SAB INSTITUTE FOR ENVIRONMENTAL AND COASTAL MANAGEMENT
Environmental Consulting and Research

University Of Port Elizabeth, P. O. Box 1600, Port Elizabeth, 6000, South Africa
041 – 5042747  041 – 5832317  iecm@upe.ac.za

KOUGA MUNICIPALITY
ST FRANCIS BAY BEACH EROSION
ENVIRONMENTAL IMPACT ASSESSMENT

SPECIALIST REPORT

ECOLOGICAL IMPACT ASSESSMENT OF THE PREFERRED TECHNICAL ALTERNATIVES

Prepared by

Norbert Klages, Eileen Campbell,
Dave Schoeman,Tris Wooldridge
 

IECM REPORT Number C69

JULY 2002
 

TABLE OF CONTENTS (Click to go to Relative Section)
Executive Summary

1. Introduction

1.1 Background

1.2. Brief and terms of reference

1.3. Study area

1.4. Approach to the study

1.5 Structure of the report

2. The affected environment

2.1. The dune environment

2.1.1. Eroded foredune complex

2.1.2. Marine Glades barrier dune

2.1.3. Vegetated dunes north of the Kromme River

2.2. The beach environment

2.2.1. Conceptual structure of a beach

2.2.2. Biotic assemblages inhabiting beaches

2.2.3. Fauna inhabiting the beaches under consideration

2.2.4. Role of beaches in the coastal ecology

2.3. The estuary environment

2.3.1. General introduction

2.3.2. Regulation of sediment accumulation in estuaries

2.3.3. The Kromme River estuary

2.4. The Sand River dune field

2.4.1. Sand River dunefield site description

2.4.2. The vegetation of the Sand River dune field

2.4.3. The fauna of the Sand River dune field

3. Sources of risk and impact assessment

3.1. Risks to and impacts on the dune environment

3.2. Risks to and impacts on the beach environment

3.3. Risks to and impacts in the estuary

3.4. Impacts of sand mining in the Sand River dune field

4. Recommended mitigation measures and management actions

4.1. Dune environment

4.2. Beach environment

4.2.1. Sources of sand

4.2.2. Transport and deposition of sand

4.2.3. Foredune formation

4.2.4. Construction and operation

4.3. Estuarine environment

4.4. Sand River dune field environment

5. Conclusions and recommendations

5.1. Conclusions from the dune perspective

5.2. Conclusions from the beach perspective

5.3. Conclusions from the estuary perspective

5.4. Conclusions from the Sand River dune field perspective

6. Cited literature

APPENDIX

EXECUTIVE SUMMARY
An assessment is provided of the environmental impact of the construction of the different engineering options to address coastal erosion problems as presented by Entech as well as the impact of increased sand deposition on the dune environment of the St. Francis Bay Beach.

The foredune complex backed by the town development of St. Francis Bay has been severely eroded to the extent that the functional dune system has been destroyed and its natural fluctuations have ceased. With the loss of dune malleability, the protective role of the foredune complex has been lost. The inland side of the Marina Glades dune is largely stable and will remain so provided the erosive action of canal water is kept low. The stability of this dune spit is extremely tenuous because it is surrounded by water and adjacent to an estuary mouth. Measures taken to replenish the sand volume of the St. Francis Bay Beach address only one of the impacts currently acting on the foredune complex. Unless action is taken to address the other erosive forces, the problems are likely to persist and continual action will be required to maintain beachfront integrity.

Beach functioning is already compromised. With the current rate of erosion, it may be concluded that failure to intervene will eventually result in the loss of the local beach as a functional ecological system.

Freshwater inflow into the Kromme River is significantly attenuated by two large storage reservoirs in the catchment, so that the estuary becomes an extension of the sea most of the time. Effective removal of accumulated sediment by occasional floods in the estuary is diminished. Species composition and distribution as well as biological processes in the estuary differ from other estuarine systems that receive an adequate supply of freshwater.

The risks to the dune environment all relate to interference with dune malleability. Developments close to sandy shores are only sustainable if the natural sand transport patterns are maintained. These patterns have been so seriously altered by developments in St. Francis Bay that the foredune dune system along much of the beach has ceased to function.

Most risks to the beach arise from the compaction of sand during construction activity as well as the addition of poorly matched sediments containing silt, clay and organic components. Erosion downstream from groynes presents a major risk.

Most impacts on the estuary relate to the removal of sand, leading to changes in mouth and river channel configuration. A further risk arises from the flooding of the Kromme River.

The sand banks near the mouth of the Kromme River estuary and the open (unvegetated) dunes around the Sand River seem best suited to providing the sand for beach restoration and nourishment. Mining of 600 000 m3 and 1 000 000 m3 of sand, respectively, is feasible. Most of the vegetation found on the Sand River dune system is exotic making its conservation value minimal. Sand in these vegetated areas, however, will contain plant litter (organic detritus). If sand is taken from the vegetated areas and deposited on the St. Francis Bay Beach, it will cause an undesirable black colour on the beach and the decomposition of litter will affect the beach ecosystem. Sand removal from the Sand River dune system should therefore be confined to unvegetated areas. This will result in loss of the only positive impact, namely the removal of alien trees, but will completely mitigate for the negative impacts. Seasonal vleis and pans should not be mined to protect the resident small vertebrate fauna. No negative environmental impacts are predicted if sand was taken from the municipal dump site next to the R330 towards Cape St Francis although the volume is small and may contain rubbish. The grey colour of the sand excavated from the canal extension is likely a result of its high sulphur content deriving from anoxic conditions thus making it the least desirable option.

