Breeding pattern of Oreochromis niloticus (Linnaeus, 1758) versus native congeneric species, Oreochromis macrochir (Boulinger, 1912), in the upper Kabompo River, northwest of Zambia

Abstract Investigating the determinants of the reproductive biology of fishes is an essential component of fisheries research. Tilapia breeding patterns were investigated to determine the impact of non‐native Oreochromis niloticus on the native congeneric Oreochromis macrochir in the upper Kabompo River in the Northwest of Zambia using the gonadosomatic index and the sex ratios. Oreochromis niloticus was the most abundant fish caught (221, 63.5%) than O. macrochir (127, 36.5%). Results showed that the overall gonadosomatic index means of O. macrochir in both sections were similar. Oreochromis macrochir bred in December and February–March, with no reproduction in June. However, O. niloticus in the invaded section indicated all year reproduction through reduced spawning in May–June, with increased spawning activity in February–March. The sex ratio (females: males) was 1:1.3 and 1:1.7 for O. niloticus and O. macrochir, respectively, and both significantly deviated from the sex ratio of 1:1 (ꭓ2 = 8.42 and 9.37, p < .05). Our study has revealed that O. niloticus was able to spawn across all sampled months with a 23% higher breeding population than O. macrochir, which might explain the suppression in the abundance of native O. macrochir. Due to the superior breeding patterns of O. niloticus, fisheries, wildlife, and aquaculture practitioners need to make contingency plans to alleviate its impacts further downstream of the Kabompo River.


| INTRODUC TI ON
The introduction of invasive alien species in an ecosystem poses a greater risk to cause ecological disruption in stable environments once it establishes (Ellender et al., 2018;Jere et al., 2021;Lowe et al., 2018;Vicente et al., 2011). This has resulted in possible harmful relationships between invasive alien species and native species (Le Roux et al., 2020;Zengeya et al., 2015). Anthropogenic activities are the leading cause of the intentional and unintentional introduction of invasive alien species over the world (Fonseca-Alves et al., 2011).
Tilapias (Cichlidae) are among the African native species that have been introduced to other continents around the world, because of their ability to reproduce frequently and attainment of early sexual maturing (Azua et al., 2017;van Wilgen et al., 2020). On the contrary, tilapias can have negative effects on ecosystems, causing serious modification on the aquatic community dynamics and further reducing biodiversity (Gozlan et al., 2010;Groombridge, 1992;Jere et al., 2021;Le Roux et al., 2020). Generally, the introduction of invasive alien species threatens the stability of aquatic environments, triggering competition and thereafter suppressing the native communities leading to the extinction of the native species (Gozlan et al., 2010;Lowe et al., 2018;Zengeya et al., 2020).
In sub-Saharan Africa, most freshwaters have been invaded by alien fish species (Zengeya et al., 2015). The characteristics of invaders include early maturity and a fast growth rate that makes them successful and impact negatively the diversity of native species (Beyruth et al., 2004;Espinola et al., 2010;Jere et al., 2021). In the genus Oreochromis, males build nests for spawning and develop secondary sexual structures (Beyruth et al., 2004;Gozlan et al., 2010;Lowe et al., 2018;Turner & Robinson, 2000). These attributes make invasive alien species to be more dominant in areas outside their normal range than native species and may create an ecosystem population imbalance. This population imbalance of native species may cause the native ecosystem to be more vulnerable to resist further impacts of invasion . Moreover, invasive alien species are difficult to control, in many instances where chemical or physical means were employed; no significant eradication has been achieved (Groombridge, 1992;Jere et al., 2021). This could suggest that a strict policy on the introduction of alien species is the best means to avoid control invasion.
Nile tilapia, Oreochromis niloticus (Linnaeus, 1758), is among the most preferred fish species for aquaculture purposes, despite its ability to reduce local biodiversity through competition with native species (Zengeya et al., 2020). However, some countries have introduced O. niloticus to increase local fish diversity and supply food fish to local people (Azua et al., 2017). The species is preferred for aquaculture because of its ability to tolerate a wide range of environmental conditions (Grammer et al., 2012;van Wilgen et al., 2020). In Zambia, O. niloticus was introduced in the late 1980s by the Department of Fisheries for research and development in aquaculture (Bbole et al., 2014;DoF, 2018;FAO, 2012;Kenzo & Mazingaliwa, 2002).
To understand the impacts of invasive alien species, breeding patterns is one way in which these impacts can be comprehended.
To achieve this, gonad staging using gonadosomatic index (GSI) is the accurate and reliable means to determine the reproductive biology of fish (Rodi & Mackie, 2001). This method is an important tool for elucidating breeding differences between congeneric fish species (Rodi & Mackie, 2001;Gozlan et al., 2010;van Wilgen et al., 2020). Therefore, results provided by the gonadosomatic index method can be used to determine the spawning events of species. This is an important predictor of the impacts of invasive alien species in the aquatic ecosystem (Lowe et al., 2018;van Wilgen et al., 2020).
The impacts of invasion by exotic species vary spatially and temporarily and tend to be context-dependent in the subtropical lotic systems (Rouget et al., 2002). One way in which exotic species impact indigenous species is through rapid spawning outside its normal range, which may lead to dominance and suppression of native species. It is known that there is a viable population of exotic O. niloticus present in the upper Kabompo River, but its reproductive biology impacts on the native species have not been reported (Bok & Bills, 2012;DoF, 2018). In this study, we investigated the breeding patterns of native O. macrochir and exotic O. niloticus to better understand reproductive activity interaction (Lowe et al., 2018) This hypothesis was tested using the gonadosomatic index (GSI) to understand the spawning biology and sex ratio to investigate the reproductive potential for both species. This method used here gives us clues on the spawning biology of species and a good understanding of the reproductive potential of the species. The study will help to clarify the increased dominance of O. niloticus over the native O. macrochir and assist in developing a fisheries management plan aimed at reducing the impacts of fish invasion.

