Assessing the Rigidity of the River Continuum Concept to Fish Species in a Fluvially Dominated Southern African Coastal System


 A continuous gradient of physical and biological processes exists within a river system from headwaters to mouth, according to the River Continuum Concept. In this study, the River Continuum Concept was used to compare patterns in fish assemblages and diversity trends to Remane predictions. The findings show that species diversity is low near the estuary's mouth and increases upstream. While the presence of several marine and estuarine species increased species diversity upstream in the river due to the dominance of freshwater fishes, it decreased downstream due to the presence of several marine and estuarine species. These patterns are consistent with the biogeographical trend of decreasing species richness along the South African coast from east to west. Based on the findings of this study, the River Continuum Concept ignores plasticity, which is evident in species that use transitional waters. As a result, the concept is insufficiently valid for the Orange River Estuarine Continuum, and a new concept known as the “River-Estuarine Continuum” has been proposed.


Introduction
Some of the ecological models used to describe the longitudinal distribution along estuarine and riverine reaches are the River Continuum Concept (RCC) and the Remane model, respectively. The River Continuum Concept describes changes in biological processes that occur along the course of rivers from their headwaters to their mouths while remaining connected throughout (Vannote et al. 1980). Whereas, the Remane model describes how species diversity changes along the transitional salinity gradient (Attrill, 2002; Whit eld et al. 2012). As a result, the River Continuum Concept was adopted to examine its usefulness in predicting the behaviour of shes in a uvially dominated southern African river system. Changes in salinity conditions along the river system drive changes in communities (Isom 1986) by controlling the distribution of species along a habitat continuum (Barnard 1998; Barletta et al. 2005).
Fishes that utilize an estuarine environment exhibit a wide range of environmental tolerances and preferences as a result of adaptation to these highly variable environments (Kennish 1990). However, several sh species may be subject to physiological limitations resulting in temporal changes in species composition, distribution, and abundance ( The river-continuum concept combined various ecological changes along the river to create a uni ed description from the headwaters to the mouth. The concept attempts to categorize species but ignores species plasticity, particularly for sh species that live in transitional waters. The Orange River Estuary (ORE) is home to a variety of freshwater, marine, and estuarine species (Lamberth 2003;Naesje et al. 2007; Ramollo 2011). As a result, ORE provides a perfect system to test the rigidity of the river-continuum concept. Therefore, this study aims to 1) assess how freshwater, estuarine and marine sh species are structured along the river-estuarine system, and 2) examine the rigidity of river continuum predictions in the Orange River system.

Overview of Remane predictions
Adolf Remane created a diagram in 1934 to depict changes in benthic invertebrate diversity along a salinity gradient within the Baltic system which is a brackish sea (Attrill 2002). Vannote et al. (1980) expanded the diagram by including shes in the original diagram, in an attempt to describe structures and synthesize a set of general hypotheses about communities along a river system. However, the diagram has received criticism due to its several limitations (Hedgpeth 1967;Odum 1988;Barnes 1989;Hudson 1990;Attrill 2002;Whit eld et al. 2012). Despite the limitations, the model has been recognized as a textbook model in stream ecology (Cummins et al. 1984;Attrill 2002;Statzner and Higler 2008). Based on the model, the diversity of freshwater taxa declines rapidly at salinities ranging from 0.5 to 5, with the minimum species richness occurring at 5 to 7. At a salinity of 6, estuarine species predominate, whereas, at a salinity of 7, marine species predominate (Remane 1934).
Over the years, several scholars have been modifying the diagram to present a generalized model for describing diversity trends in estuaries. Hedgpeth (1967), expanded the diagram to include the entire salinity spectrum. Hudson (1990) changed the shape of the original model by depicting the freshwater biota as having lower species diversity than the marine biota, as well as extending the presence of estuarine species beyond the 20 salinity limit. Whit eld et al. (2012) presented a combined revised model that demonstrates the connection between salinity and faunal trends in estuaries. Besides the revised model, its applicability to the Orange River Estuary, particularly given that it is a large uvially dominated estuarine system, is unknown.

