows a statistically significant difference between groups (p = 0.02).

Table 4. Analysis of variance applied to the two groups.

*Signif. codes: ‘*’ 0.05.

The multiple linear regression analysis revealed that sex had no significant influence on any of the morphometric parameters. However, each parameter exhibited a strong correlation with other length and weight-related variables ().

Table 5. Multiple linear regression analysis.

*Signif. codes: ‘***’ 0.001.

Supplementary provides the complete results of the multiple linear regression analyses, detailing the regression coefficients, standard errors, coefficients of determination (R²), and adjusted R² values for each model. The findings confirm that sex does not exert a statistically significant effect on any of the morphometric parameters of M.sollaudii. In contrast, robust linear associations were observed between the length and weight variables.

Table 6. Full multiple regression outputs.