| Description of the study area
The study was conducted in the upper part of the Kabompo River, and its source is in the North Western Province of Zambia ( Figure 1).
The province is endowed with numerous supplies of inland flowing waters that are suitable for fish farming and fisheries. The study area stretches approximately 45 km in length from the source of the Kabompo River (latitude 25.2414 to 25.044156 E and longitude 11.8973 to −12.369120 S, respectively). The river has a natural waterfall that acts as a barrier drops-off in the river gorge at approximately 40-to 50-m elevation and disappears at 1.5km before surfacing (AES, 2014;DoF, 2018;Jere et al., 2021). Upper Kabompo is part of the Kabompo River and is one of the major tributaries of the Zambezi River. It originates at an altitude of approximately 1500 m above mean sea level (amsl) in the highlands, which forms the watershed between the Zambezi and Congo rivers. It has a water surface area of about 1473 km 2 , with abundant fish landing sites. The study site is one of the sources of brood fish for the Aquaculture Breeding Programme in Zambia (DoF, 2018). It was also chosen because it is the most active fishing area and of economic importance to the surrounding villages as a source of livelihood (Bok & Bills, 2012).

| Sampling procedure
Six sampling points were chosen at 4.5-km intervals, covering approximately 30 km of the river stretch ( Figure 1). The study area was divided into two sections: the uninvaded section with three sampling points that was not invaded by O. niloticus following the aquatic biodiversity study report by Bok and Bills (2012)  sampling was conducted daily for 4 weeks in a month, out of which two weeks were allocated in the sampling section per month. The fish were caught in the early mornings (06:00 h) and evenings (17:00 h), because in these hours, all the fish were fresh and easy to identify before spoilage takes place.
One sampling was assumed to be invaded, while the other section was uninvaded following the aquatic biodiversity report (AES, 2014;DoF, 2018). This provided a platform for comparison of the