Study area
The Orange River ows over 2,300 km from its origin in the highlands of Lesotho downstream into the

Field sampling
Fieldwork was carried out during the typical high-ow (December-March) and low-ow (August-October) seasons (van Langenhove et al. 1998). Within a 15-year period, sampling was carried out in eight different years (2004, 2005, 2012, 2013, 2015, 2016, 2017 and 2018). Two sampling trips were conducted biannually for all years except 2012, 2013, and 2016 during the high-ow period, and also in 2017 during the low-ow period. During any of the sampling trips, however, no high-high ow was observed. This was due to river system regulation upstream by the Gariep and Vanderkloof dams which signi cantly reduced the amount of water owing downstream to the estuary (Bremner et al. 1990; Anderson et al. 2003). Throughout this study, the Orange River Estuary remained open (2004-2018).
A beach-seine (30 x 2 x 10 m) was used for sampling at 18 sites from the mouth of the estuary to 35 km upstream (Fig. 1). The mesh size of the net is 10 mm in the centre, including a cod-end (bag), and 10 m of 15 mm stretched mesh size in each of the wings. The distances between sites vary greatly, and they were chosen to re ect a progression from marine to estuarine to freshwater in uences as well as accessibility to the riverbed. Fish caught were identi ed to species level (Smith and Heemstra 1986; Skelton 2001), measured and counted at each site. After recording shes were released alive into the water where possible. All applicable guidelines for the care, collection and use of animals were followed in accordance with the ethical standards of Nelson Mandela University (NMU, South Africa). Speci cally, the ethics clearance was granted by the Research Ethics Committee (Animal), Nelson Mandela University. The ethics clearance reference number is A18-SCI-ZOO-002.
The number of sh species that were collected one haul at each site was recorded. Also, salinity (expressed as practical salinity units) was measured at each site, from the middle of the water column, using a Yellow Spring Instrument v6920/EXO 1 multi-parameter probe. The river-estuarine area was divided into salinity zones based on an adaptation of the Venice system of South African salinity zones in estuaries (freshwater: 0-0.49 PSU; oligohaline: 0.5-4.9 PSU; mesohaline: 5.0-17.9 PSU; polyhaline: 18.0-29.9 PSU; euryhaline: 30.0-35.9 PSU; and hypersaline: ≥36 PSU) (Strydom et al. 2003).
Moreover, shes recorded were grouped into estuarine associations following Potter et al. (2015). Many researchers who research in most southern African estuaries use this approach. The marine category is divided into three guilds (marine straggler, marine estuarine-opportunist, and marine estuarinedependent), the estuarine category is divided into four guilds (solely estuarine, estuarine & marine, estuarine & freshwater, and estuarine migrant), the diadromous category is divided into ve guilds (anadromous, semi-anadromous, catadromous, semi-catadromous and amphidromous), freshwater category consists of two guilds, namely the freshwater straggler and freshwater estuarine-opportunistic (Potter et al. 2015).

Data analyses
The diversity indices including Shannon-Wiener diversity (H') and the number of species (S)) were calculated using PRIMER v6 statistical software package among seasons, years, and sites (Clarke and Warwick 1994). Moreover, salinity levels measured at each site were categorized into salinity zones based on an adaptation of the Venice system (Strydom et al. 2003).
The assumptions of normality and homogeneity of variance were tested using a normal probability plot and Levene's Test. All diversity indices data revealed violation of both normality and homogeneity of variances. Mann-Whitney U-test (Zar 1999) was used to test whether diversity indices (H' and S) differed between high-ow and low-ow season. Kruskal-Wallis tests were performed separately on species diversity indices to test for differences between years and salinity zones. Stacked area diagrams were created in Microsoft Excel 2013 to show the distribution of estuarine associations based on Potter et al. (2015), as well as a salinity continuum based on the revised Venice system for salinity zone classi cation (Strydom et al. 2003). The signi cance level for all tests was set at 0.05 alpha.

Fish species composition
Over the eight-year study period, a total of 26 086 sh were caught, representing 30 species. According to the previous study by Nashima et al. (2021), the majority of sh species belongs to the freshwater category (47%) based on estuarine associations, with 11 freshwater stragglers and three freshwater estuarine-opportunist's species. The estuarine category (20%) included one solely estuarine species, ve estuarine and marine species, and no estuarine and freshwater or estuarine migrant species. The marine category (33%) included four marine stragglers, three marine estuarine-opportunists, and three marine estuarine-dependents.

