Paradigms in estuarine ecology – A review of the Remane diagram with a suggested revised model for estuaries

Estuarine, Coastal and Shelf Science 97 (2012) 78e90

Contents lists available at SciVerse ScienceDirect

Estuarine, Coastal and Shelf Science journal homepage: www.elsevier.com/locate/ecss

Paradigms in estuarine ecology e A review of the Remane diagram with a suggested revised model for estuaries A.K. Whitfield a, *, M. Elliott b, A. Basset c, S.J.M. Blaber d, R.J. West e a

South African Institute for Aquatic Biodiversity (SAIAB), Private Bag 1015, Grahamstown 6140, South Africa Institute of Estuarine and Coastal Studies (IECS), University of Hull, Hull HU6 7RX, UK c Department of Biological and Environmental Sciences and Technology, University of Salento, 73100 Lecce, Italy d CSIRO Marine and Atmospheric Research, GPO Box 2583, Brisbane, Queensland 4001, Australia e Australian Centre for Ocean Resources and Security (ANCORS), University of Wollongong, NSW 2522, Australia b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 27 July 2011 Accepted 15 November 2011 Available online 23 November 2011

Most estuarine ecology textbooks have included the so-called Remane diagram which is derived from German studies in the Baltic Sea region during the early part of the 20th Century. The model shows how aquatic species diversity changes from freshwater to more marine areas. In essence it aims to show the relative proportions of each component of the fauna (freshwater, brackish and marine) and how these change along a salinity gradient. These combined components decrease in diversity with a progression from both the freshwater and marine ends of the spectrum, with the 5e7 salinity area being dominated by a small number of true brackish/estuarine species. The way in which the Remane diagram has been interpreted (and misinterpreted) and used (and misused) in the literature is discussed here. We primarily investigate whether the model needs to be modified to help provide an understanding of current biotic distribution patterns within estuaries and how these patterns might be influenced by climate change. Using global estuarine examples for a variety of taxa we discuss the appropriateness of the Remane model beyond the zoobenthos (on which the model was originally based) and provide a revised model that is more suited to estuaries worldwide. Comment is also provided on the way in which a more appropriate estuarine biodiversity model can influence future estuarine ecotone and ecocline studies. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: estuary euryhaline stenohaline holohaline ecotones ecoclines

1. Introduction There have been recent attempts to define paradigms showing the fundamental properties of estuaries by attempting to both summarise and allow the further investigation of those properties (e.g. Elliott and Whitfield, 2011; Basset et al., in press). As one such paradigm, many estuarine ecology textbooks (e.g. Hedgpeth, 1967; Beadle, 1972; McLusky and Elliott, 2004) have included the socalled Remane diagram (Fig. 1), the basic model of which can be traced back to Remane (1934). The original diagram is a conceptual model designed to show species diversity distribution along a salinity continuum (in this case, for rivers entering Baltic waters) and displays the numbers of species with different salinity tolerances (freshwater, brackish and marine) which comprise the communities across that continuum. In this review we examine the basis for the creation and widespread acceptance of the Remane

* Corresponding author. E-mail address: a.whitfi[email protected] (A.K. Whitfield). 0272-7714/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.ecss.2011.11.026

diagram and assess whether it has application for estuaries worldwide. Based on more recent information from estuarine systems on different continents, we have constructed a more appropriate model for use in describing the relationship between salinity and biotic trends in estuaries. 2. Review of the Remane diagram According to the species diversity terminology used on the Yaxis of the Remane diagram by Remane (1934), the model suggests (although no scale is given) that the marine and freshwater components comprise an equal number of species. These components decrease with the progression into transitional waters and the space is then also occupied by a small but significant number of ‘brackish’ species, which peak at a salinity of about 6 (Fig. 1). It is important to note that salinities are here presented according to the practical salinity scale and thus no units are given (Lewis, 1980); to provide comparison with previous work it should be noted that up to a salinity of 42 the practical salinity units (psu) equate to parts per thousand (ppt or &), or g l
1.

A.K. Whitfield et al. / Estuarine, Coastal and Shelf Science 97 (2012) 78e90

Fig. 1. A redrawn version of the original Remane diagram (Remane, 1934). The slanted hashed area represents freshwater species, vertical hashed area corresponds to brackish species, and white area below the curve marine species. The vertical dashed line represents a salinity of approximately 50% seawater.

Adolf Remane’s figure legend does not confirm that we are dealing with absolute numbers for the marine, brackish and freshwater components, but the low maximum value for brackish species (Fig. 1) seems to suggest that actual numbers are being used in the diagram (rather than a scaling of species diversity for each component out of 100%). Greater clarity, however, is provided by the Remane and Schlieper (1958) version of the earlier diagram which shows the Y-axis scaled as a “per cent” (Fig. 2). This most likely refers to the percentage of freshwater and marine animals, relative to the total number of freshwater and marine animals, respectively. In this diagram, the inclusion of brackish animals is somewhat confusing, but most likely is a subset of the marine taxa (see below). The Remane model suggests that the diversity of freshwater taxa declines rapidly between a salinity of 0.5e5, with minimum

Fig. 2. A redrawn version of the Remane and Schlieper (1958) diagram, as depicted in McLusky and Elliott (2004), showing that the Y-axis has now been scaled according to percentage (not number).

79

species richness occurring at 5e7 (Fig. 1), a critical physicochemical range for aquatic organisms according to Telesh and Khlebovich (2010). At a salinity of 6, brackish species are the dominant component in the Baltic, but this ratio changes considerably above 7 when marine species become overwhelmingly dominant. Very few brackish species are recorded above a salinity of 17 (approximately 50% seawater) and none are recorded above 20. Remane (1934) classified all waters between 0.1 and 17 salinity as “brackish” and therefore his brackish group of species is effectively confined to this salinity range. Remane (1934) does not define what is meant by marine, brackish and freshwater species. Whilst marine and freshwater taxa obviously have a distribution that ranges from the marine and freshwater environments into brackish waters respectively, the brackish species quoted by Remane are presumed to be a subset of marine and freshwater taxa that have adapted to life in brackish waters. The implication from the Remane diagram is that all these groups of species are euryhaline to varying degrees and that this is the primary driver in terms of their occurrence along the salinity gradient. It is likely, however, that stenohaline taxa are also included in the Remane marine component since there is an immediate decrease in species diversity as salinities decline below 35 (Fig. 1). In addition, the marked decrease in freshwater taxa as salinities increase suggests a similar intolerance by certain freshwater species to saline waters. In essence the “brackish” sector of the Remane diagram is a reflection of the stress-subsidy continuum (Elliott and Quintino, 2007), showing the inability of many freshwater and marine species to tolerate low salinities, whereas true brackish species are highly euryhaline and capable of thriving under wide variations in salinity. This property is likely to become increasingly significant in relation to climate change and rises in sea level when estuaries and river systems may undergo major changes. Therefore understanding the vulnerability of freshwater and marine taxa in estuaries and the role and colonization potential of truly brackish species has never been more important. The Remane (1934) paper appears to be written almost verbatim (in an older form of colloquial German) from an oral presentation, and does not provide the kind of detail currently expected of scientific publications. There is also a lack of basic information, e.g. to what extent intertidal and/or sub-tidal sampling was used to compile species lists and one is left to assume that a similar sampling protocol was used in the different study areas. The model appears to be based primarily on benthic invertebrates although comment is made about similar patterns in other taxa, but no figures or tables are presented. The fact that the Remane diagram legend does not refer specifically to invertebrate diversity, and the author indicates that his paper is a general overview on the “problems and phenomena of brackish water biology”, indicates that the diagram was intended for general rather than specific use. Hence the ease with which it has been used as a paradigm to cover the biota in all estuaries. The Remane (1934) paper covered four broad regions, viz. the German North Sea, Belt Sea, southern Baltic up to northern Gotland, and the remaining northern areas of the Baltic. Remane only refers generally to the actual salinity in the different sampling areas, with surface salinities in the northern Baltic given as 4e6, southern and central Baltic 6e8, Kiel Bay 13e20 and North Sea 30e35. Based on the evidence presented it would appear that no samples were collected from estuaries sensu stricto, although Barnes (1974) considers the Baltic to be a single large stable ‘estuary’. Remane refers to the Baltic Sea as a large brackish water area that is connected to the ocean (the North Sea) and that this makes it particularly interesting because of the gradual decrease in salinity from the Kattegat and Belt Sea up to the Gulf of Bothnia

