ABSTRACT
Rocas is the first marine protected area created in Brazil. It is a Biological Reserve and therefore the only human activity allowed there is scientific research. It is an ellipsoid atoll with an internal area of about 7.5 km². Its largest axis (E-W) is 3.7 km long, and the shortest (N-S) is 2.5 km long. An algal ridge limits the reef flat, that is dominated by a coralline algae-vermetid gastropods association growing as small linear ridges. In the reef front (in some grooves), in the pools and in the lagoon, corals (Siderastrea stellata, Montastrea cavernosa and Porites sp) are found. Seismic refraction profiles revealed the presence of two subsurface strata. A 11.6 m long drill core on the western part of the reef, with the recovery rate of 40%, shows that the Holocene sequence of Atol das Rocas was primarily built by coralline algae and, subordinately, by corals, along with some encrusting foraminifer Homotrema rubrum and vermetid gastropods. The reef growth began before 4.8 ky BP with the accretion rate varying from 1.5 to 3.2 m/ky. Subaerialy exposed old reef spits, elevated above tidal range, and a beachrock cliff in one of the cays present in the atoll are evidences of a equal to or higher than present sea level in Rocas, earlier in the Holocene. Low degree of competition for space and low grazing pressure may be the ecological reasons that promoted such a strong growth of coralline algae in Rocas.

INTRODUCTION

    Rocas is a geomorphologic site because it is the only atoll in Southwestern Atlantic, and one of the smallest in the world. It was discovered in 1503 due to the sink of Gonçalo Coelho sailboat (Rodrigues 1940) . Thus, from this very first appearance in the nautical literature, its low lying profile with only two cays and some rocks outcropping from sea had always had the meaning of hazardous place.
    Rocas is a geologic site because as a reef it is a carbonate deposit that results from the building activity response of benthic organisms to environmental parameters such as light availability, hydrodynamics and relative sea level fluctuation.
    Rocas is a paleontological site because it was built dominantly by coralline algae, and only secondarily by corals. This is an important fact because it is generally accepted that coralline algae do not have potential to erect or to be a primary builder of reefs in the Quaternary (Macintyre 1997) .
    At last, but not least, Atol das Rocas is an ecological sanctuary because it is home to a great number of migratory and resident sea birds, which rest, nest and feed there. Among the most abundant birds in the atoll, there are the brown tern, the brown and masked boobies and the frigate birds. Furthermore, like Fernando de Noronha, it is a place of intense nesting activity of green turtles and feeding activity for the hawksbill turtle. Around Atol das Rocas on the top of the seamount platform there is a great quantity of commercial fishes. Mollusks and crustaceans occur in great abundance, as well. Lobsters, as an example, was one of the causes for the intense predatory fishing activity in the atoll waters in the past.
    The purpose of this paper is to present an overview description of Atol das Rocas, based mainly on Kikuchi (1994) and Kikuchi & Leão (1997) . Data on its composition and structure will be presented, showing that the encrusting coralline algae were its primary reef builders. The reasons why coralline algae predominate in such a conspicuous manner is discussed. I also present evidences of a former sea level position in the Holocene equal to or higher than the present sea level. This reinforces the general pattern of the Brazilian coastal sea level curve that exhibits a transgressive period until about 5.1 ky BP and a sea regression since that period. A brief comment on the conservation status of the atoll close this chapter.

ENVIRONMENTAL SETTING

    The atoll is built on the western side of the flat top of a seamount within the Fernando de Noronha Fracture Zone (Figure 1). It is located about 260 km east of the city of Natal, in Northeastern Brazil, and 145 km west of the Fernando de Noronha Archipelago (between 3°45’S and 3°56’S and 33°37’W and 33°56’W). The lighthouse is located at 3 51’30”S and 33 49’29”W.