Seasonal and spatial trends in species diversity
The number of species (S) and Shannon-Wiener diversity (H') differed signi cantly between the high-ow and low-ow seasons (S: Mann-Whitney U-test, z = -6.88, p < 0.001; H': Mann-Whitney U-test, z = -5.01, p < 0.001). Both diversity indices depicted high values during the high-ow season (Table 1). Species diversity indices showed signi cant differences among salinity zones but not years. Overall, both showed an increasing trend upstream into freshwater ( Fig. 2; Table 1).

Estuarine species association along the river-estuarine continuum
This study recorded a salinity range of 0.1 to 34.9. The dominance of freshwater species in freshwater during the high-ow and low-ow seasons was depicted by a rapid decline toward oligohaline and then more saline water (Fig. 2). Moreover, the majority of freshwater species were restricted to salinities below mesohaline conditions (5.0-17.9). During the high-ow season, Tilapia sparrmanii was the only freshwater species found in polyhaline water (2005 and 2007). During the low-ow season, no freshwater taxa were recorded in either polyhaline (18.0-29.9) or euhaline (30.0-35.9) ( Table 2). The highest species diversity values in the marine category were recorded in oligohaline waters during a high-ow season (Table 1), with three marine estuarine-opportunists (C. richardsonii, Pomatomus saltatrix, and Rhabdosargus globiceps), three marine stragglers (Lithognathus aureti, Austroglossus macrolepis, and Chelidonichthys capensis), and two marine estuarine-dependent (Lichia amia and Mugil cephalus) species. The absence of marine estuarine-dependent species (M. cephalus and Lithognathus lithognathus) in salinities above oligohaline was notable ( Table 2).
Chelon richardsonii and M. cephalus were the only two marine species with unrestricted distribution into freshwater reaches (Table 2). Other marine species, such as L. amia, Chelidonichthys capensis, L. aureti, and Pomatomus saltatrix, were only found up to the oligohaline zone (Table 2). Despite this, all other species, regardless of origin, were underrepresented in the Orange River Estuary in terms of percentage occurrences, except for C. richardsonii (Nashima et al. 2021).
Freshwater and all other salinity zones (i.e. oligohaline, mesohaline, polyhaline and euhaline) had different numbers of freshwater species, according to the Kruskal-Wallis pairwise test. Whilst for marine and estuarine categories, there were no differences in the number of species among salinity zones (Fig. 3). Gilchristella aestuaria, a solely estuarine species, and C. richardsonii, a marine estuarineopportunist, were widely distributed across salinity zones. Overall, the number of species recorded in the ORE, showed the dominance of freshwater species in fresh and oligohaline while estuarine species dominate polyhaline and euhaline waters (Fig. 3).

Discussion
The Remane model, derived from the Baltic region, has been consistently portrayed as the textbook model for estuarine diversity trends (Attrill 2002). However, the model was criticized for having too many limitations for general use in estuaries. This model has been extended and modi ed over the years to provide an alternative generalized model for describing diversity trends in estuaries (e.g. 2) The vast majority of freshwater species are restricted to freshwater habitats, with only a few taxa extending into mesohaline, polyhaline, and euhaline zones; 3) Marine taxa predominate in estuarine waters that are mesohaline, polyhaline, or hyperhaline. 4) Although in small numbers, marine species can be found in oligohaline estuaries and even freshwaters.