80

A.K. Whitfield et al. / Estuarine, Coastal and Shelf Science 97 (2012) 78e90

and Gulf of Finland. This is very different from the situation in estuaries, where freshwater and seawater are much closer together on the continuum, and where the gradient and fluctuations in salinity within an estuary are usually much greater than gulfs or seas (Elliott and Whitfield, 2011). It is even more different from the situation in many estuarine lagoons, where two distinct salinity gradients connect the lagoon with the neighbouring freshwater and marine ecosystems (Basset et al., in press). Hence it would appear that the Remane diagram is a transitional water model for open brackish seas, rather than one for estuaries. Consequently it is surprising that this model has been accepted, almost unconditionally, as being valid for the influence of salinity on estuarine biotic diversity. This does not imply that salinity is not a major driver in determining spatial and temporal species distribution patterns within estuaries; indeed, Elliott and Whitfield (2011) refer to the dominance of salinity as a classifying tool for estuarine biota. Remane (1934) did not compare his results with those in other parts of Europe or even other continents in the northern hemisphere. In addition, there is no attempted comparison with tropical or southern hemisphere studies, probably because no estuarine research had been conducted on this topic prior to the 1950s (Day, 1964). A number of fossil studies support the use of salinity as a primary driver of biotic diversity in transitional waters, but these papers only appeared much later (e.g. Hudson 1963, 1980). In contrast, we now have sufficient additional information to be able to test the appropriateness of the model for different taxa in a range of northern and southern hemisphere estuaries and other transitional waters. There are, of course, some important anomalies from other parts of the world that Remane (1934) would have been unaware of at the time, e.g. freshwater species disappear from the Remane diagram at salinities of >15 which is very conservative, especially as some secondary freshwater taxa from fish families such as the Gobiidae and insect families such as the Chironomidae can survive in full seawater. However, the most surprising anomaly of all is the Mozambique tilapia Oreochromis mossambicus which is a primary freshwater species that is capable of surviving in salinities as high as 116 in the hyperhaline Lake St Lucia (Wallace, 1975). Indeed, the only marine species which comes close to O. mossambicus in terms of salinity tolerance was the skipjack Elops machnata which was found in salinities of up to 112 in the same lake. It is perhaps significant that the fish species with the widest salinity tolerance (0e142) in hyperhaline portions of Laguna Madre (Texas) was also a freshwater species (sheepshead minnow Cyprinodon variegatus), with marine and estuarine fish species in the estuarine lagoon having a maximum salinity tolerance of 80 (Hedgpeth, 1967). The Remane (1934) diagram also shows marine species unable to tolerate waters with a salinity of <3. Once again there is an abundance of evidence which shows that many marine fish and invertebrate species are capable of surviving in salinities of 1e2 for prolonged periods and even in freshwater for more limited periods (Hanekom, 1989; Ter Morshuizen et al., 1996; Whitfield et al., 2006). Some of the elasmobranchs, a class not recognised for their euryhalinity, are capable of entering freshwater areas, e.g. the bull shark Carcharinus leucas is known to enter rivers and is sometimes found more than a 1000 km up systems such as the Zambezi and Limpopo (Skelton, 2001). In addition to the above example of marine species entering river systems, it is significant that diadromous species (e.g. anguillid eels) are omitted from the Remane diagram. This reduces the applicability of the model to global estuaries and is probably linked to the fact that fishes were not included in the original Baltic assessment (Remane, 1934).

3. Variations in the depiction of the Remane diagram Over the decades the original Remane diagram has been modified in the way it was originally depicted, with Remane and Schlieper (1958) changing the scale used on the Y-axis (Fig. 2), and Barnes (1974) altering the way the component graphics were presented (Fig. 3). This latter approach was reinforced by Attrill and Rundle (2002) who added ecotone and ecocline concepts to the model (Fig. 4). Hedgpeth (1967) extended the Remane diagram to include the entire salinity spectrum and also the addition of two new categories, holoeuryhaline and ultrahaline (Fig. 5). According to Hedgpeth, holoeuryhaline species, which are likely to be dominated by flagellates and other small organisms (Sylvestre et al., 2001), are species that range from freshwater to extreme hyperhaline conditions but may be absent from the sea. Ultrahaline species on the other hand are marine species that also tolerate hyperhaline conditions. A similar approach is outlined by Khlebovich (1969) who extended the Remane salinity continuum up to 100, indicating that the lowest number of species on the continuum is recorded in salinities above 70, with an absolute minimum diversity recorded between 80 and 100 (Fig. 6). Khlebovich did not use the terms holoeuryhaline or hyperhaline to describe species occupying hyperhaline waters. Instead, he followed the more conventional approach and divided the “transitional” aquatic fauna into three categories: freshwater species (freshwater origin), brackish water and euryhaline species (marine origin), and marine species (stenohaline forms) (Fig. 6). Hudson (1990) changed the shape of the original Remane diagram by showing the freshwater biota at a lower species diversity level than the marine biota, and also extending the presence of estuarine/brackish species beyond the 20 salinity limit (Fig. 7). In addition, the Remane model was improved further by Hudson (1990) inserting salinity categories on the X-axis and, as was the case with Khlebovich (1969), including proposed species diversity trends in the hyperhaline zone (Figs. 6 and 7).

Fig. 3. The Remane diagram redrawn from Barnes (1974). Note the modified graphical depiction of the component species groups when compared to the original Remane (1934) and Remane and Schlieper (1958) diagram.

A.K. Whitfield et al. / Estuarine, Coastal and Shelf Science 97 (2012) 78e90

Fig. 4. The ‘revised version’ of the Remane diagram (redrawn from Barnes, 1974) with the ecotone and ecocline concepts added by Attrill and Rundle (2002). Note that “brackish-water” species have now been changed to “estuarine” species.

4. Does the Remane diagram apply to estuaries? Now that we know that the diagram was not derived from estuarine data sensu stricto, and was also based almost exclusively on invertebrate sampling (mostly zoobenthos) in the Baltic region, the question arises as to why it has been used so frequently to explain salinity linked axial biodiversity trends in estuaries around the world. Indeed, we need to question whether the Remane diagram bears any resemblance to longitudinal changes in aquatic biodiversity within estuaries or should we be developing a more appropriate model from which to work. Less than a year after Remane published his now famous diagram, Alexander et al. (1935) indicated a similar, but not identical, trend in the Tay Estuary, Scotland (Fig. 8). This investigation, which included longitudinal diversity trends for both aquatic plants and animals, showed declining biotic diversity from the sea towards the head of the estuary (which would also have coincided with decreasing salinity). In contrast to the Remane diagram, however, there was no large increase in diversity at the freshwater end of the continuum (Fig. 8). There is still some confusion as to exactly what constitutes a “brackish-water fauna” and how this differs from the “estuarine resident” or “true estuarine” classifications preferred by some biologists. Since Remane (1934) never defined what he meant by “brackish species” it has devolved to others to describe what is meant by the term. Barnes (1999) summarizes the characteristics of brackish-water fauna as being “a suite of marine species of differing degrees of euryhalinity, that penetrate into dilute media to varying

Fig. 5. Extrapolation of Remane’s curve, with the inclusion of holoeuryhaline and ultrahaline species as depicted by Hedgpeth (1967).

81

Fig. 6. Extension of the salinity scale showing faunal changes according to salinity (redrawn from Khlebovich, 1969).

degrees, together with a few freshwater species that can withstand more saline waters”. Unfortunately this definition does not highlight any characteristics that separate “brackish-water fauna” from euryhaline marine and freshwater fauna that enter fluctuating salinity environments, e.g. in many estuaries, saline tolerant Tubificoides oligochaetes are typical of a “brackish-water” fauna (McLusky et al., 1993). Quintino et al. (in press), in sampling the whole continuum from the estuary mouth to freshwaters, indicated that the polychaete, Hediste diversicolor, the amphipod Corophium multisetosum and the isopod Lekanesphaera hookeri would be amongst the few regarded as brackish species in that they only occurred in the oligohaline area. In contrast, “estuarine-resident fauna” can be distinguished from their marine and freshwater counterparts by virtue of the fact that breeding by these species takes place within the estuarine environment and many of these taxa are confined to estuaries (Day, 1981). An additional reason for separating estuarine resident species from the brackish species used by Remane (1934) is that the former group often occur in salinities above 20 (which was the upper limit for the latter group in the Remane diagram). The absence of diadromous species (i.e. including anadromy, catadromy and amphidromy categories) in the Remane diagram is also significant and reinforces the view that this model does not apply to the whole community of estuarine systems. Indeed, in estuaries associated with certain geographic areas (e.g. New

Fig. 7. Revised version of the Remane diagram as applied to fossil biota (modified after Hudson, 1990).

82

A.K. Whitfield et al. / Estuarine, Coastal and Shelf Science 97 (2012) 78e90

uncommon in certain estuaries (Young and Potter, 2002) and other coastal waters (Lenanton, 1977), and the potential response of the biota to these conditions needs to be included in a modified form of the Remane model. Not only can certain species survive hyperhalinity, some may thrive and even breed within these waters (Bayley, 1972; Diouf et al., 2009). 5. Variations on the Remane model

Fig. 8. Composition of the biota (species number) along the Tay Estuary (redrawn after Alexander et al., 1935).