Figure 1: Location map of Atol das Rocas and the biological reserve limits

    Kikuchi (1994) shows, according to data obtained in Instituto Nacional de Pesquisas Espaciais (INPE), that rainfall is irregularly distributed along the year, with a monthly average of 860 mm, ranging from 183 mm (April/92) to 2663 mm (August/92). In this same period, the air temperature ranged from 17.5°C (April) to 35.8°C (February).
    The data of wind direction indicate that the dominant ESE winds occur along the whole year, with an average frequency of 45% of the measured days. Between June and August (southern hemisphere winter) SE winds occur in 35% of the days and the frequency of E winds is 15% in the same period. Between December and April (southern hemisphere summer) SE winds and E winds occur in about 20% of the days with data available. Winds with speed varying between 6 and 10 m/s dominate along the whole year, but during winter, winds with speed between 11 and 15 m/s are common. Velocities higher than 20 m/s were registered, more frequently, during summer.
    Tides are semi-diurnal, with a maximum range of 3.2 m.
    The drift that characterize Atol das Rocas region is the Southern Equatorial Current, which is originated on the African coast, from the Benguela Current. It flows to west at a velocity ranging from 30 cm/s to 60 cm/s (Richardson & McKee 1984; Silveira et al. 1994) .
    According to Hogben & Lumb (1967) 80% of the waves that were registered in the region where this study site is included, come from E and 15% from NE. Their periods range from 4 s to 7 s and their heights, from 1 m to 2 m. Melo (1993) , however, point out that during December and March this behavior can change, with the occurrence of waves with 15 s to 18 s period and 2 m high waves coming from the northern hemisphere.
    Water temperature in the outside of the atoll averages 27°C, with minimum value of 25.5°C and maximum value of 28°C. In the inner reef, water in the pools can reach 39°C. Available data of salinity, indicate that it averages 37 salinity units (su), varying from 35 su to 42 su. Some data on pH, measured during a couple of days during the summer of 1991, in the inner part of the reef, show values varying from 5 (at night) to 11 (at noon) (Kikuchi 1994) .
    Water visibility during good weather conditions is greater than 20 m. A good visibility is evidenced in the TM/LANDSAT image (Figure 2), where, in the blue channel, it is possible to identify bottom features in depths up to 30 m.

Figure 2: A. TM/LANDSAT satellite image from Rocas atoll, blue band. The scale bar (white) is 1 km long. 
B: Geomorphology map of Atol das Rocas.

HISTORICAL OVERVIEW

    The existence of Atol das Rocas was first registered in the XVI Century Cantino’s Map (Andrade 1959) . The first detailed map of the atoll apeared in 1852, by Leutnant Phillip Lee (Rodrigues 1940) . In that time it was still named Rocas shoal (Baixo das Rocas) or Cabras Shoal (Baixo das Cabras). It was in the bathymetric chart by Comander Vital de Farias in 1858 that Rocas was first described as an atoll (Rodrigues 1940) . Léry (1980) was the first naturalist to give a brief and faint description the atoll. It was only in Andrade (1959) that Rocas received a deeper scientific description. Among other things he described the morphology of the reef flat, the lagoon, of some pools, the old reef spits and of the beachrock that occurs in Cemitério islet. These two latter features were considered indications of a former higher relative sea level in the atoll and correlated with evidences of higher sea level in the Holocene found in the Brazilian coast in Pernambuco State (Andrade 1959) . Thus, Holocene sea level fluctuations were considered of primary importance on the build-up process of Rocas. Due to the low occurrence of corals on the reef flat, Andrade (1959) suggested that Rocas was entirely built by coralline algae (called by the author as Lithothamnium).
    The controversy about the reef's composition and its classification as an atoll began with the paper by Vallaux (1940) . He points out that the reef is composed by coralline algae and that the lagoon would not deserve this name because of its shallow depths. The issue was founded on the dispute that occurred in that time between the theories about the driving mechanism of reef evolution. Darwin's ideas about the 3 successive stages of reef evolution in the Pacific, from fringing reefs to barrier reefs and finally to atolls, all driven by isostasy was jeopardized by Daly who proposed an alternative hypothesis. According to this author, eustatic sea level changes were responsible for carbonate accretion and solution. Thus, high frequency sea level rise and fall along geologic history were considered the cause of reef evolution. The first hypothesis implied that carbonate accretion on atolls would be thick, with deposits at least as old as the early Tertiary, and the lagoons great depths were taken as evidences of this processes. According to the second, the glacial-control theory, reefs in general (and atolls in particular) would be a thin Pleistocene strata since reef substrate would not have changed its position throughout reef development. Later in this century, with the results of borings in the Pacific atolls it was proved that Darwin's ideas were correct as a general model of reef evolution but, at the same time, it was seen that the Quaternary thickness of reefs was generally thin and that eustatic sea level change had an important role in reef development. Thus, the residual blocks of reefs found on the eastern side of the atoll surface and the beachrocks present on one of the atoll cays, that indicate a higher sea-level in the Holocene, are not evidences that falsify the classification of Rocas as an atoll, as pointed out by Andrade (1959) in support to Vallaux's ideas. Based on its geomorphological aspects, it will be shown that Rocas is a true atoll, despite its shallow (6 m deep) but, nevertheless, navigable lagoon.
    The tectonic setting and substrate character of the atoll were considered by Almeida (1955), who states that Rocas atoll and Fernando de Noronha Arquipelago belonged to an alignment of seamounts, which is a branch of the meso-oceanic chain. Bryan et al. (1973) shows evidences of the continuity of this alignment through the continental margin in the Rio Grande do Norte State and into the Brazilian coast and hinterland in the Ceará State. Gorini (1981) confirms the morphology of this sea-mountains alignment and named it as Fernando de Noronha Fracture Zone. According to the last author, this fracture zone is a counterpart of the Jean Charcot Fracture Zone, in the eastern margin of the South Atlantic Ocean. Cordani (1970) obtained ages in Fernando de Noronha rocks spanning from 12 to 1,8 m. y. B.P. He adds, though, that the volcanic activity of the Archipelago might have begun earlier, around 39 m. y. B.P. Thus, sitting to the west of Fernando de Noronha, Rocas' sea-mount must have begun to form earlier. Nevertheless the volcanic activity in the seamount do not necessarily stopped much earlier.
    Andrade (1960) shows that the beachrock grain size is similar to that of the islets' sediment. Coutinho & Morais (1970) study samples collected on the bottom surface of Rocas and Fernando de Noronha and observe that it is “biogenic” carbonate sand, composed mainly by coralline algae (Melobesioidae, Family Coralinaceae), plates of the green calcareous algae Halimeda and benthic foraminifers (mainly Amphistegina radiata e Archaias sp). Tinoco (1972) studies the foraminifers fauna of bottom surface sediments of Rocas and found that Amphistegina radiata and Peneroplis proteus dominates the sediment to depths greater than 45 m but Archaias angulatus is more abundant near the atoll and inside it.