Mitigation of the most important impacts is possible. Fatal flaws preventing the proposed development to go ahead were not discovered. Construction of a set of three relatively short groynes, combined with initial beach nourishment, presents the environmentally preferred alternative.

1. INTRODUCTION

1.1. Background
In January 2002 the SAB Institute for Environmental and Coastal Management at the University of Port Elizabeth was appointed by SRK Consulting to assist with the ecological components of an Environmental Impact Assessment (EIA) for a suite of proposed engineering measures aimed at improving coastal protection at St. Francis Bay. The IECM’s brief was subsequently expanded (in July 2002) to include ecological assessments of the impacts of sand sourcing for the proposed beach remediation.

Currently St. Francis Bay Beach is eroding at 2.5 - 3 metres a year, threatening seafront properties and infrastructure. Preceding technical studies have indicated that the construction of groynes and/or beach nourishment would be a means of reducing long-term beach erosion problems experienced on the St Francis Bay Beach. Several alternative mitigation measures to reduce coastal erosion, such as offshore structures and revetments, were also considered.

The principal objectives of the EIA are to:

• Assess the potential impacts associated with the construction and operation of the proposed groynes and/or beach nourishment alternative at St Francis Bay;

• Recommend appropriate and practicable mitigation measures to minimise the negative impacts and maximise potential benefits; and

• Indicate the environmentally preferred alternative.

The scoping phase of the EIA process was completed in 1999 with the submission of a Scoping Report to the Department of Economic Affairs, Environment and Tourism (DEAET). Inter alia the Scoping Report proposed that a sandy beach, a dune and an estuarine ecological impact assessment of the preferred technical alternative(s) be undertaken.

This report assesses the environmental impact of the construction of the different engineering options as presented by Entech (2002a) as well as the impact of increased sand deposition on the dune environment of the St. Francis Bay Beach. Also assessed are the ecological impacts of sand sourcing from the lower Kromme estuary and from the Sand River dune field as well as two other smaller sources, the municipal dumpsite and a stockpile excavated from marina canal extensions. Details of this proposed sand mining are given in Entech’s Technical Study of Sand Sources (2002b).

1.2. Brief and Terms of Reference
A. Review available data regarding the existing ecology of each of the following environments:

• Dune ecology, particularly during construction (including the spit of land between the canals and the beach, and dunes to the north of the Kromme River mouth).

• Marine (sandy beach) ecology (including St Francis Bay Beach and beaches north of the Kromme River mouth);

• Estuarine ecology of the Kromme River mouth and estuary;

• Dune ecology of the Sand River dune field.

 
1.3. Study area
The study area where the various beach restoration options were considered is referred to as the St. Francis Bay Beach and stretches unbroken in a northeasterly direction from the rocky headland at Hewlett’s Reef near the boundary of Portion 2 of Goed Geloof 745 of St. Francis Bay to the Kromme River mouth. Comment is also included on the dune system to the north of the Kromme River mouth (Brakfontein 701 of St. Francis Bay) bearing in mind the nature both of the erosion problem and of the proposed interventions (Figure 9). The eastermost portion of the Sand River dune complex, extending approximately 3.5 km south-westward from the tarred R330 (see Map 1), was considered in detail as the major source of sand for beach restoration. This is part of a headland bypass dune field of more than 10 km in length that consists of mobile wind-blown mounds of sand. The sandbanks in the lower Kromme River estuary were also investigated as sources of sand, as was a dune located on the municipal waste dump site next to the R330 towards Cape St Francis and a stockpile excavated from extensions of the Marina Glades canal system.

1.4. Approach to the study
Approach and information sources
The following activities were undertaken in order to meet the objectives of this specialist report:

• Site visits to the lower reaches of the Kromme estuary, St. Francis Bay Beach and the lower Sand River to determine the nature of the environment and help identify potential issues of concern.

• The review of the ecology of the St. Francis Bay Beach, the adjacent dune environment, the Kromme River estuary and the lower Sand River is based on available information and did not include new investigations.

• A review of the literature addressing the nature of sandy beach ecosystems and the ways in which they react to human disturbance of different types; and

• A qualitative assessment of the likely effects of the various intervention strategies proposed.

Assumptions and limitations
The assumption is made that increased sand deposition will follow the construction of groynes or breakwaters, and/or that the sand volume will be increased by artificial beach nourishment as proposed. This follows from the Entech Coastal Erosion Impact Assessment (2002a), which concluded that only two erosion control measures are viable for implementation at St Francis Bay:

1. Beach restoration and continuous nourishment: This intervention involves the initial “restoration” of the beach by addition of roughly 1 000 000 m3 of sand, thereby moving the waterline about 30 m seaward of its current position. Thereafter, the process of “nourishment” will add 50 000 to 100 000 m3 of sand to the beach each year in order to replace sand lost to natural erosion processes.

A set of three relatively short groynes distributed uniformly along the beach: Under this scenario, three groynes would be constructed perpendicular to the coast. Each would be about 250 m long and they would be about 850 m apart. During the construction phase, some initial beach restoration is also envisaged.

It should be noted that because very few ecological data have been collected from the beaches under consideration (only a single, anonymous, confidential report could be located), most findings contained in this report have had to be inferred from a general understanding of South African beach and dune ecology, aided by professional judgment. This understanding was obtained from a thorough review of the available literature describing aspects of beach and dune ecology and from many years of personal experience in ecological research. The assessment of potential impacts and mitigation of negative impacts is based on coastal dune management principles founded on the definitive work of Tinley (1985) as taken into the Council for the Environment (1991) policy document.