Zealand), fish species collectively grouped into the diadromous category can dominate these systems (McDowall, 1976). This group should therefore be included in any revision of the Remane diagram for estuaries. The Remane model does not seem to fit the macrobenthic invertebrate patterns recorded in Australian coastal lagoons (Powis and Robinson, 1980; Dye and Barros, 2005) or Portuguese rias (Quintino et al., in press). Moreover, in Mediterranean lagoon ecosystems the species richness of the zoobenthos is affected more by physiographical and hydromorpological parameters rather than salinity (Basset et al., 2006). In addition, the Remane model tends to portray a static picture of longitudinal diversity patterns, assuming similar patchiness at every level across the freshwater-marine ecotone. This is not the case in Mediterranean lagoons where patchiness is extremely high both within and among lagoons, with low a but high b and g diversities being recorded (Basset et al., in press). Indeed, in certain Italian lagoons, almost 90% of the macroinvertebrate biodiversity is due to among-lagoon patchiness (Barbone and Basset, 2010), with most of the within-lagoon biodiversity due to a patchiness of benthic habitats. Furthermore, Quintino et al. (in press) showed that the oligohaline species minimum for macrobenthos does occur when the whole continuum has been sampled but that the more mobile species (shown by the leaf-bag technique) shows the pattern better than does core sampling of the more sedentary forms. Using an analysis based mainly on fish data, but also including invertebrates, Bulger et al. (1993) suggest that the Remane species minimum at salinities of 5e8 needs to be challenged. This was based on the assumption that no freshwater forms were included in the Remane original analysis, and also that the salinity gradient in the Baltic includes a temperature gradient that would have influenced species distribution patterns. Again, only recent studies such as Quintino et al. (in press) and Rodrigues et al. (2011) have started producing data to test these patterns by assessing the complete continuum. Despite this, an initial assessment of their data suggests that the Remane model is only partially met. The absence of biotic diversity associated with hyperhaline waters in the original Remane diagram also needs to be questioned but is understandable when considering the absence of a hyperhaline zone in the Baltic. However, hyperhaline conditions are not

Bulger et al. (1993) showed that biological changes occur along the estuarine salinity gradient and attempted, as with the Venice classification, to superimpose categories on a continuum (using a Principle Components Analysis to give the continuum but over which boundaries were superimposed). They suggested that estuarine species are not evenly distributed across the salinity gradient and they identified a salinity of 24 rather than 30 as the lower limit of salinity penetration by marine species. They also suggested that a salinity of 4 was the tolerance limit of stenohaline (sic) freshwater fishes. Attrill (2002) provided a very powerful argument for salinity variation to be used as a primary criterion for determining species distribution and diversity within estuaries rather than absolute salinity tolerance. The alternative linear model developed by Attrill was based on mean salinity range at any point in the estuary versus mean diversity of macroinvertebrates and meiofauna from sub-tidal populations in the same region. In the Thames Estuary, Attrill and Rundle (2002) showed that the species diversity minimum for both the macrofauna and meiofauna occurred where mean salinities of approximately 10 were recorded, and that this region was also subjected to large fluctuations in salinity. This minimum is relatively close to the minimum in the Remane diagram (Fig. 1) but Attrill and Rundle (2002) argue strongly that in the case of the Thames Estuary it represents the lower ends of two ecoclines, one from the freshwater and the other from the marine side. They also contend that so-called brackish or true estuarine species do not exist (at least in the Thames Estuary) and therefore their linear model differs fundamentally from that of Remane (1934). Roy et al. (2001) have summarised many data for south eastern Australian estuaries and investigated the relationship between estuary geomorphology, water quality and species diversity and abundance. Although they adopt the general model of Remane and Schlieper (1958), they also emphasise the importance of estuary type (Roy, 1984) and estuary zonation (Rochford, 1951) in determining environmental conditions (including salinity) along the estuarine gradient, which in turn influence the species diversity and abundance. Roy et al. (2001) also highlight the large-scale role that flood events can have on species composition. These issues are explored further below. One of the consequences of fluctuating environmental conditions is a paucity of estuarine resident taxa. Bamber and Henderson (1988) have argued that the physical variability of estuarine environments has acted to select for generalist fish genotypes, able to adjust their morphology, physiology and behaviour to a wide range of conditions. The lack of diversity of true estuarine fish species (i.e. those that do not usually occur outside of estuaries) has also been raised by Whitfield (1994) who suggested that the low diversity of resident species is linked to the ephemeral nature of individual estuaries and often unpredictable nature of physico-chemical conditions within these systems, all of which does not promote speciation on a large scale. Within each of the phyla that occur within estuaries, however, there is usually a small proportion of species that do not occur in either the adjacent marine or freshwater environments and could therefore be defined as estuarine. Examples of estuarine species include the eelgrass, Zostera capensis, which is endemic to southern Africa and confined to

A.K. Whitfield et al. / Estuarine, Coastal and Shelf Science 97 (2012) 78e90

estuaries on the subcontinent (Edgcumbe, 1980). A similar situation occurs for Zostera capricorni in eastern Australia, which is, for the large part, restricted to estuaries (West et al., 1989). Certain South African invertebrate and fish taxa are also found almost exclusively in estuaries, e.g. the copepod Pseudodiaptomus hessei, amphipod Corophium triaenonyx, isopod Cyathuria estuaria, polychaete Dendronereis arborifera, bivalve Brachidontes virgiliae, cumacean Iphinoe truncata, brachyuran Hymenosoma orbiculare, tanaid Apseudes digitalis and clupeid Gilchristella aestuaria. Overall, Remane was correct in allocating only a relatively small proportion of the Remane curve to this group, but his restriction of the ‘brackish’ species to salinities between zero and 20 is incorrect, especially as estuarine taxa appear to be well adapted to a wide range of salinities, including hyperhaline waters (Whitfield et al., 2006; MacKay et al., 2010). A slightly different perspective on the existence of “brackish” water species in estuaries is provided by Cognetti and Maltagliati (2000). They suggest that many of the so-called euryhaline marine species in “brackish” (estuarine) waters are, in fact, differentiated sibling forms of marine species in the adjacent ocean. They contend that overlooking these brackish forms has arisen from the traditional “macro-systematic” level of classification and that fine morpho-physiological and genetic analyses reveal a different picture, with sub-populations of euryhaline marine species showing differentiation, through selection, in order to thrive in unpredictable “brackish” waters. 6. Estuarine biotic groups and salinity 6.1. Plankton Telesh et al. (2011) have recently shown that, for both phytoand zooplankton in Baltic waters, the Remane diagram is inappropriate. In the case of protists and metazooplankton the species richness reaches a maximum in the horohalinicum (5e8 salinity zone) whereas the Remane model suggests a minimum diversity in this region (Fig. 1). These authors argue that planktonic organisms thrive in stable brackish environments such as the Baltic and that their advanced osmoregulation strategies give them a distinct advantage in adapting to low salinity conditions. The situation in widely fluctuating salinity environments such as estuaries appears to be somewhat different from that described for the Baltic Sea above, and Gómez et al. (2004) recorded a dominance of the upper and middle reaches of the Río de la Plata Estuary by riverine phytoplankton, declining in diversity towards the mixohaline zone. A more recent study on the phytoplankton in the Río de la Plata region (Calliari et al., 2009) showed an increase in the number of representative taxa from the freshwater and upper estuary stations towards the more saline coastal shelf stations. Zooplankton diversity in estuaries seems to follow similar trends to that described for phytoplankton. Duggan et al. (2008) documented a significant decrease in zooplankton species richness from the mouth to the river sites of a northern Australian estuary. Species diversity of zooplankton in South African estuaries often reaches a peak in the lower reaches (Grindley, 1981), primarily due to the contribution of both estuarine and marine species to the overall population within this region (Wooldridge, 1999). Therefore species diversity is higher at the mouth when compared to the head of an estuary (Wooldridge and Deyzel, 2009). Consequently the zooplankton species decline from the marine towards the oligohaline upper reaches of certain estuaries (Fig. 9) follows that predicted by the Remane model but may be a consequence of proximity to marine species richness rather than salinity per se. There is also no evidence of high zooplankton species diversity in freshwater entering estuaries, as predicted by the Remane diagram.

83

Fig. 9. Diversity of zooplankton from samples collected between the head and mouth of the Swartkops Estuary (redrawn after Grindley, 1981).

Taylor (1993) also found a gradient in the occurrence of zooplankton species along the length of the Forth Estuary (Scotland). This gradient could be linked to salinity and other hydrographic features, with both estuarine and incursive marine species present. The abundance of many of the taxa was not only related to salinity but also water temperature (as shown by the Bary diagrams that he presents thus illustrating the synergistic effects of environmental variables), with some species increasing in abundance with increasing water temperature and others decreasing in abundance with increasing water temperature. His data also illustrates the forced migration of community components depending on the freshwater inflows (and thus salinity), whereby the dominant freshwater-tolerant community (dominated by the calanoid copepod Eurytemora affinis) which occurs only at the head of the estuary under summer, lowflow conditions, then is migrated/ pushed downstream into the middle reaches following high autumnal and spring rainfall and thus lowered salinity. This emphasises the temporal displacement of the Remane curves depending on the prevailing hydrographic patterns. 6.2. Macrophytes and algae Remane (1934) made reference to the greater diversity of aquatic plants in seawater when compared to low salinity areas, as well as the exclusion of most freshwater macrophyte species from brackish areas of the Baltic. In South Africa, marine and estuarine plants tend to dominate the more saline lower reaches of permanently open estuaries (Adams et al., 1992), declining with a progression up the estuary towards the head where macrophytes of freshwater origin are more diverse (Fig. 10). Based upon the recorded salinity ranges of freshwater, estuarine and marine aquatic macrophytes in South African estuaries it appears that plant species diversity is highest in salinities between 25 and 30, declining rapidly once they increase above 35 (Fig. 10). Similarly, the diversity of algae also declines with increasing salinity. Sapozhnikov et al. (2009) describe how the diversity of benthic algae in the Aral Sea declined from 159 species in 2000 to 38 species in 2005 as salinities increased from oligohaline to hyperhaline. In South Africa the highest aquatic macrophyte diversity is recorded in coastal freshwater lakes and lagoons that have limited or no contact with saline waters (Howard-Williams, 1980). This is similar to the pattern for aquatic angiosperms (i.e. submerged and emergent flowering plants) in Australia that are more species rich in freshwater when compared to the number of seagrass species in marine waters (Sainty and Jacobs, 1994). A similar pattern of macrophyte zonation according to salinity was found in the Damietta Estuary (Egypt) where seven vegetation