SITE DESCRIPTION

Geomorphology

    Atol das Rocas is a reef that developed on the flat top of a seamount (Figure 1), where depth ranges from 15 m to 30 m. It is an atoll with elliptical shape, opened on its western and northern parts. Its greater axis, oriented E-W, is about 3.7 km long and the minor axis, oriented N-S, is about 2.5 km long (Figure 2). Despite its small dimensions, the reef front, reef flat and a lagoon can be clearly distinguished and subdivided into discrete features, such as reef front, reef flat and lagoon (Figure 2A).
    The reef front has two distinct features: on the windward side (eastern and southeastern side), it is an abrupt, nearly vertical wall, from the reef edge to depths of about 10 m where a talus deposit occurs on its foot down to depths of 15 m. At this level, there is a flat surface colonized by fleshy and coralline alga, corals and sponges, which extends up to 1 km to the east and south of the atoll. In spite of the fact that this surface is colonized dominantly by green and brown algae, with little or no sediment accumulation, colonies of Mussismilia hispida and Millepora alcicornis, together with sponges and rhodoliths are found on it. This is possibly the surface of the substrate of Atol das Rocas. On the lee side a spur-and-groove system develops from the reef edge to depths of 18 m.
    The reef flat is subdivided in two components: the reef proper and a sandy deposit (Figure 2B). The reef proper (reef ring) is 100 to 800 m wide, interrupted occasionally by pools and opened by two channels in the leeward side of the atoll, one on the western side and another at the northern side. These two channels divide the reef proper into the windward arch and the leeward arch.
    In the reef flat there are features like channels, pools and sand cays. The reef proper or the reef ring is the outer reef flat and contour the sandy deposit and the lagoon. The surface of the reef proper is composed of linear ridges of encrusting coralline algae associated with vermetid gastropods, most of which are approximately parallel to the reef edge (Figure 2B). Branching coralline algae (Jania sp., Amphiroa sp.), green and brown macroalgae also grow on these linear ridges. Residues of a former higher reef building (Figure 3) appear at the windward arch of the atoll, more frequently on the eastern part of the reef flat, but can be found also on its southwestern part. This feature, named old reef spits or féo in the literature (Battistini et al. 1975) stands up 2 to 3 m above the surface of the reef flat, in Rocas (Figure 3). Notches on the spits feet (Figure 3) indicate that the mean high water springs (MHWS) is about 0.5 m above the reef flat surface. They are composed mainly of encrusting coralline algae. Vermetid gastropods and the foraminifer Homotrema rubrum occur as subordinate components. The sandy deposit corresponds to a greater part of what Andrade (1959) calls as “very shallow lagoon”. It is composed mainly by a coralline algae debris, medium to fine sand-grained. More than 50% of the fragments are algal debris, with foraminifers tests and mollusk fragments appearing subordinated (the mean frequency reaches about 10% each). In some places there appear discontinuous and asymmetric ripples, produced by tidal flow on the reef flat.