Unfortunately, the constraints imposed on the study permitted no new measurements or sampling of any kind. Therefore, should any intervention strategy be accepted for implementation, the collection of detailed data in accordance with an appropriate sampling design should be considered a priority.

Likewise, available information on the flora and fauna of flood tidal delta systems in South Africa is limited. Discussion on the flora and fauna of sandy substrata in the lower Kromme River is mainly based on previous studies undertaken in other estuaries. Additional work on the flora and fauna of sandy substrata in the lower estuary was desirable and would have improved confidence levels when reporting on the ecology of this reach of the estuary.

Entech’s Technical Study of Sand Sources (2002b) identified two potential sources of sand that are technically deemed to be without conservation value for the purposes of this ecological impact assessment. As the potential environmental impacts of the Marina Glades canal extensions do not form part of this investigation the sand from there was treated as a stockpiled resource. The investigation was therefore restricted to the suitability of the excavated sediment for beach remediation. The remnant dune present on the municipal dump site is already severely impacted and has lost its conservation value entirely. Under these circumstances it is felt that it too can be treated as a stockpiled resource requiring no special consideration as an affected environment.

1.5. Structure of the report
This report is divided into five chapters:
 

 
2. THE AFFECTED ENVIRONMENT

2.1. THE DUNE ENVIRONMENT
There are two different dune environments in the St. Francis Bay Beach system and a third north of the Kromme River mouth:
The eroded foredune complex backed by housing development (Fig. 1);
The barrier dune between the beach and the Marina Glades Ski Canal (Fig. 2); and
The vegetated dunes north of the Kromme River (Fig. 3).

2.1.1. Eroded foredune complex
The foredune complex backed by the town development of St. Francis Bay is approximately 1.7 km long. The foredunes have been severely eroded (Fig. 1), to the extent that the functional dune system has been destroyed and its natural fluctuations have ceased. With the loss of dune malleability [pliability, ability to absorb impact], the protective role of the foredune complex has been lost.

The foredune complex would have consisted of a cordon of hummock dunes [mounds of sand] that function as the first sand trap for sand washed ashore and blown inland. Highly specialised plants grow just above the spring high water line. The presence of these plants results in a localised reduction in the near-surface wind velocity (Hesp 1981) causing sand deposition. Plants grow rapidly to remain emergent from the sand, and this process of sand deposition and plant growth leads to the formation of mounds of sand stabilised by dune hummock plant species. Because of the high growth rate of these plants, as well as the harsh environmental conditions (salt spray, mobile sand, low water availability) the plants are generally herbaceous to succulent [non-woody to water-preserving]. They are particularly sensitive to any disturbance (Heydorn 1986) and are usually the first component of the foredune complex to be lost under seaward-based erosive conditions. The dominant plant species on the hummocks was most likely Scaevola plumieri (L.) Vahl; Ipomoea pes-caprae (L.) R. Br.; Gazania rigens (L.) Gaertn. var. uniflora (L. f.) Rössl. and Tetragonia decumbens Mill. There remains no evidence of the natural occurrence of these species at the St. Francis Bay Beach in this western 1.7 km section. This indicates the complete loss of this component of the dune system.

Behind the hummock dune cordon, reduced sand deposition and lower salt spray results in the development of sufficient vegetation cover to stabilise the sand. The first stabilised dune is the youngest and will have smaller fluctuations in sand volume. This dune is referred to as the primary foredune ridge and it is the most sensitive of the foredunes (Tinley 1985) because of the higher plant cover. More plant species will tolerate these conditions compared to the hummock dunes and the species diversity is generally higher on the primary foredune ridge compared to hummocks (Lubke and De Moor 1998). Taller shrubs and creepers grow on the foredune ridge. They have an even greater ability to trap sand than hummock species. The apparently tenuous role of plant growth in holding the sand in place on the foredune cannot be overemphasised (Heydorn 1986). The primary foredune ridge still exists at St. Francis Bay Beach, but its hypersensitivity to disturbance is clearly evident at this site. Seemingly trivial development features, such as a wrongly located footpath, have had far-reaching and large-scale erosive effects.

Disturbance to the primary foredune ridge has occurred from the seaward side (the main concern to be addressed by the proposed construction according to the Entech report of 2002), but also from the landward side. The combined effects have resulted in large-scale erosion of the ridge. The disturbances acting on the foredune ridge are as follows:

Alteration of natural sand transport
The major feature of a sandy shoreline backed by dunes (also known a littoral active zone) is the continuous movement of sand, rendering the system highly malleable (Tinley 1985). The maintenance of the system is dependent on this malleable state. The stabilisation of the St Francis Bay (Santareme) headland bypass dune to allow development of St. Francis Bay has resulted in alteration of the natural fluctuations of the system (WPR 1993, Entech 2002a). The WPR report of 1993 clearly outlines how this has resulted in narrowing of the beach (Fig. 5.2 a in WPR 1993).

Stormwater runoff
In a natural system, dune sands are highly porous and rainfall soaks away gently into the aquifer from where freshwater seeps into the surf-zone in the intertidal beach (Campbell and Bate 1991). Paving of roads, walkways and parking areas results in stormwater drainage in the form of fast-flowing water. Uncontrolled stormwater damage of the primary foredune ridge is evident at the site (Fig. 4).

Access footpaths
The key feature controlling rampant sand movement in dunes is the highly specialised dune vegetation. These plant species are particularly sensitive to any physical disturbance such as trampling (Heydorn 1986). They die rapidly if impacted, resulting in unobstructed movement of sand by wind or water (Fig. 5).

