Extracts from my MSc Thesis on Phosphate Pollution in Aqaba, Red Sea

I have only included the abstract, general conclusions, selected tables and images. Any more is available on request as this is the only time it has been published.

OBSERVATIONS ON THE EFFECT OF PHOSPHATE (APATITE) DUST ON THE COMMUNITY STRUCTURE OF A REEF ON THE JORDANIAN COAST OF THE GULF OF AQABA (RED SEA).

Thesis submitted in accordance with the requirements of the University of Liverpool for the degree of Master of Science by Brian Richard Maudsley,

October 1989

Abstract.

During the loading of phosphate (apatite ore) into ships at Aqaba, clouds of dust are produced, much of this falls on the sea and nearby reefs. A sewage treatment plant also discharges in the vicinity. A reef about 0.5km from the dock has been compared with a control reef using random 10m transects parallel to the coast along which all sessile organisms were noted. Urchin surveys were carried out concurrently using quadrats. The study shows that there are several important differences between the two sites, both in abundance and type of organisms, and indicates that the phosphate may be the main cause of the differences. This is probably due to a combination of its forming an insoluble sediment as well as decreasing the N:P ratio, The importance of the main crustose coralline algae (Porolithon onkodesand Lithophyllum kotschyanum) is discussed in relation to reef building and degeneration processes.

GENERAL DISCUSSION AND CONCLUSIONS.

It is evident that the reef flat communities in the polluted (P) and relatively unpolluted (MS) sites are different. The differences detected in this investigation appear to be primarily in relative abundance of organisms but, if more detailed species determinations had been carried out, then greater differences in species composition might also be revealed. The most obvious differences between the environments at the sites are the nearly continuous fall-out of apatite ("phosphate") and the sewage outlet at site P. According to published reports there are high levels of dissolved phosphate in the seawater at this site, but it is not clear from the data whether this enrichment comes from the sewage or from the dust. Most earlier studies have assumed that the phosphate dust is the cause of enrichment and, even though this material is very insoluble in seawater of the pH found in the Gulf of Aqaba, studies have shown that phosphate particles and apatite can dissolve to produce concentrations of phosphate sufficient to support algal growth (Golterman et al 1969, Smith et a! 1978). As high dissolved phosphate levels have been found at the sewage outlet this could have been the main source of enrichment. The nitrogen levels have been reported as being typical of the whole coast, ie not elevated, (Hutings and Abu-Hilal 1983). There is a study in progress by workers at the Marine Station aimed at assessing the relative importance of these sources as the sewage outflow was stopped in June 1987, after the bulk of the present work was carried

out. Other differences between the sites also occur, such as slope of the coastline and, related to that, width of the reef fiat and the presence of pebble filled gullies at the polluted site which penetrated through the reef to the beach. These structural differences could lead to differences in water circulation and strength of wave action which could affect community structure.

Factors affecting back reef community structure

1.Currents and wave motion.

It has been shown that there are definite zones on the reefs of the Gulf of Aqaba (Mergner and Schuhmacher 1974). The reef crest is exposed to more active water movement than the back reef and develops a distinctive community including Millepora dichotoma and excellent growth of Porolithon onkodes. The gullies that penetrate through the polluted reef (P) will allow waves to penetrate as far as the beach and thus the sides of the gullies might be expected to have a reef crest type of community. The present study confirms this, transects crossed by gullies were always closer to the control reef (MS) in community structure. The steep reef front of the polluted area would be expected to cause greater water movement at the polluted site which should affect the whole back reef area to a much greater extent than at the unpolluted site, yet the non-gully back reef transects show a distinctive community quite different from the reef front type.

2. Sediment.

Sediments have often been implicated in changes in community structure causing coral death (Bak 1978) or qualitative changes in coral types, notably promoting a greater proportion of branched rather than massive types (Loya 1976 a,b), as sediment is easier to remove from the former requiring less metabolic energy. The relationship between sediment and growth of algal turf has also been noted (Wanders 1977), each augmenting the other. Sediment will not settle where water movement is high and more or less continuous thus the lower energy back reefs. Lagoons and crevices are the places where sediment accumulation might occur. The settlement tiles do show this pattern as far as the back reef is concerned although the effect was not clear cut and the loss of the tiles from polluted site made a proper analysis difficult. Similarly, the enormous variation in the amounts of sediment on the tiles meant that there was no significant difference between the two sites. There is certainly far more sediment around at the polluted site, this is clear from the lower visibility as well as from the transect data where 5.4% of cover was noted as "sediment" but only 0.74% at the unpolluted site. In addition, the algal turf at P was usually a thick mat growing on sediment. There was a great deal of variation in the cover of tiles from a thick wad of sediment and algal turf (which was never seen at the unpolluted site) to tiles scoured clean, which even the coralline algae had not colonised successfully after three weeks. This variation must be an expression of the mosaic nature of the site with very large differences in currents and herbivore populations over a small area.

The high cover of P.onkodes in reef front and gully transects and the healthy growth of crustose coralline algae on the loose tiles at P and in reef front crevices indicate that they can grow at P (there is neither any evidence that P.onkodes was in a state of degeneration nor simply the remnants of a previous healthy population) despite the high phosphate levels. There is also a slightly higher proportion of branched corals, when compared to massive ones, in the polluted area which also implies sediment pressure (Loya 1976 a,b). Thus it is concluded that sediment has an important effect on community structure in this area.