Figure 3: Photography of reef flat in the windward arch, E part of the atoll

    On the edge of the windward arch there is a continuous algal ridge, about 20-30 m wide and nearly 0.5 m high. It is exposed above sea level during the low tides. This is higher energy environment of the reef due to the fact that the energy of the dominant wave train is dissipated on the reef edge. The algal ridge is absent from the edge of the leeward arch, in the northwest side or Farol cay.
    A shallow lagoon, with maximum depth of 6 m, is seen on the northeastern side of the reef. It communicates with open sea through the northern channel (Figure 2), called Barreta Grande. This channel is about 100 m wide and almost 100 m long. Its depth varies from 6 m on its inner part to 10 m on its outer side. This channel resembles a set of meandering small channels due to the growth of several pinnacles in it (Figure 4). The corals Montastrea cavernosa, Siderastrea stellata and Porites sp grow on the walls of the pinnacles. The bottom surface of the spaces between the pinnacles is composed of gravelly sand sediment. The flow through this channel is tidally controlled. The western channel on the other hand, is much smaller and short (less than 50 m in each dimension). The flow through this channel is constantly to outside of the atoll.

Figure 4: Passage between two pinnacles in Barreta Grande. Depth is about 6 m and bottom surface is made of gravel

    Several pools occur on the reef proper (Figure 2B). They are 3 m deep at maximum, during the low tides and can attain up to 400 m in length. This is the case of the Turtle pool, on the eastern side of the atoll. The borders of the pools develop overhangs and the coalescence of several individual pinnacles can also be seen (Figure 5). This is an evidence of the process that originated the reef ring. In the greater pools, isolated pinnacles can be seen as well. The bottom of the pools are covered with sand.

Figure 5: Pool on the south part of the atoll. Note the flat top of the pinnacles at water surface level on the central part of the photo and the transition to the reef flat.

    There are two sand cays on the western part of the atoll (Figure 2). The southern cay is called Cemitério and has a cross bedded beachrock cliff 1.5 m high, on its northeastern side. The height of this cay is about 2 m above the reef flat level. A lighthouse was built on the northern cay called Farol. This cay is about 3 m higher than the reef flat level and there is not beachrock developed on it.
    Rocas is composed of zones frequently found in the Caribbean atolls despite its reduced dimensions and formation of an semi-closed ring (Figure 2B, Table 1). This latter characteristic is not a common feature found in the described atolls (Stoddart 1965). However, comparing Rocas with other Atlantic atolls, many similarities in the general morphology of the reef can be seen, as shown in Table 1. Examples of this latter are the slope of the fore reef, that can be compared to that found in Alacrán reef (Kornicker and Boyd 1962); the width of the reef proper (ring), that is of the same order found in the Nicaraguan atolls (Milliman 1969); and the thickness of the Holocene section and the physiographic origin, that can be matched with that of Hogsty reef (Milliman 1967).

Table 1: Comparison of morphologic characteristics between Rocas and the other Atlantic atolls

Reef structure and composition

    The results of the seismic survey, allowed the detection of 3 strata. Kikuchi (1997) published a reevaluation of the data presented in Kikuchi (1994), in which the mean velocities of each strata were the following:

v₀ = 0.33m/ms
v₁ = 2.50m/ms
v₂ = 4.70m/ms

from the shallower to the deeper strata.