Sand removal by wave action
The remedial aspect at which the proposed construction is aimed is that of sand starvation of the beach. Dune systems naturally lose sand by water and wind and it follows that maintenance of dune integrity includes sand deposition as part of replenishment. The sand-starvation of the beach has been described (WPR 1993, Entech 2002a) and the movement of the beach landward is clearly evident as erosion by high tide wave action (Fig. 6). It is important to note, however, that the proposed construction only addresses one of the causes of system failure at the St. Francis Bay Beach, albeit it is the main cause.

It is clear from the foregoing that if no action is undertaken to remedy the situation, it is highly probable that the combination of seaward and landward erosive forces will destroy the remainder of the foredune, the tarred road immediately landward of it, and the first row of houses. Some action will obviously be necessary to prevent this from happening.

2.1.2. The Marina Glades barrier dune
The eastern third of the St. Francis Bay Beach (ca. 1 km) is backed by a 2 to 4 m high dune cordon that forms a barrier between the beach and the waterways and canals of the marina (Fig. 2 and 7). The barrier dune extends into the Kromme Estuary mouth in the form of a sand spit (Fig. 8). On the seaward side of this dune, much of the natural foredune processes still occur: several foredune hummocks break the first force of the wind, particularly in the east; and the primary foredune still functions as a sand trap (Fig. 7). The malleable nature of the dune is fundamentally intact on the seaward side, although the western portion suffers similar erosion as the foredune complex (see previous section).

The barrier dune has low plant species diversity: the vegetation is heavily infested with the exotic rooikrans (Acacia cyclops A. Cunn. ex G. Don). The tick-berry bush or bietou (Chrysanthemoides monilifera (L.) T. Norl.) is the most abundant indigenous plant species. The plants are essential to the maintenance of the stability of this barrier dune.

The inland side of the dune is largely stable and will remain so provided the erosive action of canal water is kept low.

The eastern extreme of the dune is affected by the Kromme estuary discharge and the erosive force of the water flow is aggravated by the presence of the marina canal and associated dredging activities. These actions have placed severe pressure on the maintenance of the dune (Fig. 8) and artificial stabilisation efforts are in evidence. The stability of this dune spit is extremely tenuous because it is surrounded by water and adjacent to an estuary mouth.

2.1.3 Vegetated dunes north of the Kromme River
The dunes north of the Kromme River are much higher than those of the St. Francis Bay Beach and the volume of sand contained in the system is large. The dunes are well vegetated, thereby ensuring surface stability of the sand and, due to the large sand volume, alterations of sand supply and transport will have a much smaller effect than on a low foredune with a small sand volume.

2.2. THE BEACH ENVIRONMENT

2.2.1. Conceptual structure of a beach
For the purposes of this study, a sandy beach will be defined as that portion of the sandy oceanic coastline that falls between the outer breakpoint in the surf zone and the drift-line closest to the base of the primary dunes (McLachlan et al. 1981a). Together with the adjacent foredunes, the beach forms the littoral active zone, which is an integrated geomorphic system, characterised by processes related to wind- and wave-driven sand transport (Brown and McLachlan 1990). Despite this close linkage, however, the beach proper, which is marine dominated and is controlled primarily by wave energy, remains an ecological unit that is discrete from the dune system, which is inhabited by terrestrial life-forms and is more strongly influenced by wind energy (McLachlan 1983). It is therefore appropriate to consider separate management objectives for the beach and dunes of any given coastal system.

The beach itself may conveniently be sub-divided into two components (McLachlan 1983). The first is the surf zone, which extends from the line of breakers furthest from shore to the low water mark. This area is exclusively marine and is not physically or ecologically separated from the nearshore ocean. The second component is the intertidal, which is more clearly demarcated, extending between the spring high- and low-water marks, thereby forming an area of transition between marine and terrestrial coastal ecosystems.

It is clear that all constituent components of healthy sandy beaches are dynamic over a range of scales in space and time. For example, the extent of the surf zone varies constantly as a result of differing incident swell height and direction, whereas the intertidal zone varies in width over the spring-neap tidal cycle. By the same token, the effects of swell height and tidal cycles vary among times and places on the basis of nearshore topography and sediment characteristics. Resident fauna mirror this instability in that they are highly mobile and are indeed precluded from constructing permanent burrows both by the turbulent energy associated with a wave dominated environment and by its dramatic alterations over small spatial and temporal scales (Brown and McLachlan 1990). For similar reasons, beaches are also characterized by the absence of rooted plants of any sort.

2.2.2. Biotic assemblages inhabiting beaches
Almost all information pertaining to beach communities has been acquired from intertidal infaunal assemblages [see for example Brown (1964), Defeo and De Alava (1995), Jaramillo (1987), Jaramillo and McLachlan (1993), McArdle and McLachlan (1991, 1992), McLachlan (1977, 1983, 1996, 2001), McLachlan and Hesp (1984), and McLachlan et al. (1981a, 1981b, 1993, 1996)]; considerably less is known about the functioning of surf zone ecosystems [see for example Bentley and Cockcroft (1995), Du Preez (1990), McLachlan and Bate (1985) and McLachlan et al. (1981a, 1984)], with the exception of their phytoplanktonic component (Campbell 1996, Campbell and Bate 1996). This is primarily a function of the ease with which the respective habitats can be sampled, but it also reflects human activity patterns and the perceived susceptibility of these compartments to anthropogenic disturbances. Of greatest importance to the intent of this report is the fact that many of the species inhabiting the surf zone of beaches are also inhabitants of the deeper waters of the nearshore ocean. By contrast, species inhabiting the intertidal zone are unique to this spatially limited habitat, being found neither above the high water mark nor below the low water mark in any significant abundance. Because there is no clear ecological boundary between the surf zone and the adjacent nearshore ocean and because so little is known about the structure and function of surf zone ecosystems, the remainder of this report will focus primarily on intertidal ecology. Nevertheless, pertinent remarks on surf zone ecology will be made wherever they are deemed necessary.

