Reefs and coral carpets in the northern Red Sea as models for organism-environment feedback in coral communities and its reflection in growth fabrics

Coral framework construction and resultant growth fabrics in response to environmental factors were studied in the northern Red Sea, and the Gulfs of Suez and Aqaba. The dependence of growth fabric types on sea-floor topography, oceanography and the ecology of constituent coral species was investigated. Five types of coral frameworks and their growth fabrics were differentiated: Acropora reef framework (platestone to mixstone facies); Porites reef framework (domestone facies); Porites carpet (columnar pillarstone facies); faviid carpet (mixstone facies); Stylophora carpet (thin pillarstone facies). Two non-framework community types were found: Stylophora-Acropora community and soft coral communities. Reef frameworks and resultant growth fabrics show a clear ecological zonation along depth and hydrodynamic exposure gradients. Coral carpets build a framework lacking a distinct internal zonation since they only grow in areas without pronounced gradients. In the northern Red Sea they show a gradual change with depth from Porites (pillarstone) to faviid (mixstone) dominance. The initiation of frameworks was governed by bottom topography (reefs on steep slopes and highs, coral carpets in flat areas). According to environmental conditions, different coral communities produce different framework and growth fabric types. In step with framework growth the environment is modified. The modified environment in turn modifies the coral communities. Thus an environment-organism-environment feedback loop exists. ~ ~, Coral reefs, both fossil and modem, have long been recognized as systems of intricate environment-organism interaction (Rosen 1975; Longman 1981; Frost 1981; Hopley 1982; Done 1982,1983; Perrin et at. 1995; Riegl & Piller 1997; Wood 1999) that are largely brought about by the corals' ability to build a solid reef structure (Insalaco 1998). The ecological structure of the reef and its imprint in the geological record is strongly dependent on a combination of environment, species availability and ecology, both for initiation and as shaping factors during its growth (Longman 1981; Leinfelder 1997; Guozhong 1998). Once a solid reef structure is established, this in turn modifies the environment. As the reef structure grows and modifies its own environment, constituent coral communities and accretion rates change (Montaggioni & Faure 1997; Smith et at. 1998) creating an organism-environment feedback. The composition of coral communities, which in the fossil record is recorded by its growth fabric (Insalaco 1998), is influenced not only by physico-chemical factors in the water column, but also by basin geometry and bottom topography which govern the availability of space for the development of reefs or non-reef building communities (Hopley 1982; Done 1982; Kleypas 1996; van Woesik & Done 1997; Insalaco 1998). In the northern Red Sea, coral reef development follows mainly tectonically generated topographic highs and the mostly steep continental margin (Strasser et al. 1992; Gvirtzman 1994; Piller & Pervesler 1989). However, in shallow shelf areas, extensive framework-building coral communities exist in addition to reefs (Piller & Pervesler 1989; Riegl & Piller 1997, 1999). We call these communities 'coral carpets' in accordance with Reiss & Hottinger (1984). The systems 'reef' and 'coral carpet' differ both in their ecological and frame-building response to environmental factors, as well as in their influence on the environment and their representation in the geological record (Riegl & Piller 1999). However, they should not be seen as mutually exclusive systems but rather as different stages into which frame-building coral communities can develop according to environmental constraints. In this paper we provide a model for how organismic response to the physical environment translates into different types of reef and non-reef frameworks which in turn change their From: INSALACO, E., SKELTON, P. Wo & PALMER, ToJ. (eds) 2000. Carbonate Platform Systems: components and interactions. Geological Society, London, Special Publications, 178,71-88.0305-8719/00/$15.00@The Geological Society of London 2000. 