84

A.K. Whitfield et al. / Estuarine, Coastal and Shelf Science 97 (2012) 78e90

Fig. 10. Cumulative coastal aquatic macrophyte diversity, comprising species of freshwater and marine/estuarine origin according to recorded salinity ranges for each taxon (data extracted from Ward, 1976; Howard-Williams, 1980; Adams et al., 1999).

types were identified (Khedr, 1998). Ordination of the vegetation types identified a strong longitudinal gradient from the mouth to 20 km upstream, with the lower saltmarsh groups being separated out from the freshwater marsh groups. Even at the species level, the role of salinity in affecting macrophyte distribution in estuaries has been clearly shown, with Ribeiro et al. (2011) identifying this parameter as a primary driver of Crinum americanum occurrence in the Massaguaçu Estuary (Brazil). Red and brown macroalgae (i.e. seaweeds) are generally confined to the lower reaches of permanently open South African estuaries that have a strong tidal prism and where seawater salinity prevails, whereas green algae (especially filamentous forms) are much more widespread along the length of estuaries (Adams et al., 1999). This is similar to the pattern in the Thames Estuary (UK) where more than 20 algal species are present in salinities above 30, between 10 and 15 species in salinities ranging from 10 to 30, and less than 5 species in salinities <5 (Kaiser et al., 2005). A review by Wilkinson et al. (1995) of algal colonization of temperate European and North American estuaries indicates that red and brown algae have differing proportions of species in estuaries that may be linked to the effect of temperature on salinity tolerance. Wilkinson (1980) also noted a reduction in intertidal algae (mainly marine) as a result of selective attenuation in an upstream direction, firstly red and then brown algae, with green algae being present throughout most estuaries. Wilkinson (op. cit.) also noted that very few macroalgal species occur in the middle or upper reaches of estuaries and that this appears to be a global trend. More recently, Wilkinson et al. (2007) have shown that the reliability of environmental quality indices linked to macroalgae depends on the responses by the species to salinity and other environmental factors. 6.3. Zoobenthos Indications are that the Remane model closely reflects the general pattern of macrobenthic invertebrate diversity from the

lower reaches (euhaline) to the head (oligohaline) of typical estuaries both in South Africa (Schlacher and Wooldridge, 1996, Fig. 11) and in Europe (Quintino et al., in press; Rodrigues et al., 2011). Unfortunately equivalent invertebrate species diversity information from the freshwater region above these estuaries is not always available and therefore we cannot ascertain whether there is a secondary increase in diversity in the inflowing river (as predicted in the Remane diagram). This break in sampling continuity is a major problem in attempting to assess the complete zoobenthic continuum from freshwater through an estuary to the adjacent marine area. Further support for the Remane model in terms of European benthic invertebrate diversity trends is provided by Elliott and Kingston (1987) and Mouny et al. (1998) who found declining species richness from the polyhaline to oligohaline zones of the Forth and Seine estuaries respectively. For example, the macrobenthos distribution given by Elliott and Kingston (1987) (Fig. 12) shows the classical minimum in the upper reaches, followed by an exponential increase over the main salinity transition zone, before decreasing slightly to a background marine diversity nearer the sea. The elevated species richness at the seaward end of the estuary before a slight decrease to the background marine levels was interpreted as showing the position of the ecotone, defined as the merging of two separate communities (the estuarine and marine ones). The error bars indicate, however, the inherent variability at any particular region of the estuary, probably reflecting the influence of habitat heterogeneity on macrobenthos (Fig. 12). Similarly, Rosa-Filho et al. (2004), by combining macrobenthic data from Brazilian estuaries were able to use multivariate models to describe and then predict the change of species along estuaries using salinity but required other habitat descriptors (such as vegetation type and depth) to explain more of the variability in the data. Jones et al. (1986) and Jones (1987) investigated the spatial and temporal patterns in the macrobenthic communities in the Hawkesbury River system near Sydney (Australia) in relation to salinity

Fig. 11. Average freshwater, estuarine and marine macrobenthic invertebrate diversity in the headwaters (salinity 1e11), upper reaches (salinity 8e28) and lower reaches (salinity 17e35) of four permanently open South African estuaries (data from Day, 1981).

A.K. Whitfield et al. / Estuarine, Coastal and Shelf Science 97 (2012) 78e90

Fig. 12. Macrobenthic species richness (mean  SE for each location) along the Forth Estuary and Firth of Forth showing overall declining diversity with increasing proximity to freshwater supply (modified from Elliott and Kingston, 1987).

and sediment types. They found that, although a consistent decline in species richness along the system never occurred, the more saline locations usually did support the majority of species and the upstream locations had the least species (i.e. conforming to the Remane model). The authors also noted that major patterns in the macrobenthos were often related to salinity (Jones et al., 1986) and that flood and drought conditions impacted species diversity and community structure in the higher saline areas (w20e30) more so than in the low salinity sites (w0e2) (Jones, 1987). Teske and Wooldridge (2003, 2004) determined that both sediment type and salinity influence macrobenthic communities in a range of South African estuarine types. They found that whilst the distribution of estuarine endemic fauna was driven mainly by sediment type, that of the marine and oligohaline fauna was driven by salinity. This trend was so strong that marine zoobenthos was numerically common only in permanently open estuaries with a strong tidal presence, and oligohaline species were only abundant in the upper reaches of large river dominated open estuaries or low salinity temporarily closed estuaries. The dominant role of sediment and salinity in driving the distribution of invertebrates in aquatic ecosystems is also provided by Nanami et al. (2005). They examined the zoobenthic community structure and spatial distribution patterns of individual species in the brackish Lake Hinuma (Japan). Results from cluster and canonical correspondence analyses revealed that median sediment grain size and salinity have significant effects on the distribution of 11 of the 14 dominant species recorded during the study. The Río de la Plata Estuary (South America) had the highest zoobenthos species diversity in the region closest to the open ocean, declining towards the mixohaline zone further up the system (Cortelezzi et al., 2007). An indication that salinity is a primary driver of macrobenthic invertebrate distribution in estuaries is also provided by de Villiers et al. (1999) who depict little change in stenohaline, euryhaline marine and estuarine species diversity from the mouth to the head of the Kariega Estuary (South Africa), a system renown for being euhaline along its entire length (Allanson and Read, 1995). Again, Quintino et al. (in press) detected a decrease in macrobenthic species richness with distance upstream using core sampling but a decrease and then increase of the mobile species using an experimental leaf-bag sampling technique. 6.4. Fishes As with the macroinvertebrates in the Río de la Plata Estuary (see above), its fish assemblages are separated according to salinity (Jaureguizar et al., 2003). According to these authors, three fish assemblages could be identified along the main axis of the estuary,

85

the boundaries of which coincided with the highest salinity gradients. These ‘boundaries’ were not static, and expanded or contracted between seasons, driven primarily by salinity. Jaureguizar et al. (op. cit.) suggested that, for large estuaries, fish assemblages would be good indicators of the spatial scale of habitat heterogeneity, as defined by salinity and water column stratification. Although the Remane model predicts low fish species diversity in salinities of 5e10, this is not correct. For fishes in Indo-West Pacific estuaries, particularly in the sub-tropics and tropics, these waters can support diverse fish assemblages, e.g. in the subtropical Kosi Estuary in South Africa (Blaber and Cyrus, 1981), the tropical Embley Estuary in Australia (Blaber et al., 1989), and the tropical Lupar Estuary in Borneo (Blaber, 2000). The situation is similar in temperate eastern Australian coastal rivers and estuaries (Gray et al., 1990), e.g. West and Walford (2000) sampled fish every 2e5 km from freshwater to marine dominated locations in both the Clarence and Richmond systems in northern New South Wales. They found that fish species numbers decreased from approximately 35e40 at the entrance sites to about 15e20 at the most upstream locations. However, even sites which had mean salinities of 0e10 supported diverse fish assemblages. It was also noted that the abundances of various fish species changed between what were categorised as marine, brackish and freshwater sites, and most species tolerated a wide range of salinity conditions. In a parallel study of shallow vegetated and bare sites along the Clarence system, West and King (1996) noted a trend of decreasing species diversity between marine, brackish and freshwater locations, exacerbated by recruitment of marine species during the warmer months. Again the brackish sites supported very diverse fish communities. It is only when the salinity approaches zero and river flow is high that fish diversity declines markedly (Whitfield and Harrison, 2003). In those estuaries where there is little or no longitudinal salinity gradient, the greatest species diversity is still recorded in the lower reaches (Ter Morshuizen and Whitfield, 1994), suggesting that food resources, sediment types and geographical position within an estuary may also be important in determining species distribution patterns within these systems (Richardson et al., 2006). It is also important to note that estuaries with little or no river inflow often have a higher number of recorded species but reduced fish abundance when compared to those systems with a more conventional salinity gradient (Whitfield et al., 1994). The higher species diversity is associated with an increased presence of stenohaline marine taxa entering the euhaline estuary. Support for the above fish species responses to river flow/ salinity changes is provided by the findings of Sheaves et al. (2007) in the Ross Estuary, Australia. They recorded 88 fish species in the estuary during a dry climatic period and 69 species during a wet period. In addition, during the dry period ichthyofaunal composition was not related to the position of sites whereas during the wet period there were clear upstream gradients, with upstream sites heavily influenced by freshwater species and downstream sites being dominated by marine and euryhaline species. Although for many studies the data showing the species progression into the freshwater areas above estuaries are absent, overall there is a low freshwater fish diversity in rivers above the head of South African estuaries (Skelton, 2001), especially when compared to the diversity of marine species in the sea adjacent to those same estuaries (Heemstra and Heemstra, 2004). Consequently the shape of a typical South African fish diversity diagram (Fig. 13) indicates that marine species are overwhelmingly dominant in mesohaline, polyhaline and euhaline water whereas estuarine and freshwater species combined are more diverse in oligohaline waters (Fig. 13). As expected, the catadromous species