In general, macrofaunal species richness and abundance increase across any given beach from the high water mark seaward (McLachlan 2001). This is a natural response to the gradient in aerial exposure, which poses the greatest physiological demands on organisms living near the high water mark (Cockcroft 1990). However, these patterns are often modified by the tidal migratory behaviour of many beach organisms (McLachlan et al. 1979, McGwynne and McLachlan 1985, Brown and McLachlan 1990), which use their motility to follow the waterline up and down the beach as the tide rises and recedes. In this way, the typical zones that are evident on beaches at low tide (Jaramillo 1987, McLachlan and Jaramillo 1995, McLachlan et al. 1996) might become compressed or even superimposed as the tide rises and marine organisms can penetrate further up the intertidal. This is of particular concern on St Francis Bay beach, where significant portions of the intertidal are separated from the foredunes by revetments. In many places, these revetments fall within the intertidal zone during spring tides. While the implications of this disturbance for the intertidal macrobenthos are unquantified, it is possible that they might be so severe as to negatively affect species richness and abundance.

In terms of comparisons among beaches occurring in any given region, there are a number of features that can be used to predict macrobenthic abundance and diversity. These are: incident wave energy, sand particle size and beach slope (McLachlan et al. 1993, McLachlan 1996, 2001). For any given incident wave energy (assuming that this parameter is fairly constant within a region), macrobenthic abundance and diversity tend to increase as sand particle size becomes finer and/or beaches become flatter.

Flat, fine-grained beaches are referred to as dissipative because they tend to have wide surf zones over which incident wave energy is dissipated as waves break sequentially on ever shallower sandbars before forming a bore that runs up the intertidal (known as a swash). This type of beach morphology is considered to be ecologically benign, allowing it to harbour abundant, rich faunas. At the opposite end of the spectrum are reflective beaches, which have very narrow surf zones, so that waves break directly onto the steep intertidal, with the result that much of this energy is reflected back into the water column. Reflective beaches are considered to be ecologically harsh and to have depauperate, sparse faunas. Between the dissipative and reflective extremes are various intermediate forms. The St Francis Bay Beach falls toward the dissipative range of intermediate morphologies.

One of the prevailing paradigms of beach ecology (McLachlan 2001) suggests that the most dissipative beaches in a region tend to harbour all macrofaunal species present in that region, but that species are progressively lost as beaches become more reflective and less hospitable. This principle, together with the perceptions that intertidal species are in general distributed over large geographical ranges and that they disperse propagules over large spatial scales, suggest that beach faunas could be expected to exhibit low levels of endemism. In addition, it has frequently been suggested that beach faunas are so well adapted to regular physical disturbance by wind, waves and tides that the impacts of anthropogenic activities are negligible (Jaramillo et al. 1987, Rakocinski et al. 1996). The overall impression, therefore, is that beaches are highly resilient to mild to moderate human disturbance and that more emphasis in terms of coastal zone management should be placed on the adjacent dune systems (McGwynne and McLachlan 1992). Nevertheless, where anthropogenic disturbances are severe, as is the case here, specific cognizance must be taken of beach ecology.

2.2.3.The fauna inhabiting the beaches under consideration
The only quantitative data describing the fauna of St Francis Bay Beach are contained in a single, anonymous, confidential report. Furthermore, because these data were collected in March 1990, at which time the beach was already eroding, they do not represent ideal controls from which to gauge the natural state of the beach. Nevertheless, the faunas described were not unexpected. The intertidal communities at St Francis Bay Beach were fairly rich in terms of the number of species present (15-17), and infaunal communities were dominated by small, mobile crustaceans (amphipods and isopods), although whelks, clams, polychaete worms and surf zone shrimps were also common. An interesting feature of these communities was the high representation by scavengers, probably due to the presence of a rocky shore immediately up current. By contrast, the fauna inhabiting Paradise Beach, adjacent to the headland on the northern bank of the Kromme River mouth, was poor, with only 5 species present; all at very low abundances. This was attributed to the nature of the coastline, which consists of mixed sand and rock, and to the nature of the sediment, which comprised a thin layer of fine sand overlying a bed of coarse shell grit. In such sediments, burrowing is difficult, and rapid percolation of water through the upper layers results in low moisture contents and high desiccation stress for resident organisms. Rich faunas would therefore not be expected to develop under these conditions. However, it should be borne in mind that St Francis Bay Beach was already eroding at this time, and that Paradise Beach was probably receiving very little input of fine sediments via the alongshore current. This could explain the stratified nature of sediments at Paradise Beach and its attendant impoverished fauna.

2.2.4. The role of beaches in the coastal ecology
Beaches and dunes act in combination to form a coastal defence system against the effects of the sea (Brown and McLachlan 1990). In addition, beaches act as massive digestive systems (analogous to biological filters in aquaria) that play a crucial role in processing groundwater nutrients, which seep into the sea from the water table outcrop in the low intertidal, as well as nutrients associated with stranded wrack [stormcast plant or animal material] or oceanic upwelling. Placing the importance of beaches into perspective, it has been estimated that South African beaches filter more than 10 000 litres of sea water for every meter of shoreline each day. As a result of this filtering, all of the nutrients in the surf zone are completely recycled every eight days or so (McLachlan et al. 1981a).