72 BERNHARD RIEGL & WERNER E. PILLER (leeward, S-facing) reef sides (see windrose on Fig. 1). environments. We examine: (1) the different coral framework types in the northern Red Sea; (2) the ability of coral carpets to build frameworks (3) the expected lithological representation in the fossil record; (4) the interaction of the environment with the frame-building coral communities; (5) the feedback of the organisms to the environment via different frame-building capacity. Material and methods Terminology For the purpose of this study, coral carpets were defined as laterally more or less continuous veneers of coral framework following the existing sea-floor morphology. They do not create a distinct three-dimensionality and are therefore ecologically relatively uniform (Fig. 2). Riegl & Piller (1997) used the term 'coral carpet' in a broader sense for all low-relief coral communities in Safaga Bay, irrespective of frameworkbuilding potential. Riegl & Piller (1999) defined only communities with framework-building potential as carpets. From a geological perspective, coral carpets form biostromes (Cumings 1932; Kershaw 1994). Reefs were defined as distinctly three-dimensional structures producing a stronger ecological differentiation of animal and plant communities than coral carpets. Our definitions are compatible with and expand that of Wainwright (1965) quoted in Stoddart (1969), namely that in '. ..structural coral reefs" corals are actively contributing by skeletal accumulation to the topographic development of the reef.. .'. The definitions of Rosen (1990), which is a condensation from several other definitions '. ..organic framework, raised relief, wave resistance, photic zone restriction and tropical (or warm water) distribution. ..' -and Longman (1981, p. 10) -'.. .biologically influenced buildup of carbonate sediment which affected deposition in adjacent areas. ..and stood topographically higher than surrounding sediments during deposition. ..' also support our claim that carpets are indeed different from reefs, mainly because of the absence of clear relief since they may frequently be at almost the same level of the surrounding sediment (see also Wood (1999): '. ..a discrete carbonate structure. ..that develops topographical relief upon the sea floor. ..'). Fagerstrom (1987, p. 13) remarks that '. ..among Holocene photic and aphotic zone reefs there is an unbroken size gradation from isolated solitary corals, to weakly colonial, to strongly colonial of various sizes, to clusters of large, massive coralla, to coral knolls, knoll reefs, and patch reefs with entire reef systems' where 'reefs rarely occur in isolation. ..they occur in variously spaced clusters; each cluster is commonly called a "reef system" or if the cluster is enormous... it may be called a "reef province". Reefs plus reef systems (provinces) and the adjacent (external) sediments constitute a "reef complex". ..' (Fagerstrom 1987, p. 7). It Study area Coastal and offshore sites, representing most coral habitats available in the northern Red Sea, were investigated in the Gulfs of Aqaba and Suez as well as the Egyptian Red Sea (Fig. 1). The quantitative sampling sites were located in the Straits of Gubal and the Hurghada area. Additionally, qualitative observations were made in the Gulf of Aqaba north to Eilat in Israel, to Ain Sukhna in the Gulf of Suez and in the main Red Sea basin south to Ras Banas (Egypt). Geomorphological features in the northern Red Sea region are mainly controlled by tectonism and salt diapirism (Dullo & Montaggioni 1998). Since the Oligocene, the sedimentary evolution of the Red Sea basin has been tectonically controlled, as evidenced by the orientation of major fault structures and the orientation and shape of the reefs which frequently follow and are determined by such structures. Salt diapirism, which again frequently follows the major tectonic lineaments, is another factor creating highs suitable for extensive coral settlement (Orszag-Sperber et al. 1998). Purser et al. (1998) showed how the tilting of fault blocks influenced carbonate and siliciclastic sedimentation, where fault-line depressions funnel silicilastics while carbonates develop on top of, or on the seaward sides of, structural highs (mostly tilt blocks or diapiric structures). These processes also provide the structural diversification into highs with reefs and moderately deep «50 m) shelves settled by coral carpets as discussed in this paper. Meteorologically, the region enjoys stable conditions which result in stable oceanographic conditions over most of the year. Dominant wind and swell direction for about 80% of the year is from the northwest with an average speed of 10 knots (Roberts 1985). In winter, eastwardtravelling depressions can cause changes in wind direction to SE or E (Edwards 1987). It is therefore possible to talk about mostly exposed (windward, N-facing) and mostly sheltered REEFS AND CORAL CARPETS IN THE RED SEA 73 Fig.!. Location map indicating the study area (Egyptian Red Sea) and names mentioned in the text. Coral carpets are tied to shallow shelf areas. In the Gulf of Suez they replace fringing reefs north of Zafarana. In the Gulf of Aqaba they occupy the sloping seafloor in the fore reef areas. BERNHARD RlEGL & WERNER E. PILLER 74 (B) CORAL CARPET = BIOSTROME