| Fish sampling
Fish were collected separately from the runs, riffles, vegetative thicket points, and open pools using different fishing gears. The following fishing gears were used in the fish collection: gill nets ranging in mesh size from 2 to 6.5 cm with 3-m depth and 30-m length; two double-ended The caught fish species were sorted and identified where possible using the reference material by Kenzo and Mazingaliwa (2002).
A total of 1548 fish were collected, of which 921 were O. niloticus and 627 were native O. macrochir. The collected specimens were immediately sexed, after which they were stored in ice and taken to the laboratory for further examination. For the specimens that we were able to dissect in the field, their gonads were weighed to the nearest 0.01 gm, fixed in 10% formalin, and taken to the laboratory for microscopic examination.
The preserved gonad samples from the field were removed from the formalin solution in the laboratory for GSI determination. Microscopic staging using SHT was also used in the processing of the samples under a light microscope for examination of sex and gonad development from gametes. This process is important as it supplements the GSI method in determining the reproductive timing of the fish (Hilge, 1977). To do this, the whole specimens were blotted dry and reweighed to the nearest 0.01gm. Individual gonad transverse portion was removed and then processed using standard histological techniques (SHT) to give 5to 8μm sections that were stained for microscopic examination using the procedure by Harris's hematoxylin and eosin (H&E) (Hilge, 1977).
The removal of the transverse portion of gonads from the middle region of one lobe was done to determine whether uniformity exists in the development of gametes. Only gonads of 689 examined specimens that showed mature males and females represented in all reproductive stages (1-6, females and 1-5, males) were considered. These measurements were collated in a developed Microsoft Excel spreadsheet for data entry for computation of GSI and sex ratio. Gonadosomatic index (GSI) of the female and male for each species of the collected samples was determined separately using the formula.

| Statistical analyses
Data were entered into Microsoft Excel 2007 (summation, average, and percentages) for computation of GIS and sex proportion of O. niloticus and O. macrochir in the upper Kabompo River. The GSI mean data satisfied the parametric assumptions after subjecting them to the normality and homoskedastic tests (Sokal & Rohlf, 1995).
Thereafter, two-way ANOVA using interaction contrast was then used to test the differences in the means of a gonadosomatic index of both males and females of O. niloticus and O. macrochir (Sokal & Rohlf, 1995). Equally, a chi-squared test was performed to test for whether each species mean sex ratio differed from the expected sex ratio of 1:1 (Kings, 1995). All statistical tests were performed using the R software package (R Core Team, 2019).  (Figures 2 and 3, respectively). In contrast to the mean  100 males (45.3%) ( Figure 6). Females were more numerous in all the 3 sampled months for both species (Figure 2). Sex ratio was 1:1.34

| D ISCUSS I ON
One way in which exotic species impact indigenous species is through rapid spawning outside its normal range, which may lead to dominance and suppression of native species. Therefore, our study for males. The findings from a similar study on the Kafue River, the southern region of Zambia, were similar (Bbole et al., 2014;Marshall, 2011). Similarly, Lake Kariba that receives water inflow from the Zambezi River showed high catches of O. niloticus than the native congeneric species (Marshall, 2011;Zengeya et al., 2020).  Silva & Chandrasoma, 1980;Zengeya et al., 2015). Stewart (1988) and Gomez-Marquez (1998) (Gomez-Marquez, 1998;Le Maitre et al., 2020;Lowe et al., 2018).
Oreochromis niloticus is associated with environmental tolerance, early sexual maturity, and rapid colonization (Ellender et al., 2018;Pérez et al., 2006;Russell et al., 2012). The aforementioned characteristics are considered critical in facilitating a successful invasion and establishment of this fish in the river, which may ultimately affect the dominance of the native congeneric species (O. macrochir).
In this study, the observed GSI values observed in all the sampled periods are consistent with year-round spawning and are similar to the GSI values reported by Lowe et al. (2018) Fryer and Iles (1972) also cited that the sex ratio varies significantly from species to species and across months of the year; however, in the majority of cases it is relatively close to 1 in the same population. The sex ratio variation indicated in different studies points to the fact that fertilization of the eggs may have been concluded and males possibly emigrated from the spawning sites to the feeding areas leaving females behind incubating the young (Zengeya et al., 2020). The sex ratio helps to shape and maintain an ecological species population balance and may help to resist invasion by invasive alien species (Zengeya et al., 2020 We recommended that further studies understand the possibility of ecosystem-wide consequences of invasion and determine the extent to which other native fish assemblages respond throughout the continually expanding range of O. niloticus in the upper Kabompo River.

ACK N OWLED G M ENTS
We thank the Department of Aquaculture and Fisheries (Aquaculture

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that have been used in this study are available, and the Dryad data repository will be used to archive the data.