3. Phosphate and other mineral pollution.

There are elevated levels of dissolved phosphate at the polluted site but no unusual levels of nitrogenous materials have been detected (Hulings and Abu Hilal 1983). High levels of dissolved phosphate have been shown to interfere with coral calcification (Simkiss 1964) and calcification of the articulated coralline alga Corallina ofricinalis (Craven 1982). This effect would impair growth of these organisms but as noted above, corals and coralline algae do grow in the polluted area although coral death (Stylophora pistillata) has been reported from a site about 500m closer to the loading dock (Walker and Ormond 1982), There is a greater growth of algae at the polluted site (which had increased dramatically a year after the present study) including Padinasp. and Cladophora sp. on the reef platform and Enteromorphasp and Ulva lactucain the channel next to the beach. The latter organisms are characteristic of eutrophic conditions (Maragos 1972). Similarly, the algal turf was better developed although there was a discrepancy between the natural turf, which was primarily made up of the blue-green alga Lyngbya aestuarii, and the colonisers of the tiles which were mainly Giffordiasp followed by Herposiphonia sp. a year later. It is quite well established that variations in the N:P ratio can strongly affect the growth of algae in a selective way; in particular an imbalance in favour of phosphorus has been found to favour blue-green algae (Rhee 1978, Nalewajko and Lean 1980). The heavy growth of algal turf at the polluted site would seem to confirm this, indicating a definite nutrient effect. Hulings and Abu-Hital (1983) suggest that phosphate concentrations control the seasonal growth of algae, if this is so then it might be expected that phosphate pollution would disrupt the pattern. It is probable that the thick natural sediments are anaerobic (nearby sand was graphite grey just under the surface - an indicator of anaerobiosis (Sorokin 1973)) and hydrogen sulphide is produced, this might discourage non cyanophytes (Maragos 1972). Thus, the eutrophic conditions certainly have an effect but it is not easy to assess the relative importance of the phosphate effect from a sediment effect in the present study. The two factors probably augment each other and are in turn closely related to the urchin population.

4. Sea Urchins.

There was a decline in numbers of all the major urchin taxa except for Diadema setosum over the study period. The proportions of urchins were different in the two study areas although the actual species present were the same. Urchins and other grazers exert a strong positive effect on the survival of crustose coralline algae by grazing uncalcified competitors (Dart 1972), This was shown very clearly in the present study both from the settlement tile data and from the transplant data. It is likely that the richer growth of Porolithon onkodes and Lithophyllum kotschyanum at the unpolluted site is a result of the large numbers of Diadema setosum, Tripneustes gratilla and Echinometra mathaei.

OnlyTripneustes was more common at the polluted site; the reasons for this and the lower numbers of the other urchin taxa there, are not clear. The richer growth of algae might be expected to support a much greater urchin population, this is true of Tripneustesonly, individuals of which are very much larger here than at the unpolluted site (personal observation). The overall fall in populations indicates that that there are negative factors acting at both sites. Whatever these are, the reduction in numbers could cause the back-reef community to change quite considerably if it continues; the tendency would be towards the reduction of the cover of coralline algae and hard corals and increase of the cover of soft corals and macroalgae (Littler and Littler 1984).

5. Other factors.

The destructive effect of 'catastrophic low tides' upon coral reefs has been noted by Loya (1976) as have the problems of recovery of a back-reef community after widespread coral death caused by these tides under conditions of chronic pollution from phosphate and oil. Phosphate pollution is certainly chronic at the polluted site but oil pollution seems never to occur, although the potential is there with the docks being so close. It is possible that this reef area has been exposed to the low tide/phosphate combination within the last few years (Marine Station records are too recent to help). Given the recovery problem it is likely that degradation of this back-reef area could occur in steps, each very low tide making total recovery less likely and allowing algal growth to build up to the present level. The death of corals and coralline algae does not usually occur through simple overgrowth of the organism by filamentous turf algae. These only settle and grow after damage to the other organism, it is possible that they could then spread but in healthy corals regeneration soon occurs (Walker and Ormond 1982, Loya 1976). Coralline algae can be damaged by filamentous boring algae, which are present in the study area, by desiccation, by smothering with sediment or other organisms and by physical damage from grazing (eg by parrot fish) or mechanical abrasion (see Appendix 10). Another factor that affects the reef communities of this area is the pronounced seasonality. The development of the macro algal species follows this in a very marked way with the maximum growth during the winter. Many of these algae can also grow in the summer but are prevented from doing so by the more active grazers (Mergner and Svoboda 1977), the caged settlement tiles show this very clearly and nitrate is not considered as normally limiting in the area (Hulings and Abu-Hilal 1983). Sedentary reef animals will thus be under greater stress during the winter when metabolic and growth rates will be lower and the danger from overgrowth by macroalgae is greater.

Summary

The composition of the back-reef community in the phosphate polluted study site is the result of several factors the most important of which is probably the continual deposition of insoluble apatite dust causing chronic sedimentation. The process of cleaning this sediment places a metabolic load on the sedentary animals, low winter temperatures with high macroalgal growth and/or extreme low summer tides, can exacerbate the damage. As phosphate is generally considered as being limiting in the Gulf of Aqaba (Weikert 1984) the higher than normal dissolved phosphate but 'typical' nitrate levels encourage turf algae to grow, favouring blue-green algae, and occupy the polluted areas. Regeneration of, or recolonization by, corals or coralline algae is thus made more difficult. A further complication has arisen in that the general reduction of sea urchin populations which occurred during the study period will allow greater algal growth (that this is possible is clearly seen from the caged settlement tiles, the results of which also indicate that nitrate was not limiting at that time). The community structure, which is very heterogeneous in terms of spatial dispersion of organisms compared to the unpolluted reef, and the presence of unstable, pebble filled, gullies in the reef, could indicate that reef degradation is in progress possibly leading to localised destruction.