To prevent physical damage to the shoreline, both beach and dune systems must be present. However, if beaches are not maintained in a healthy state, they cannot function properly, in which case they might become stagnant and contribute additional contaminants to the coastal ecosystem. While this scenario is considered fairly unlikely in medium-grained, wave dominated beaches, such as the one in St Francis Bay, this notion is contingent on the amount of nutrients being added to the system. Should nutrients reach the beach via seepage from ill-maintained sewerage systems, or from any alternative source, imbalances in the nutrient flow cycles might result either in potentially harmful microbial blooms or sulphurous plumes in the surf zone.

In addition to the roles that beaches play in protecting the coast from the actions of the sea and in maintaining coastal water quality, they are also areas of unique biodiversity. Species found within the intertidal of sandy beaches are generally absent from all other ecosystems. Although beaches are relatively resilient to moderate anthropogenic disturbance (McGwynne and McLachlan 1992), care should be taken not to overstress these environments; otherwise their ecological services to the coastal zone as a whole may be lost.

With the current rate of erosion on St Francis Bay beach, it may be concluded with some certainty that failure to intervene will eventually result in the loss of the local beaches as a functional ecological system. Most of the likely impacts from such a course of action are listed in the Entech report (2002a, p. 18). However, one additional ecological impact resulting from the loss of the intertidal beach is the inability of the intertidal community to assist in processing nutrients introduced to coastal waters via groundwater seepage and wrack stranding. Under certain circumstances, this might result in significant local deterioration of coastal water quality (e.g. murkiness, smell, health risk).

Some form of intervention will obviously be necessary to prevent further erosion. However, given the current state of the beach and the likelihood that its ability to function properly might already be compromised, it is strongly suggested that active measures be put in place to prograde [widen] the beach.

2.3. THE ESTUARY ENVIRONMENT

2.3.1. General introduction
An estuary is broadly defined as that part of a river system influenced by the sea. A salinity gradient becomes established that typically ranges from freshwater at the tidal head (near-zero salt content) to seawater at the mouth (salt concentration around 35 parts per thousand or 35 ppt). However, a number of factors operating over different time scales lead to temporal variations in the horizontal salinity profile; for example:

• the ebb and flow of the tides operate over time scales of a few hours,

• seasonal changes in evaporation and rainfall operate over time scales of months and

• shifts between wet and dry cycles operate over time scales of years.

Vertical differences in salinity are also common. The volume of freshwater inflow, the degree of tidal mixing and the topography of the estuary structure the vertical salinity profile. Tide-generated currents also have a major influence on sediment characteristics and the type of organisms found along the estuary. Where strong currents prevail, the substrate will be coarse (sand or gravel). Finer particles (e.g. silt) settle where waters are calm and the currents weak. The state of the tidal inlet also exerts an influence on the salinity gradient. In South Africa, many estuaries close off periodically (about 72%) as a result of the accumulation of sediment in the lower estuary.

The variable nature of estuarine habitats, especially defined by fluctuating salinity and substrate type, make estuaries a stressful and rigorous environment in which to live. Estuarine organisms must cope with problems unlike those of plants and animals living in purely marine or fresh waters. In general, estuaries are inhabited by far fewer species compared to adjacent ecosystems, but they rank among the most productive environments on earth. This is only possible if estuaries maintain functional links with both rivers and the sea. Very few organisms are able to tolerate the full salinity range found in estuaries and are broadly separated into three groups:

1. A marine associated component is the largest in terms of the number of species and are unable to tolerate, or barely able to tolerate salinity changes. This component is restricted to the tidal inlet region where salinity remains close to that of seawater.

2. Typical estuarine organisms that are able to tolerate varying degrees of salinity reduction below 30 ppt and can penetrate some distance up-estuary. This category includes estuarine endemic species. Many can tolerate salinity values down to about 15 ppt, with a few hardy species tolerating levels down to 3 ppt.

3. Those organisms tolerant of low salinity values found at the top of the estuary where salinity does not exceed about 5 ppt.

Although species richness is generally much lower compared to adjacent marine or freshwater habitats, a number of species are unique to estuaries and are well adapted to survive the variability of the physical environment.

2.3.2. Regulation of sediment accumulation in estuaries
Precipitation in southern Africa is unevenly distributed in space and time, resulting in highly variable river discharge patterns. Severe and prolonged droughts occur and river floods often break these. In South Africa, the semi-arid nature of the climate (leading to greater variability and unpredictability in rainfall) and the increasing demand for freshwater has resulted in the construction of very large storage reservoirs. At present, major dams have a combined storage capacity that exceeds 50% of the total mean annual runoff (Department of Water Affairs 1986). Farm dams account for a further reduction in freshwater flow along river systems.