86

A.K. Whitfield et al. / Estuarine, Coastal and Shelf Science 97 (2012) 78e90

Fig. 13. Species diversity changes of freshwater, catadromous, estuarine and marine fish species in the St Lucia system according to recorded salinity ranges (data from Skelton et al., 1989; Whitfield et al., 2006).

are thinly and evenly spread over the salinity range from freshwater to seawater. The overall greater diversity of marine taxa in higher salinity waters follows the trend depicted in the Remane diagram but this diversity declines markedly when salinities rise above 50 (Fig. 13), with the hyperhaline category not covered by the Remane model. The freshwater fish fauna of tropical Australia is also relatively depauperate and hence their contribution to the estuary-related fauna is correspondingly low (Blaber, 2000). Similarly, due to the relative isolation of the Philippine Islands, only a few freshwater species occur in this region and their contribution to the estuarine ichthyofauna is also low (Fig. 14). According to Pauly (1982) this

Fig. 14. The relationship between salinity and fish species diversity in the Philippines (after Pauly, 1982).

depauperate freshwater fish fauna has allowed a number of marine fish species to become secondarily adapted to freshwater, to the extent that “holoeuryhaline” marine species have altered the shape of the Remane diagram as applied to estuarine systems associated with these islands (Fig. 14). In contrast to the depleted freshwater fish fauna described above, in high rainfall mainland tropical areas there is often a diverse freshwater fish assemblage present (e.g. parts of West Africa and South America) and their contribution to the overall estuarine fish community may be much higher (Whitfield, 2005). For example in the Fatala Estuary in Guinea freshwater fishes make up 24% of the ichthyofauna (Baran, 1995) and in the Lupar Estuary in Sarawak 14% of the fishes are freshwater species (Blaber, 2000). Under these conditions, the Remane diagram for the Baltic is likely to be very similar to that recorded in these systems, i.e. high freshwater species diversity in the rivers entering the estuary and extensive oligohaline areas within each estuary due to elevated river run-off. Barletta and Blaber (2007) concluded that the significant difference between many Indo-West Pacific estuaries and those of the tropical western Atlantic is a result of the much higher diversity of freshwater species in tropical South America, e.g. the Caeté Estuary of northeastern Brazil is dominated, especially in its upper and middle reaches (salinities <10), by freshwater species (Barletta et al., 2005; Barletta and Blaber, 2007). Truly estuarine fishes also appear to play a much greater role in the tropics, comprising 31% of the diversity in the Fatala Estuary (Baran, 1995), 30% in the Lupar Estuary (Blaber, 2000) and 50% in the Caeté Estuary (Barletta and Blaber, 2007). This may be a function of many large piscivorous marine fish species being excluded from the freshwater-dominated upper regions of these systems. There are, however, a number of euryhaline marine fish species that do penetrate low salinity, or even freshwaters, and this is likely to be related to the osmoregulatory abilities of each individual species (Whitfield et al., 2006). It would appear that, particularly in the sub-tropics and tropics, many euryhaline marine species can tolerate very low salinities (<4) (Albaret et al., 2004) and that these species can even dominate the headwater region of estuaries with strong riverine inputs (Ter Morshuizen et al., 1997). As indicated above for the macrobenthos and zooplankton, the Remane model for fishes is further complicated by reflecting the distribution of species which are the result of several environmental factors rather than just salinity. For example, salinity tolerance by fishes is strongly related to the interaction between temperature and salinity, with the osmoregulatory abilities of even euryhaline species being compromised at extreme low and high temperatures (Blaber, 1973; Whitfield and Blaber, 1976). Nevertheless, there is interesting evidence that some estuarine resident (truly brackish) species can adapt to freshwater over periods of time and this may be significant in relation to future predicted changes in sea levels. For example, the completely landlocked freshwater Lake Sibaya in south-east Africa has a relict fauna of several species of fishes (e.g. Gilchristella aestuaria and Croilia mossambica) that are usually found only in estuaries. It has been postulated that the isolation of Lake Sibaya since the Pleistocene allowed gradual osmoregulatory adaptations to occur (Allanson et al., 1962; Blaber, 2000). Another aspect not covered by the Remane model is how biomass changes with salinity. Kolpakov and Milovankin (2010) established that fish biomass in the Razdol’naya Estuary increases with a rise in salinity. Furthermore, they determined that freshwater stenohaline (sic) species dominated in the upper part of the estuary, freshwater euryhaline and semi-anadromous species in the middle part, and semi-anadromous and marine species in the lower estuary.

A.K. Whitfield et al. / Estuarine, Coastal and Shelf Science 97 (2012) 78e90

87

7. A revised model for estuaries? The foregoing discussion emphasises that the Remane diagram was based primarily on benthic invertebrate diversity in a brackish sea and its river mouths, not an estuary. Plant and fish species diversity were not included in the Remane diagram yet it has been consistently portrayed in books as though it applies to all biotic groups and therefore become a central paradigm in estuarine ecology. Based on the above we conclude that we need to create a revised Remane diagram that more accurately reflects the biodiversity situation for estuaries than the current model. Alexander et al. (1935) provide an early example of the numbers of species for marine, estuarine and freshwater plants and animals from the Tay Estuary, Scotland (Fig. 8). They indicated that marine species in the lower reaches reached hundreds whereas freshwater species towards the head of the estuary are into the tens, a trend that is repeated in a number of subsequent investigations in other parts of the world where conventional estuarine salinity gradients prevail. In assessing the distributions of species, Bulger et al. (1993) used the salinity ranges of species as indicated in the literature for different species and their life stages. However, this is not the same as using actual salinity tolerances and it is notable that there are few wide-scale studies in which salinity tolerances have been determined for estuarine species occurring in estuaries (e.g. Blaber 1973). South African invertebrate and fish species trends (Figs. 11 and 13), together with information outlined in this review, indicate that overall diversity declines from the euhaline, polyhaline and mesohaline parts of an estuary towards the oligohaline headwater region. However, even in this latter region, marine taxa are likely to be dominant in terms of species diversity provided salinities remain above 3. At the opposite salinity extreme, most mobile species move or perish in saline waters above 50. For example, Grindley and Heydorn (1970) describe how the demise of a wide range of fish, crustaceans, molluscs and other invertebrates in salinities between 49 and 89 at Lake St Lucia (South Africa) laid the foundation for a bloom of the planktonic flagellate Noctiluca scintillans and the proliferation of chironomid midges. Grindley and Heydorn (op. cit.) postulated that the organic compounds from the fauna killed by the high salinities promoted the development of bacteria on which Noctiluca feeds, thus leading to the dinoflagellate ‘red tide’. Similarly the lack of fish and other predators on larval chironomids, due primarily to hyperhaline conditions in the lake, favoured the proliferation of the midges to pest proportions. There are considerable parallels between biotic trends in hyperhaline St Lucia and the situation recorded in the Casamance Estuary (Senegal). Due to a prolonged drought, the lack of river flow resulted in the upper reaches of the Casamance Estuary attaining a salinity of 170 some 210 km from the sea. As expected, biotic components such as mangroves, foraminiferans, zooplankton and fishes showed a considerable reduction in species number with increasing salinity. As was the case at St Lucia, populations of the different groups became almost monospecific due to the loss of species to high salinity (Debenay et al., 1989). Hedgpeth (1967), in his study of the fauna of Laguna Madre (Texas), depicts a considerable decline in biotic diversity with increasing salinity (particularly above 60), with only a handful of species capable of surviving in salinities above 80. Similarly, Vega-Cendejas and de Santillana (2004) describe how fish species richness and density decline from the mouth of the Rio Lagartos Lagoon (Mexico) to the inner zone where salinities >80 prevailed and where competitive interactions decreased. In contrast to the above situations, the diversion of freshwater from the Farakka Barrage on the Ganaga River (India) into the

Fig. 15. Proposed conceptual model for estuarine biodiversity (species) changes covering the salinity continuum from freshwater to hyperhaline conditions.