Thus, the layers thickness are:

z₀ = 1.7m
z₁ = 10.0m

were z₀ and z₁ summed up together are thickness of the Holocene section of the reef. The bedrock (v₂) upper limit depth in the investigated site, thus, has a minimum value of 11.7m (Figure 6).
    The upper two seismic layers of the reef sequence can be identified as only one layer of Holocene reefrock, based on data from the core (Figure 6). The low velocity that defined the upper seismic layer, could represent the occurrence of a cap of reef with water and air in its pores, resulting from the low level of the tide at the moment of the seismic registers. The Holocene section of Rocas atoll was almost completely cored. The ages yielded by coral skeletons confirmed the Holocene age for this section. The core hole drilled to a depth of 11.69m, with a recovery rate of 40%. Encrusting coralline algae was the most abundant organism found in the core samples, forming more than 60% of the recovered rock. This prominent role of coralline algae in the building up of reefs is a common feature of Brazilian reefs, such as seen in Abrolhos and in the northern coast of Bahia. A similar characteristic is observed in reefs of Bermudas (boilers), studied by Ginsburg (1973) , and also in the reefs of Antilles, studied by Adey & Burke (1977) . Coral skeletons appeared only subordinately. They were fragments or small colonies of the species Siderastrea stellata, Favia gravida, Mussismilia hispida, Agaricia sp and Porites sp and made up about 10% of the rock. Vermetid gastropods and the encrusting foraminifer Homotrema rubrum are responsible, each of them, for about 6% of the core and occur associated with the coralline algae.

Figure 6: Schematic W-E profile of Rocas atoll, with interpretation of seismic survey. The core hole is represented on the profile, together with its schematic composition. The ages of the coral skeletons from the core are also plotted

    The pre-Holocene substrate of Rocas can be compared to the volcanic rocks that appear in the Fernando de Noronha Archipelago (ultramafic to intermediate volcanic rocks, according to Almeida, 1955 ). The seismic velocity that characterize this lower layer is of the same order of those from the basalts described in the subsurface of Bikini atoll (Dobrin et al. 1949; Raitt 1954), in Kwajalein and Sylvania guyot (Raitt 1953) e no atol Eniwetak (Raitt 1957). This basal section (Figure 6) may be Tertiary, considering the ages obtained in the volcanic rocks of Fernando de Noronha Archipelago dated in Cordani (1970).

Reef accretion and sea level position

    The ¹⁴C ages obtained from the core are 4.86 ky BP at the depth of 11.2 m, 4.41 ky BP at the depth of 10.5 m, 3.06 ky BP at the depth of 7 m and 0.84 ky BP at the surface (Figure 6, Table 2), and they indicate that the 2.5 m/ms layer corresponds to a Holocene reef sequence. Consequently reef growth, at the cored site, must have begun at about 5 ky BP and grew up to present sea level with an average accretion rate of 2.8 m/ky (from 1.5 m/ky to 3.2 m/ky, Table 3). The age of a skeleton of S. stellata, of 2.02 ky BP, found in life position on one small old reef spit on the southwestern part of the windward arch, 0.5 m above the level of the reef flat surface, indicates that parts of the reef reached the present level of the sea at about 2.0 ky BP. The dates yielded by the beachrock ranging from 1.91 ky BP to 2.83 ky BP (Table 1), are concurrent with the age of the coral on the reef spit, what reinforces the hypothesis that some parts of the reef have already reached the present position of sea level between 3.0 ky BP and 2.0 ky BP.

Table 2: ¹⁴C ages of coral skeletons (Ss = Siderastrea stellata, Fg = Favia gravida) and mollusk shells (mol) from the reef surface, the core samples and the Cemitério island beachrock.

    The oldest ¹⁴C date obtained from the core (4.8 ky BP, Table 2) may not represent the very beginning of reef development in Rocas. There may be a difference in the reef age between the windward arch and the leeward arch, where the core was bored. This difference is suggested by the following: i) the continuos and well developed algal ridge on the windward arch, contrary to what happens in the leeward arch, where the algal ridge is a feeble and discontinuous feature; ii) the presence of reef residues (old reef spits) higher than today’s sea level, on the windward arch (Figure 3); iii) the ages of the skeletons collected on the reef flat, on the old reef spit and in the beachrock indicate a possible difference of about 2.0 ky (Table 1) between the time when the reef surfaced at the windward part and the time when it surfaced at the leeward part. Considering that the reef accreted at the same average rate (2.8 m/ky, Table 3) in the two arches, reef growth should, thus,  have begun earlier on the windward arch, say at about 6.0 ky BP (Figure 6). It developed as an open atoll and the heights of the old reef spits indicate that at 2.0 ky BP the reef should have reached up to 3 m above the today reef flat surface (Table 1). The leeward arch began to develop at about 5.0 ky BP, with an accretion rate increasing from 1.5 m/ky to 3.2 m/ky. Consequently, the reef may have attained its semi-closed shape only recently, after the leeward arch reached its present level, after 1.0 ky BP (Table 2).