5)
Estuarine species are more diverse in mesohaline and polyhaline waters, but they can also be found in oligohaline, euhaline, and hyperhaline waters.
6) A reduction in species diversity from high saline to low saline waters. Also, diversity starts to decline above a salinity of about 40.
In contrast to the ORE, the system is distinct in that it has high productivity but low diversity in comparison to other estuaries in the cool-temperate biogeographical region. Furthermore, the system has a limited number of estuarine species (Brown 1959). Low species diversity corresponded to the South African coast's biogeographical trend of decreasing species richness from east to west (Lamberth 2003). The ORE is dominated by freshwater biota rather than marine and estuarine species, with species diversity declining along a salinity continuum. Based on the estuarine association categories the distribution of freshwater, estuarine and marine species differed signi cantly among salinity zones but not guilds.
The solely estuarine G. aestuaria was present among all salinity zones along the river-estuarine continuum. Freshwater stragglers dominated the freshwater zone and progressively declined from the oligohaline water toward more saline water up to a salinity of 20.70 (polyhaline), while estuarine species progressed into the euhaline water. Three T. sparrmanii were recorded in mesohaline/polyhaline water during the high-ow season. As expected, Tilapias are known as tolerant of higher salinity stress (Popma and Lovshin 1994). Three species of freshwater estuarine-opportunists Clarias gariepinus, Pseudocrenilabrus philander and Oreochromis mossambicus were present in fresh and oligohaline but the latter species penetrated further into mesohaline but not in the polyhaline and euhaline reaches of the Orange River Estuary. Mozambique tilapia O. mossambicus has been recorded in salinities ranging from 0-100 (Whit eld 1998; Naesje et al. 2007), despite its highly euryhaline characteristics, it was absent in polyhaline and euhaline reaches of the Orange River Estuary. Oreochromis mossambicus may be avoiding estuarine reaches associated with an open mouth.
Estuarine & marine species were the second dominant species (i.e. Syngnathus temminckii, Caffrogobius nudiceps, Caffrogobius saldanha, Clinus superciliosus and Atherina breviceps) in freshwater and declined gradually toward polyhaline (10.0-29.9) and thereafter it increased into euhaline water. The presence of marine estuarine-dependent species M. cephalus, L. lithognathus and L. amia in fresh and oligohaline water are not surprising as they are known to penetrate the freshwater (especially if trapped) for one or two years and mature before moving out to sea to breed (Lamberth 2003;Naesje et al. 2007). The presence of piscivorous predators such as L. amia and P. saltatrix can indicate feeding usage of the ORE by the two species. Overall, the low number of marine sh species diversity in the Orange River Estuary may be related to the high dominance of riverine in uences throughout the estuary. However, C. richardsonii is dominant throughout the ORE.
This study concludes that the ideas presented in both the Remane (Remane 1934) and the revised model (Whit eld et al. 2012) do not accurately re ect the biodiversity situation for ORE, taking into account the plasticity observed in species that use the river-estuarine system. As a result, this study developed a new concept suitable for the River-Estuarine Continuum (REC) in the ORE. The following features are included in the proposed new REC: 1) Both marine and estuarine species are present in all salinity zones (i.e. fresh, oligohaline, mesohaline, polyhaline, and euhaline waters); 2) Freshwater species outnumber marine and estuarine species in terms of taxa; 3) The majority of freshwater species are con ned to freshwater, oligohaline, and mesohaline waters, with only a few taxa extending into polyhaline waters. No freshwater species were found in euhaline waters; 4) Estuarine species are more diverse in euhaline waters than in freshwater, oligohaline, mesohaline, and polyhaline waters; 5) The ORE is dominated by a single marine species (C. richardsonii).
6) There is a decrease in species diversity as one moves from low to high salinity waters.
In comparison to the original and revised models (Remane 1934; Whit eld et al. 2012), it is clear that sh diversity in the ORE is distinct, as more saline water lacked the substantial amount of marine taxa as hypothesized by pre-existing models. Nonetheless, sh assemblages that utilize the river-estuarine environment, particularly the dominant ones, use a wide salinity range (i.e. plasticity) along the ORE (i.e. C. richardsonii and G. aestuaria). Overall, the observed patterns were in uenced by the biogeographical trend of decreasing species richness on the South African coast from east to west (Lamberth 2003). Furthermore, differences in species tolerance and preference for changes in salinity along the ORE can dictate the presence of sh species along the river system.

Conclusion
The Orange River Estuary continues to be ecologically important for a variety of aquatic species, particularly marine and freshwater sh. The ndings of this study revealed that the River Continuum Concept's general applicability is insu cient for the ORE due to signi cant variation in sh diversity and composition along a salinity gradient. As a result of this research, a new river-estuarine concept with new features was proposed. The idea presented in the new concept, however, needs to be tested in other estuarine systems to determine its applicability and whether this trend is unique to the Orange River Estuary.
Declarations Figure 1 The Orange River Estuary is shown on the map from site 1 to 11, while the river section is shown from site 12 to 18.

Figure 2
Seasonal changes in estuarine functional grouping covering the salinity continuum from freshwater to euhaline conditions based on an adaptation of the Venice system for the classi cation of salinity zones (Strydom et al., 2003).