A major consequence of river impoundments is the reduced frequency and amplitude of flood events (Reddering and Esterhuysen 1987, Whitfield and Bruton 1989). These episodic, high discharge events perform the vital function of scouring and removing non-consolidated marine sediment that accumulates naturally in lower estuarine regions (Reddering 1988). Sand accumulation is caused by flood-tide dominance and is typical of many estuaries along the southeast coast of South Africa. Although some sediment is flushed out on the ebb, the rate of sediment scour by ebb-tidal currents is much less than the rate of flood-tidal sediment input. If dams significantly or totally retain the full flood discharge, this has the effect of artificially protracting the interval between river floods. Under such conditions, disproportionately large flood-tidal deltas develop. This requires a flood of greater erosive capacity to remove the increased volume of sediment (Reddering and Esterhuysen 1987, Reddering 1988, Whitfield and Bruton 1989). Although opening and closing of tidal inlets is a natural occurrence, many estuaries are closing more frequently and for longer periods owing to less effective scouring of inlet channels during floods.

2.3.3. The Kromme River estuary
The Kromme is a good example of an estuary where freshwater inflow from the river is significantly attenuated by large storage reservoirs. The Kromme River is highly regulated and the estuary becomes an extension of the sea most of the time (with respect to salinity). It is therefore not surprising that species composition and distribution as well as biological processes in the estuary differ from other estuarine systems that receive an adequate supply of freshwater.

River and estuary characteristics
The Kromme River rises some 95 kilometres from the sea in the Langkloof between the coastal Tsitsikamma mountains and the Kouga mountain range. The total catchment area is between 936 km2 and 1125 km2, depending on the source reference (Bickerton and Pierce 1988). Quartzite forms the largest part of the geological substrate in the catchment basin. The Geelhoutboom tributary drains the Humansdorp area where the underlying geological formation is Bokkeveld slate that is readily eroded (Reddering and Esterhuysen 1983). As a result, the river carries a relatively high sediment load when it enters the estuary about 8 km from the mouth.

Two major reservoirs are located in the main catchment. The larger Mpofu Dam (construction completed in 1983 with a capacity 100 x 106 m3) is 18 km from the coast and 4 km above the tidal head of the estuary. A second reservoir (Elandsjagt) is located higher up in the catchment. The combined storage capacity of the two reservoirs is 133 x 106 m3, exceeding the MAR (Mean Annual Runoff) of 106 x 106 m3 from the catchment basin (Department of Water Affairs 1986). Rainfall is distributed throughout the year with spring and autumn peaks. In addition, there are numerous farm dams situated on tributaries. Mean annual precipitation in the Kromme catchment varies from 700 to 1200 mm.

Present management policy provides for a total annual freshwater allocation of 2 x 106 m3 for the estuary, unless natural overtopping of the dam occurs. However, overtopping is infrequent, and years may pass between overspill events. The annual allocation for the estuary is therefore <2% of the Mean Annual Runoff from the catchment. Severe drought at the end of the 1980s and early 1990s resulted in the reservoir levels falling below 30% of capacity (Jury and Levey 1993). Freshwater was then released on a monthly basis in order to prevent hypersalinity developing in the upper estuary. During the latter part of the drought and up to the present time, no freshwater was released for environmental purposes. Consequently, river flow below the Mpofu dam became erratic and the estuary received little or no freshwater, except for local runoff after heavy rains. Because of the severe reduction in the natural supply of freshwater, marine conditions now dominate the estuary for extended periods (years). During summer, the upper reaches tend to become hypersaline as its salt content exceeds that of seawater (Reddering and Esterhuysen 1983, Wooldridge and Callahan 2000).

The 14 km long Kromme Estuary (Fig. 9, Table 1) has a constricted but permanently open tidal inlet. In the upper reaches the estuary in narrow and incised into bedrock. Tides are semi diurnal with a small diurnal inequality. Mean spring tide differences outside the inlet is about 1.75 m, while neap tides average 0.57 m. A flood tidal delta extends 5 km from the mouth, but additional sand is derived from an adjacent dunefield. Aperiodic floods of sufficient magnitude scour estuarine channels, but reservoirs in the Kromme catchment dampen or filter out all floods smaller than the 1-in-30 year event (Bickerton and Pierce, 1988). Other physical characteristics of the estuary are given in Table 1.

Table 1.

Physical characteristics of the Kromme estuary. Data from Baird and Ulanowicz (1993), Baird and Heymans (1996) and Bickerton and Pierce (1988). The tidal prism is the volume of water shifted up and down the estuary each full tidal cycle.

Sediments become progressively finer-grained in an upstream direction, mainly because of the decreasing velocity of tidal currents (Reddering and Esterhuysen 1983). The lower reach of the estuary (circa 5 km) is relatively shallow and sandy with well-developed intertidal flats (Fig. 9). This section of the estuary is very dynamic and channels continually change their position. Most of this sand is of marine origin. A further source of sand influx is via the Sand River that enters the estuary approximately 2 km from the mouth (Fig. 9). The Sand River drains an extensive dune field to the southwest and carries sand into the lower estuary during occasional floods.

After stabilization of the coastal dunes in the 1960s, sand moving through the inlet diminished and rate of accumulation inside the estuary decreased. However, with the construction of the two large reservoirs in the catchment, effective removal of accumulated sediment by occasional floods in the estuary also diminished.

Fine muddy sediments in the middle and upper estuary have a fluvial origin [of river origin], although the two large storage reservoirs probably stop most fluvial sediment input via the main tributary. However, sediment input via the Geelhoutboom tributary is not influenced by the reservoirs and fine material continues to be deposited into the estuary. According to Reddering and Esterhuysen (1983), the mixing of marine sand and mud results in a compact mass that is not easily removed by floods.

The position of the erosion base with respect to the bed surface is a good indicator whether sediment is accumulating in an estuary or not (Reddering and Esterhuysen 1983). This erosion base is recognized as a layer of coarse material such as pebbles, shell, bedrock or even highly compacted clay. This layer is covered by finer material at times of sediment accumulation. In the Kromme estuary, the erosion base is not commonly encountered, suggesting that sediment is generally accumulating. However, its extent and significance is not known (Reddering and Esterhuysen 1983).