Table 3: Reef growth rates, calculated from the ages of the coral skeletons obtained in the core. Depth calculated with reference to reef flat level.

Reef building flora and fauna

    The reef surface is mainly covered by soft algae and an association of living coralline alga and vermetid gastropods. A study of the corallines by Gherardi (1999) indicates the occurrence of the genera Porolithon, Lithophyllum and Sporolithon, and Lithoporella among encrusting coralline algae. Massive corals, such as Siderastrea stellata, Montastrea cavernosa and Porites sp, only occur in protected areas, mainly in the lagoon, within the pools and in some grooves of the reef front. (Echeverría et al. 1997) published a list of cnidarians of Rocas in which, among the reef building corals, Madracis decactis, Agaricia agaricites, Porites astreoides, Porites branneri, Favia gravida, and Mussismilia hispida are cited, besides those species mentioned before. The almost absolute dominance of Siderastrea stellata among the hermatypic coral species is a striking characteristic of Rocas.
    Two aspects about the diversity of reef building organisms must be assessed. First of all, it is the present dominance of coralline algae, which occurred along whole development of the reef. Second, among reef building corals there is a conspicuous dominance of a unique species, Siderastrea stellata.
    Although it is presently recognized that coralline algae has a limited role in reef building (Macintyre 1997) due to ecological and environmental constraints and to their low accretion rate (see Steneck 1986 and references wherin), a 11 m thick of reef rock formed primarily by corallines accreted in a quite rapid average rate at Rocas (2.8 m/ky). Figueiredo (1997) shows that in Abrolhos also coralline algae growth is more elevated than in reef environments elsewhere. Besides high hydrodynamic energy, low inter-specific competition and a low degree of herbivory are conditions that favor coralline algae development (Steneck 1997) . It is possible that all these three conditions are met in Rocas. The few number of coral species and the low cover rate of the reef surface by corals that characterize Rocas, may result in a lower degree of competition for space between corallines and corals than that found in other reef sites, and this may have resulted in the enhancement of the coralline algae development. On the other hand, although the biomass of grazers at Rocas seems to be equivalent to the biomass of grazers in other reef sites in Brazil or in the Caribbean region only one genera of coralline algae grazer parrot fish (Sparisoma) is found at Rocas (Rosa & Moura 1997) . The fishes of the genera Sparisoma have weaker mouth muscles than those of the Scarus genera, the commoner reef grazer, which are a more powerful coralline algae grazer than Sparisoma. This difference in the composition of the fish fauna may have contributed, also, to the enhancement of the coralline algae growth and preservation since the grazing activity is one of the most important ecological constraint to the development of coralline algae (Steneck 1986) .
    Considering the dominance of Siderastrea stellata, Echeverría (1997) suggest that the tolerance to environmental conditions such as high wave energy and to high temperature variations may be the reason for the success of this species in the atoll. The contribution of Geomorphology must be stressed, as well. Due to the reduced dimensions of Rocas, water warmed during the low tides (39°C to 40ºC), may influence the lagoon as well. The lagoon, besides the pools is where corals can grow in profusion in the atoll.