Hooghly Estuary resulted in a considerable decline in salinity throughout the estuary, with the estuarine zone being moved seawards and the marine zone being confined to the mouth area (Sinha et al., 1996). This resulted in major changes in plankton dynamics and a sharp decline in the marine fishery but increase in the freshwater fishery in the upper estuary. From the St Lucia, Casamance and Hooghly examples it is apparent that salinity may affect individual species, taxonomic groups and guilds to such an extent that it can completely restructure the trophic dynamics within an ecosystem and is therefore of major interest to management. Using information presented in this review, it is possible to suggest a conceptual species diversity change model for estuaries, as opposed to the Remane diagram which we emphasise applies more to brackish seas. The revised version of the diagram (Fig. 15) has the following set of features: (1) The freshwater assemblage in a particular estuary is generally not as species rich as marine taxa in the same estuary; (2) The majority of freshwater species do not penetrate saline waters, i.e. they are mainly confined to the freshwater area above the estuary, with only a few taxa extending into mesohaline, polyhaline and euhaline areas; (3) Marine species extend into oligohaline estuarine and sometimes even freshwaters but in relatively small numbers when compared to their diversity in more saline waters; (4) Marine species overwhelmingly dominate (in terms of taxa) the mesohaline, polyhaline, euhaline and hyperhaline estuarine waters; (5) Estuarine/brackish species are most diverse in mesohaline and polyhaline waters but can also be present in limnetic, oligohaline, euhaline and even hyperhaline waters; (6) Biotic diversity in estuaries starts to decline above a salinity of about 40, with most species unable to survive or moving out of areas where salinities are above 50; (7) There are very few freshwater, estuarine or marine species that can be termed holohaline, i.e. those species that are capable of occupying waters with a salinity range of 0e100þ. 8. Conclusions Based on patterns worldwide and the examples quoted above, we conclude that the Remane model has been both misinterpreted and indiscriminately applied to global estuaries and to all biotic components. There is general acceptance that the Remane diagram cannot be applied quantitatively to any transitional water, including estuaries (Attrill, 2002). Given that the model is based on

88

A.K. Whitfield et al. / Estuarine, Coastal and Shelf Science 97 (2012) 78e90

findings from the Baltic region where tidal regimes are very different to European estuaries, and where river mouths enter a brackish sea rather than a marine ocean, it is a surprise that even the qualitative aspects of the Remane diagram have gained such widespread acceptance in estuarine textbooks. There is, however, merit in adapting the Remane diagram to better fit the estuarine situation since it can incorporate two layers of information that are most useful in providing a broad description of biotic compositions in estuaries. The first layer pertains to the guilds (e.g. marine, estuarine, freshwater) that occupy estuaries and the second layer relates to salinity tolerances or preferences of the species found in estuaries and adjacent ecosystems. Thus freshwater species would be those taxa confined to the limnetic and perhaps oligohaline portions of the estuary, euryhaline freshwater, estuarine and marine species would be most abundant in the oligohaline, mesohaline and polyhaline reaches of the system, stenohaline marine species in the euhaline reaches, and holohaline species over the complete range of salinity conditions, from limnetic to extreme hyperhaline (Fig. 15). Although the estuarine biodiversity distribution diagram proposed in this review is only a qualitative assessment of longitudinal species diversity changes within a typical estuarine ecosystem, it does help to explain the dominant longitudinal species diversity patterns that appear to have salinity as one of the primary drivers of fauna and flora along the estuarine gradient. Once the existence of these biodiversity trends have been established for individual estuaries, it then becomes possible to test ecological theories pertaining to the various drivers (Barnes, 1999; Elliott and Whitfield, 2011; Basset et al., in press), as well as supporting the development of realistic estuarine quality indicators that are based on fundamental attributes of these ecosystems. Acknowledgements We are indebted to Dr Helga Barthels-Hardege for her translation and help with the interpretation of Remane (1934), Mrs Susan Abraham for drawing the figures used in this review and the National Research Foundation (NRF) for funding this work. We also thank the anonymous reviewers for their comments. References Adams, J.B., Bate, G.C., O’Callaghan, M., 1999. Primary producers. In: Allanson, B.R., Baird, D. (Eds.), Estuaries of South Africa. Cambridge University Press, Cambridge, pp. 91e117. Adams, J.B., Knoop, W.T., Bate, G.C., 1992. The distribution of estuarine macrophytes in relation to freshwater. Botanica Marina 35, 215e226. Albaret, J.-J., Simier, M., Darboe, F.S., Ecoutin, J.-M., Raffray, J., de Morais, L.T., 2004. Fish diversity and distribution in the Gambia Estuary, West Africa, in relation to environmental variables. Aquatic Living Resources 17, 35e46. Alexander, W.B., Southgate, B.A., Bassindale, R., 1935. Survey of the river tees, part II e the estuary, chemical and biological. DSIR, Water Research Technical Paper 5, 1e171. Allanson, B.R., Read, G.H.L., 1995. Further comment on the response of Eastern Cape Province estuaries to variable freshwater flows. Southern African Journal of Aquatic Sciences 21, 56e70. Allanson, B.R., Hill, B.J., Boltt, R.E., Schultz, V., 1962. An estuarine fauna in a freshwater lake in South Africa. Nature 209, 532e533. Attrill, M.J., 2002. A testable linear model for diversity trends in estuaries. Journal of Animal Ecology 71 (2), 262e269. Attrill, M.J., Rundle, S.D., 2002. Ecotone or ecocline: ecological boundaries in estuaries. Estuarine, Coastal and Shelf Science 55 (6), 929e936. Bamber, R.N., Henderson, P.A., 1988. Preadaptive plasticity in atherinids and the estuarine seat of teleost evolution. Journal of Fish Biology 33 (Suppl. 1), 17e23. Baran, E., 1995. Dynamique spatio-temporelle des peuplements de poissons estuariens en Guinée e relations avec les milieu abiotique. PhD thesis, Université de Bretagne Occidentale, France. Barbone, E., Basset, A., 2010. Hydrological constraints to macrobenthic fauna biodiversity in transitional waters ecosystems. Rendiconti Lincei 21 (4), 301e314.