Physico-chemical characteristics
Numerous studies have shown increased marine influence in summer with hypersaline conditions developing in the upper reaches of the estuary (Hecht 1973, Baird et al. 1981. Marais 1983, Bickerton and Pierce 1988, Wooldridge 1999). Hanekom (1982), recorded salinity values during a three-year study of close to 35 ppt during each summer, with four floods (July-August 1979 and March and June 1981) flushing the estuary on each occasion. In 1979, surface salinity values in the upper reaches were still below 10 ppt after the August flood (Hanekom 1982). In summary, available data indicates that hypersaline conditions were recorded between 1972 and 1983 in the upper reaches of the Kromme estuary and prior to completion of the Mpofu dam (the Elandsjacht reservoir was already present). However, marine dominance has probably become more persistent because of the construction of the second reservoir only 4 km above the estuary. In a study over 36 consecutive months (1989 – 1991), salinity rarely fell below 33 ppt in the upper estuary (Wooldridge and Callahan 2000).

The nutrient status of the Kromme estuary by comparison to other Eastern Cape estuaries is generally described as low (Emmerson et al. 1982). Similar conclusions are drawn by other authors (e.g. Bickerton and Pierce 1988, Baird et al. 1992). However, studies conducted in the marina canals indicate elevated levels of nitrates and phosphates during the summer holiday season when occupation of houses increases significantly (Baird et al. 1992). These authors state that the nutrients are not toxic in the concentrations measured and no indication of anoxia [absence of oxygen] was present during the period of study (1989 -1990). However, there is some concern that effluent perceived to come from septic tanks might become problematical (Bickerton and Pierce 1988), particularly if water exchange between canals and the main estuary decreases. Currently, canals are efficiently flushed (Schumann and de Meillon 1992), although water temperatures were 1-2 oC warmer in the inner canals compared to the main estuary at the time of sampling. Current rate of flushing in the canals leads to marine dominance, although hypersaline conditions have not been recorded to date.

Flora
The phytoplankton community in the Kromme estuary and Geelhoutboom tributary consists of diatoms, dinoflagellates, picoflagellates and euglenoids (Baird et al. 1992). By comparison, the standing stock and production levels in the system expressed as carbon (413 mg C m-2 and 28.3 mg C m-2 day-1 respectively) are much lower when compared to the Sundays or Swartkops estuaries. Low levels recorded were ascribed to lack of freshwater inflow to the estuary and low concentrations of dissolved inorganic nutrients. These conclusions were supported in a later study (Adams et al. 1999) that also commented on the low diversity of species in the system.

No spatial separation of macrophyte communities along the length of the Kromme estuary is apparent, although some variation occurs. This is due to the absence of a salinity gradient that normally structures the species composition as salinity decreases upstream (Adams et al. 1992). For example, Zostera capensis (eelgrass) currently extends into the upper reaches in comparison to its usual association with lower reaches of estuaries. During the period 1983 to 1992, Adams and Talbot (1992) registered a four-fold increase in the standing biomass of eelgrass. This was primarily ascribed to reduced inflow of freshwater, lack of sedimentary disturbances, stable salinity values and reduced turbitities. Talbot et al. (1990) consider flooding and associated sedimentary to be the over-riding forcing function determining the state of submerged macrophytes in small estuaries subject to occasional floods.

Intertidal saltmarsh plants cover about 115 ha in the Kromme estuary (Adams et al. 1999). The largest saltmarsh occurs immediately below the road bridge on the eastern side (see Figure 9). Dominant plants include Spartina maritime, Chenolea diffusa, Sarcocornia decumbens, S. perennis, S. pillansiae and Limnonium linifolium.

Other species of macrophytes recorded in the Kromme include the reed Phragmites australis (present at freshwater seepage points or at the confluence of small streams), Caulerpa filiformis, Chaetomorpha sp., Codium tenue, Gelidium pristoides, Gracilaria verrucosa and Triglochin sp. (Adams et al. 1999, Bickerton and Pierce 1988, Day 1981). In the calm waters of the marina, several species of seaweeds are found.

Fauna
The biomass and diversity of estuarine zooplankton in the Kromme is low (Wooldridge 1999). The estuary has the lowest average copepod biomass (mg dry mass m-3) of six tidal estuaries for which extensive data sets are available, and this is directly linked to the low average longitudinal salinity gradient. Salinity distribution impacts directly on copepod distribution and abundance in estuaries. Reservoir retention of river water artificially maintains euhaline (salinity near that of seawater) conditions throughout the Kromme and a result, spatial and temporal shifts in the natural distribution of resident copepod species is prevented. Some species such as Acartia natalensis disappear from the plankton because of the lack of low salinity habitats.

Anthropogenically induced changes in copepod distribution and abundance will also have a negative ripple effect on the plankton-based foodweb within the estuary. Regulation of the river has therefore deprived the Kromme Estuary of a key mechanism that regulates temporal variability of copepod distribution and abundance.

Although the number of macrobenthic species (animals living in the substrate and retained by a sieve mesh size of 0.5 mm) of local estuaries can exceed 300 (de Villiers et al. 1999), the macrobenthos is typified and dominated by few species only. In the Kromme, Bickerton and Pierce (1988) list 56 species compiled from a number of sources. The sandprawn Callianassa kraussi is one of the most widespread species, attaining densities of over 100 in