CONSERVATION STATUS

    Atol das Rocas is a biological reserve that belongs to the State of Rio Grande do Norte. It is the first marine protected area established in Brazil, enforced by Federal Act n° 83.549 of July 5^th, 1978. In the National System of Conservation Units (Sistema Nacional de Unidades de Conservação – SNUC), biological reserve is a protected area category created for full conservation of biodiversity. No recreational activity nor resources exploitation is admited within these areas. However, visiting for scientific research and educational purposes is allowed in special cases, with prior allowance of Brazilian Institute of the Environment and of the Natural Renewable Resources (Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis – IBAMA ).
    The Biological Reserve of Atol das Rocas includes not only the reef itself but all seamount top limited by the 1000 m isobath. Its total area sums up to 360 km². Though it was created in 1978, it was not until 1990 that the conservation activities began for real. The first reserve station camping was established under the auspices of Marine Turtle Foundation (Fundação Pró-TAMAR) and of Manatee Project (Projeto Peixe-Boi Marinho – IBAMA).
    It takes generally 26 hours to arrive the atoll, from the city of Natal (Rio Grande do Norte). Research teams composed generally of 8 people shift every 25 days. Each team is composed of 2 park rangers (from IBAMA) and 6 more people (scientists, students and volunteers). Communication with the continent are made with the use of SSB and VHF radio systems.

ACKNOWLEDGMENTS

    I would like to thank the organizers of this book for the invitation and the opportunity to write about a beloved object. Also, I’d like to thank Prof. Dr. Zelinda M. A. N. Leão for opening the doors for me to the study of the carbonate environment. Mr. Gilberto Sales, former chief, and Ms. Zélia Brito, present Chief of Reserva Biológica do Atol das Rocas gave me the opportunity of developing the fieldwork and helped during the field work in Rocas. The TM/LANDSAT image was acquired in Instituto Nacional de Pesquisas Espaciais (INPE).

BIBLIOGRAPHIC REFERENCES 

Adey, W.; Burke, R.B. 1977. Holocene bioherms of Lesser Antilles - Geologic control of development. In: S.H. Frost; WEISS, M.P.; SAUNDERS, J.B. eds.  Reefs and related carbonates - Ecology and Sedimentology.Tulsa, American Association of Petroleum Geologists. p. 67-81.

Almeida, F.F.M. 1955. Geologia e Petrologia do Arquipélago de Fernando de Noronha. Rio de Janeiro, DNPM/DGM. 181 p.

Andrade, G.O. 1959. O recife anular das Rocas (Um registro das recentes variações eustáticas no Atlântico equatorial). Anais da Associação dos Geógrafos Brasileiros, XI: 29-61.

Battistini, R., et al. 1975. Éléments de terminologie récifale Indopacifique. Téthys, 7: 1-111.

Bryan, G.M.; Kumar, N.; Castro, P.J.M. 1973. The north Brazilian ridge and the extension of equatorial fracture zones into the continent. In: CONGRESSO BRASILEIRO DE GEOLOGIA, XXVI, Belém, Anais..., SBG-Norte, p. 133-144.

Cordani, U.G. 1970. Idade do vulcanismo do Atlântico Sul. Boletim do Instituto de Geciências e Astronomia - USP, 1: 9-76.

Coutinho, P.N.; Morais, J.O. 1970. Distribución de los sedimentos en la plataforma continental norte-nordeste del Brasil. In: SYMPOSIUM ON INVENTORY OF RESOURCES OF THE CARIBBEAN SEA AND ADJOINING REGION. Curaçao, 1970. Proceedings..., p. 273-284.

Dobrin, M.B.; Perkins Jr, B.; Snavely, B.L. 1949. Subsurface constitution of Bikini Atoll as indicated by a seismic refraction survey. Bulletin of the Geological Society of America, 60:807-828.

Echeverría, C.A.; Pires, D.O.; Medeiros, M.S.; Castro, C.B. 1997. Cnidarians of the Atol das Rocas. In: INT. CORAL REEF SYM, 8th, Panama, 1996. Proceedings... ISRS. V. 1, p. 443-446.

Figueiredo, M.A.O. 1997. Colonization and growth of crustose coralline algae in Abrolhos, Brazil. In: INT. CORAL REEF SYM, 8th, Panama, 1996. Proceedings... V. 1, p. 689-694.

Gherardi, D.F.M.; Bosence, D.W.J. 1999. Modeling of the ecological succession of encrusting organisms in recent coralline-algal frameworks from Atol das Rocas, Brazil. Palaios, 14: 145-158.

Ginsburg, R.N.; Schroeder, J.H. 1973. Growth and submarine fossilization of algal cup reefs, Bermuda. Sedimentology, 20: 575-614.

Gorini, M.A. 1981. The tectonic fabric of the Equatorial Atlantic and adjoining continental margins: Gulf of Guinea to Northeastern Brazil. In: Asmus, H.E. ed.  Estruturas e tectonismo da margem continental brasileira e suas implicações nos processos sedimentares e na avaliação do potencial de recursos minerais.Rio de Janeiro, PETROBRÁS, CENPES, DINTEP. p. 11-116.