Barletta, M., Blaber, S.J.M., 2007. A comparison of fish assemblages and estuarine functional guilds in different habitats of tropical estuaries from the Indo-West Pacific and West Atlantic. Bulletin of Marine Science 80, 647e680. Barletta, M., Barletta-Bergan, A., Saint-Paul, U., Hubold, G., 2005. The role of salinity in structuring the fish assemblages in a tropical estuary. Journal of Fish Biology 66, 45e72. Barnes, R.S.K., 1974. Estuarine Biology. Edward Arnold, London. Barnes, R.S.K., 1999. The conservation of brackish-water systems: priorities for the 21st century. Aquatic Conservation: Marine and Freshwater Ecosystems 9, 523e527. Basset, A., Sabetta, L., Fonnesu, A., Mouillot, D., Do Chi, T., Viaroli, P., Giordani, G., Reizopoulou, S., Abbiati, M., Carrada, G.C., 2006. Typology in Mediterranean transitional waters: new challenges and perspective. Aquatic Conservation: Marine and Freshwater Ecosystems 16 (5), 441e455. Basset, A., Barbone, E., Elliott, M., Li, B.-L., Jorgensen, S.E., Lucena-Moya, P., Pardo, I., Mouillot, D., A unifying approach to understanding transitional waters: fundamental properties emerging from ecotone ecosystems. Estuarine, Coastal and Shelf Science, in press. Bayley, I., 1972. Salinity tolerances and osmotic behaviour of animals in athalassic saline and marine hypersaline waters. Annual Review of Ecology and Systematics 3, 233e268. Beadle, L.C., 1972. Physiological problems for animal life in estuaries. In: Barnes, R.S.K., Green, I. (Eds.), The Estuarine Environment. Applied Science Publishers, London, pp. 51e60. Blaber, S.J.M., 1973. Temperature and salinity tolerance of juvenile Rhabdosargus holubi (Steindachner) (Teleostei: Sparidae). Journal of Fish Biology 5, 593e599. Blaber, S.J.M., 2000. Tropical Estuarine Fishes. Ecology, Exploitation and Conservation. In: Fish and Aquatic Resources Series 7, Blackwell Science, Oxford: 372 pp. Blaber, S.J.M., Brewer, D.T., Salini, J.P., 1989. Species composition and biomasses of fishes in different habitats of a tropical northern Australian estuary: their occurrence in the adjoining sea and estuarine dependence. Estuarine, Coastal and Shelf Science 29, 509e531. Blaber, S.J.M., Cyrus, D.P., 1981. A revised checklist and further notes on the fishes of the Kosi system. In: Lammergeyer No. 31, pp. 5e14. Bulger, A.J., Hayden, B.P., Monaco, M.E., Nelson, D.M., McCormick-Ray, M.G., 1993. Biologically-based estuarine salinity zones derived from a multivariate analysis. Estuaries 16 (2), 311e322. Calliari, D., Brugnoli, E., Ferrari, G., Vizziano, D., 2009. Phytoplankton distribution and production along a wide environmental gradient in the south-west Atlantic off Uruguay. Hydrobiologia 620, 47e61. Cognetti, G., Maltagliati, F., 2000. Biodiversity and adaptive mechanisms in brackish water fauna. Marine Pollution Bulletin 40 (1), 7e14. Cortelezzi, A., Capítulo, A.R., Boccardi, L., Arocena, R., 2007. Benthic assemblages of an estuarine system in South America: transition from a freshwater to an estuarine zone. Journal of Marine Systems 68, 569e580. Day, J.H., 1964. The origin and distribution of estuarine animals in South Africa. In: Davis, D.H.S. (Ed.), Ecological Studies in Southern Africa. Monographia Biologicae 24, pp. 159e173. Day, J.H., 1981. The estuarine fauna. In: Day, J.H. (Ed.), Estuarine Ecology with Particular Reference to Southern Africa. A.A. Balkema, Cape Town, pp. 147e178. Debenay, J.P., Pages, J., Diouf, P.S., 1989. Ecological zonation of the hyperhaline estuary of the Casamance River (Senegal): Foramanifera, zooplankton and abiotic variables. Hydrobiologia 174, 161e176. de Villiers, C.J., Hodgson, A.N., Forbes, A.T., 1999. Studies on estuarine macroinvertebrates. In: Allanson, B.R., Baird, D. (Eds.), Estuaries of South Africa. Cambridge University Press, Cambridge, pp. 167e207. Diouf, K., Guilhaumon, F., Aliaume, C., Ndiaye, P., Chi, T.D., Panfili, J., 2009. Effects of the environment on fish juvenile growth in West African stressful estuaries. Estuarine, Coastal and Shelf Science 83 (2), 115e125. Duggan, S., McKinnon, A.D., Carleton, J.H., 2008. Zooplankton in an Australian tropical estuary. Estuaries and Coasts 31 (2), 455e467. Dye, A.H., Barros, F., 2005. Spatial patterns in meiobenthic assemblages in intermittently open/closed coastal lakes in New South Wales, Australia. Estuarine and Coastal Shelf Science 62, 575e593. Edgcumbe, D.J., 1980. Some preliminary observations on the submerged aquatic Zostera capensis Setchell. Journal of South African Botany 46, 53e66. Elliott, M., Kingston, P.F., 1987. The sublittoral benthic fauna of the Estuary and Firth of Forth. Proceedings of the Royal Society of Edinburgh 93B, 449e465. Elliott, M., Quintino, V., 2007. The estuarine quality paradox, environmental homeostasis and the difficulty of detecting anthropogenic stress in naturally stressed areas. Marine Pollution Bulletin 54, 640e645. Elliott, M., Whitfield, A.K., 2011. Challenging paradigms in estuarine ecology and management. Estuarine, Coastal and Shelf Science 94, 306e314. Gómez, N., Hualde, P.R., Licursi, M., Bauer, D.E., 2004. Spring phytoplankton of the Río de la Plata: a temperate estuary of South America. Estuarine, Coastal and Shelf Science 61, 301e309. Gray, C.A., McDonnall, V.C., Reid, D.D., 1990. By-catch from prawn trawling in the Hawkesbury River, New South Wales: species composition, distribution and abundance. Australian Journal of Marine and Freshwater Research 41, 13e26. Grindley, J.R., 1981. Estuarine plankton. In: Day, J.H. (Ed.), Estuarine Ecology with Particular Reference to Southern Africa. A.A. Balkema, Cape Town, pp. 117e146. Grindley, J.R., Heydorn, A.E.F., 1970. Red water and associated phenomena in St Lucia. South African Journal of Science 67, 210e213.

A.K. Whitfield et al. / Estuarine, Coastal and Shelf Science 97 (2012) 78e90 Hanekom, N., 1989. A note on the effects of a flood of medium intensity on macrobenthos of soft substrata in the Swartkops estuary, South Africa. South African Journal of Marine Science 8, 349e355. Hedgpeth, J.W., 1967. Ecological aspects of the Laguna Madre, a hypersaline estuary. In: Lauff, G.H. (Ed.), Estuaries. American Association for the Advancement of Science, Washington, D.C., pp. 408e419. Publication No. 83. Heemstra, P., Heemstra, E., 2004. Coastal Fishes of Southern Africa. NISC and SAIAB, Grahamstown: 488 pp. Howard-Williams, C., 1980. Aquatic macrophytes of the coastal wetlands of Maputaland. In: Bruton, M.N., Cooper, K.H. (Eds.), Studies on the Ecology of Maputaland. Rhodes University, Grahamstown and Wildlife Society, Durban, pp. 42e51. Hudson, J.D., 1963. The recognition of salinity controlled mollusc assemblages in the Great Estuarine Series (Middle Jurassic) of the Inner Hebrides. Palaeontology 6, 318e326. Hudson, J.D., 1980. Aspects of brackish-water facies from the Jurassic of north-west Scotland. Proceedings of Geologists’ Association 91 (1e2), 99e105. Hudson, J.D., 1990. Salinity from faunal analyses and geochemistry. In: Briggs, D.E.G., Crowther, P.R. (Eds.), Palaeobiology, a Synthesis. Blackwell Scientific, Oxford, pp. 406e408. Jaureguizar, A.J., Menni, R., Bremec, C., Mianzan, H., Lasta, C., 2003. Fish assemblage and environmental patterns in the Río de la Plata Estuary. Estuarine, Coastal and Shelf Science 56 (5e6), 921e933. Jones, A.R., 1987. Temporal patterns in the macrobenthic communities of the Hawkesbury estuary, NSW. Australian Journal of Marine and Freshwater Research 38, 607e624. Jones, A.R., Watson-Russell, C.J., Murray, A., 1986. Spatial patterns in the macrobenthic communities of the Hawkesbury estuary, New South Wales. Australian Journal of Marine and Freshwater Research 37, 521e543. Kaiser, M.J., Atrill, M.J., Jennings, S., Thomas, D.N., Barnes, D.K.A., Brierly, A.S., Polunin, N.V.C., Raffaelli, D.G., Williams, P. le B, 2005. Marine Ecology. Processes, Systems, and Impacts. Oxford University Press, Oxford: 557 pp. Khedr, A.-H.A., 1998. Vegetation zonation and management in the Damietta Estuary of the River Nile. Journal of Coastal Conservation 4 (1), 79e86. Khlebovich, V.V., 1969. Aspects of animal evolution related to critical salinity and internal state. Marine Biology 2 (4), 338e345. Kolpakov, N.V., Milovankin, P.G., 2010. Distribution and seasonal changes in fish abundance in the estuary of the Razdol’naya River (Peter the Great Bay), Sea of Japan. Journal of Ichthyology 50 (6), 445e459. Lenanton, R.C.J., 1977. Fishes from the hypersaline waters of the stomatolite zone of Shark Bay, Western Australia. Copeia 1977 (2), 387e390. Lewis, E.L., 1980. The practical salinity scale 1978 and its antecedents. IEEE Journal of Oceanic Engineering 5 (1), 3e8. MacKay, C.F., Cyrus, D.P., Russell, K.-L., 2010. Macrobenthic invertebrate responses to prolonged drought in South Africa’s largest estuarine lake complex. Estuarine, Coastal and Shelf Science 86, 553e567. McDowall, R.M., 1976. The role of estuaries in the life cycles of fishes in New Zealand. Proceedings of the New Zealand Ecological Society 23, 27e32. McLusky, D.S., Elliott, M., 2004. The Estuarine Ecosystem; Ecology, Threats and Management, third ed. Oxford University Press, Oxford: 216 pp. McLusky, D.S., Hull, S.C., Elliott, M., 1993. Variations in the intertidal and subtidal macrofauna and sediments along a salinity gradient in the Upper Forth estuary. Netherlands Journal of Aquatic Ecology 27, 101e109. Mouny, P., Dauvin, J.C., Bessineton, C., Elkaim, B., 1998. Biological components from the Seine Estuary: first results. Hydrobiologia 373/374, 333e347. Nanami, A., Saito, H., Akita, T., Motomatsu, K.-I., Kuwahara, H., 2005. Spatial distribution and assemblage structure of macrobenthic invertebrates in a brackish lake in relation to environmental variables. Estuarine, Coastal and Shelf Science 63 (1e2), 167e176. Pauly, D., 1982. The fishes and their ecology. In: Pauly, D., Mines, A.N. (Eds.), Smallscale Fisheries of San Miguel Bay, Philippines: Biology and Stock Assessment. ICLARM Technical Report 7, pp. 15e33. Powis, B.J., Robinson, K.I.M., 1980. Benthic macrofaunal communities in the Tuggerah lakes, New South Wales. Australian Journal of Marine Freshwater Research 31, 803e815. Quintino, V., Sangiorgio, F., Mamede, R., Ricardo, F., Sampaio, L., Martins, R., Freitas, R., Rodrigues, A.M., Basset, A., The leaf-bag and the sediment sample: two sides of the same ecological quality story? Estuarine, Coastal and Shelf Science, in press. Remane, A., 1934. Die Brackwasserfauna. Verhandlungen Der Deutschen Zoologischen Gesellschaft 36, 34e74. Remane, A., Schlieper, C., 1958. Die Biologie des Brackwassers. Schweizerbart’sche Verlagsbuchhandlung, Stuttgart: 348 pp. Ribeiro, J.P.N., Matsumoto, R.S., Takao, L.K., Peret, A.C., Lima, M.I.S., 2011. Spatial distribution of Crinum americanum L. in tropical blind estuary: hydrologic, edaphic and biotic factors. Environmental and Experimental Botany 71 (2), 287e291. Richardson, N., Whitfield, A.K., Paterson, A.W., 2006. The influence of selected environmental parameters on the distribution of the dominant demersal fishes in the Kariega Estuary channel, South Africa. African Zoology 41 (1), 89e102. Rochford, D.J., 1951. Studies in Australian estuarine hydrology. I. Introductory and comparative features. Australian Journal of Marine Freshwater Research 2, 1e116. Rodrigues, A.M., Quintino, V., Sampaio, L., Freitas, R., Neves, R., 2011. Benthic biodiversity patterns in Ria de Aveiro, Western Portugal: environmental biological relationships. Estuarine, Coastal and Shelf Science, doi:10.1016/ j.ecss.2011.05.019.