Hogben, N.; Lumb, F.E. 1967. Ocean wave statistics. London, National Physical Lab., Ministry of Technology. 263 p.

Kikuchi, R.K.P. 1994. Geomorfologia, Estratigrafia e Sedimentologia do Atol das Rocas (Rebio-IBAMA/RN). Salvador, 144 p. (Dissertação de Mestrado, Pós-Graduação em Geologia, Universidade Federal da Bahia).

Kikuchi, R.K.P.; Leão, Z.M.A.N. 1997. Rocas (Southwestern Equatorial Atlantic, Brazil): an atoll built primarily by coralline algae. In: INT. CORAL REEF SYM, 8th, Panama, 1996. Proceedings... V. 1, p. 731-736.

Kornicker, L.; Boyd, D.W. 1962. Shallow-water geology and environments of Alacran Reef Complex, Campeche Bank, Mexico. Bulletin of the American Association of Petroleum Geologists, 46: 640-673.

Léry, J.d. 1980. Viagem à terra do Brasil. São Paulo, Editora Itatiaia/Ed. Universidade de São Paulo. 303 p.

Macintyre, I.G. 1997. Reevaluating the role of crustose coralline algae in the construction of coral reefs. In: Proc 8th Int Coral Reef Sym. Panamá, 1996, ISRS. V. 1, p. 725-730.

Melo, F., Eloi; Alves, J.H.G.M. 1993. Nota sobre a chegada de ondulações longínquas à costa brasileira. In: Simpósio Brasileiro de Recursos Hídricos, X, Gramado, 1993. Anais..., ABRH, p. 362-369.

Milliman, J.D. 1967. The Geomorphology and history of Hogsty Reef, a Bahamian atoll. Bulletin of Marine Science, 17: 519-543.

Milliman, J.D. 1969. Four southwestern Caribbean atolls: Courtown Cays, Albuquerque Cays, Roncador Bank and Serrana  Bank. Atoll Research Bulletin, 129:1-26.

Raitt, R.W. 1954. Seismic-refraction studies of Bikini and Kwajalein atolls. Bikini and nearby atolls. Part 3. Geophysics. US Geological Survey Professional Paper, 260S: 507-527.

Raitt, R.W. 1957. Seismic-refraction of Eniwetak atoll. Bikini and nearby atolls. Part 3. Geophysics. US Geological Survey Professional Paper, 260S: 685-698.

Richardson, P.L.; McKee, T.K. 1984. Average Seasonal-Variation of the Atlantic Equatorial Currents From Historical Ship Drifts. Journal of Physical Oceanography, 14: 1226-1238.

Rodrigues, O.A.A. 1940. O Atol da Rocas. Revista Marítma Brasileira, LIX: 1181-1228.

Rosa, R.S.; Moura, R.L. 1997. Visual assessment of reef fish community structure in Atol das Rocas Biological Reserve, off northeastern Brazil. In: INT. CORAL REEF SYM, 8th, Panama, 1996. Proceedings... V. 1, p. 983-986.

Silveira, I.C.A.; Miranda, L.B.; Brown, W.S. 1994. On the origins of the North Brazil Current. Journal of Geophysical Research, 99: 22501-22512.

Steneck, R.S. 1986. The ecology of coralline algal crusts: Convergent Patterns and Adaptive Strategies. Annual Review of Ecology and Systematics, 17: 273-303.

Steneck, R.S. 1997. Crustose corallines, other algal fuctional groups, herbivores and sediments: complex interactions along reef productivity gradients. In: INT. CORAL REEF SYM, 8th, Panama, 1996. Proceedings... V. 1, p. 695-700.

Stoddart, D.R. 1965. The shape of atolls. Marine Geology, 3: 369-383.

Tinoco, I.M. 1972. Foraminíferos dos bancos da costa nordestina, Atol das Rocas e Arquipélago de Fernando de Noronha. Trabalhos do Instituto Oceanográfico da Universidade Federal de Pernambuco, 13: 49-60.

Vallaux, C. 1940. La formation atollienne de Rocas (Brésil). Bulletin de L'Institut Océanographique, 37: 1-8.