89

Rosa-Filho, J.-S., Bemvenuti, C.E., Elliott, M., 2004. Predicting biological parameters of estuarine benthic communities using models based on environmental data. Brazilian Archives of Biology and Technology 47 (4), 613e627. Roy, P.S., 1984. New South Wales estuaries e their origin and evolution. In: Thom, B.G. (Ed.), Developments in Coastal Geomorphology in Australia. Academic Press, New York, pp. 99e121. Roy, P.S., Williams, R.J., Jones, A.R., Yassini, I., Gibbs, P.J., Coates, B., West, R.J., Scanes, P.R., Hudson, J.P., Nichol, S., 2001. Structure and function of south-east Australian estuaries. Estuarine, Coastal and Shelf Science 53, 351e384. Sainty, G.R., Jacobs, S.W.L., 1994. Water Plants in Australia. Sainty and Associates, Darlinghurst, NSW. Sapozhnikov, F.V., Ivanishcheva, P.S., Simakova, U.V., 2009. Modern assemblage changes of benthic algae as a result of hypersalinization of the Aral Sea. Journal of Marine Systems 76 (3), 343e358. Schlacher, T.A., Wooldridge, T.H., 1996. Axial zonation patterns of subtidal macrozoobenthos in the Gamtoos estuary, South Africa. Estuaries 19, 680e696. Sheaves, M., Johnston, R., Molony, B., Shepard, G., 2007. The effects of impoundments on the structure and function of fish fauna in a highly regulated dry tropics estuary. Estuaries and Coasts 30 (3), 507e517. Sinha, M., Mukhopadhyay, M.K., Mitra, P.M., Bagchi, M.M., Karamkar, H.C., 1996. Impact of Farraka Barrage on the hydrology and fishery of Hooghly Estuary. Estuaries 19 (3), 710e722. Skelton, P.H., 2001. A Complete Guide to the Freshwater Fishes of Southern Africa. Struik, Cape Town: 395 pp. Skelton, P.H., Whitfield, A.K., James, N.P.E., 1989. Distribution and diversity of Mkuze swamp fishes during a summer flood. Southern African Journal of Aquatic Sciences 15, 50e66. Sylvestre, F., Beck-Eichler, B., Duleba, W., Debenay, J.-P., 2001. Modern benthic diatom distribution in a hypersaline coastal lagoon: the Lagoa de Araruama (R.J.), Brazil. Hydrobiologia 443, 213e231. Taylor, C.J.L., 1993. The zooplankton of the Forth Estuary. Aquatic Ecology 27 (2e4), 87e99. Telesh, I.V., Khlebovich, V.V., 2010. Principal processes within the estuarine salinity gradient: a review. Marine Pollution Bulletin 61, 149e155. Telesh, I.V., Schubert, H., Skarlato, S.O., 2011. Revisiting Remane’s concept: evidence for high plankton diversity and a protistan species maximum in the horohalinicum of the Baltic Sea. Marine Ecology Progress Series 421, 1e11. Ter Morshuizen, L.D., Whitfield, A.K., 1994. The distribution of littoral fish associated with eelgrass Zostera capensis in the Kariega estuary, a southern African system with a reversed salinity gradient. South African Journal of Marine Science 14, 95e105. Ter Morshuizen, L.D., Whitfield, A.K., Paterson, A.W., 1996. Influence of freshwater flow regime on fish assemblages in the Great Fish River and estuary. Southern African Journal of Aquatic Sciences 22, 52e61. Ter Morshuizen, L.D., Whitfield, A.K., Paterson, A.W., 1997. Distribution patterns of fishes in an Eastern Cape estuary and river with particular emphasis on the ebb and flow region. Transactions of the Royal Society of South Africa 51, 257e280. Teske, P.R., Wooldridge, T.H., 2003. What limits the distribution of subtidal macrobenthos in permanently open and temporarily open/closed South African estuaries? Salinity vs. sediment particle size. Estuarine, Coastal and Shelf Science 57, 225e238. Teske, P.R., Wooldridge, T.H., 2004. Affinities of some common estuarine macroinvertebrates to salinity and sediment type: empirical data from Eastern Cape estuaries, South Africa. African Zoology 39 (2), 183e192. Vega-Cendejas, M.E., de Santillana, M.H., 2004. Fish community structure and dynamics in a coastal hypersaline lagoon: Rio Lagartos, Yucutan, Mexico. Estuarine, Coastal and Shelf Science 60 (2), 285e299. Wallace, J.H., 1975. The estuarine fishes of the east coast of South Africa. Part 1. Species composition and length distribution in the estuarine and marine environments. Part 2. Seasonal abundance and migrations. Investigational Report of the Oceanographic Research Institute 40, 1e72. Ward, C.J., 1976. Aspects of the ecology and distribution of submerged macrophytes and shoreline vegetation at Lake St Lucia. In: Heydorn, A.E.F. (Ed.), St Lucia Scientific Advisory Council Workshop Meeting e Charters Creek 15e17 February 1976. Natal Parks, Game and Fish Preservation Board, Pietermaritzburg. West, R.J., Larkum, A.W.D., King, R.J., 1989. Chapter 8: the seagrasses of south eastern Australia. In: Larkum, A.W.D., McComb, A. (Eds.), Seagrass Ecosystems e An Australian Perspective. Elsevier, Amsterdam. West, R.J., Walford, T.R., 2000. Estuarine fishes in two large eastern Australian coastal rivers e does prawn trawling influence fish community structure? Fisheries Management and Ecology 7, 523e536. West, R.J., King, R.J., 1996. Marine, brackish, and freshwater fish communities in the vegetated and bare shallows of an Australian Coastal River. Estuaries 19, 31e41. Whitfield, A.K., 1994. Fish species diversity in southern African estuarine systems: an evolutionary perspective. Environmental Biology of Fishes 40, 37e48. Whitfield, A.K., 2005. Preliminary documentation and assessment of fish diversity in sub-Saharan African estuaries. African Journal of Marine Science 27 (1), 307e324. Whitfield, A.K., Blaber, S.J.M., 1976. The effects of temperature and salinity on Tilapia rendalli Boulenger 1896. Journal of Fish Biology 9, 99e104. Whitfield, A.K., Harrison, T.D., 2003. River flow and fish abundance in a South African estuary. Journal of Fish Biology 62, 1467e1472. Whitfield, A.K., Paterson, A.W., Bok, A.H., Kok, H.M., 1994. A comparison of the ichthyofaunas in two permanently open eastern Cape estuaries. South African Journal of Zoology 29, 175e185.

90

A.K. Whitfield et al. / Estuarine, Coastal and Shelf Science 97 (2012) 78e90

Whitfield, A.K., Taylor, R.H., Fox, C., Cyrus, D.P., 2006. Fishes and salinities in the St Lucia system e a review. Reviews in Fish Biology and Fisheries 16, 1e20. Wilkinson, M., 1980. Estuarine benthic algae and their environment: a review. In: Price, J.H., Irvine, D.E.G., Farnham, W.F. (Eds.), The Shore Environment. Ecosystems, vol. 2. Academic Press, London and New York, pp. 425e486. Wilkinson, M., Telfer, T., Grundy, S., 1995. Geographical variation in the distributions of macroalgae in estuaries. Aquatic Ecology 29 (3e4), 359e369. Wilkinson, W., Wood, P., Wells, E., Scanlan, C., 2007. Using attached macroalgae to assess ecological status of British estuaries for the European Water Framework Directive. Marine Pollution Bulletin 55, 136e150.

Wooldridge, T.H., 1999. Estuarine zooplankton community structure and dynamics. In: Allanson, B.R., Baird, D. (Eds.), Estuaries of South Africa. Cambridge University Press, Cambridge, pp. 141e166. Wooldridge, T.H., Deyzel, S.H.P., 2009. Temperature and salinity as abiotic drivers of zooplankton community dynamics in the Great Berg Estuary. Transactions of the Royal Society of South Africa 64 (2), 219e237. Young, G.C., Potter, I.C., 2002. Influence of exceptionally high salinities, marked variations in freshwater discharge and opening of estuary mouth on the characteristics of the ichthyofauna of a normally-closed estuary. Estuarine, Coastal and Shelf Science 55 (2), 223e246.