IMIS | Flanders Marine Institute

Flanders Marine Institute

Platform for marine research


Publications | Institutes | Persons | Datasets | Projects | Maps
[ report an error in this record ]basket (0): add | show Printer-friendly version

Les récifs coralliens et le lagon de l'île Mayotte (Archipel des Comores, Océan Indien): géomorphologie, sédimentologie, hydrologie, foraminifères
Guilcher, A.; Berthois, L.; Le Calvez, Y.; Battistini, R.; Crosnier, A. (1965). Les récifs coralliens et le lagon de l'île Mayotte (Archipel des Comores, Océan Indien): géomorphologie, sédimentologie, hydrologie, foraminifères. ORSTOM (Office de la Recherche Scientifique et Technique Outre-Mer): Paris. VIII,210 ,16 pl. pp.

Available in  Authors 
    VLIZ: Geology and Geophysics GEO.42 [10328]


Authors  Top 
  • Guilcher, A.
  • Berthois, L.
  • Le Calvez, Y.
  • Battistini, R.
  • Crosnier, A.

    1° Narrative of the expedition. This research was carried out in August and September, 1959. Mayotte, which lies to the North of Mozambique Channel, was selected because this island is surrounded by the finest barrier reef and lagoon in the Indian Ocean, which were never investigated before. The expedition was sponsored by the French National Center for Scientific Research (C. N. R. S.). The vessel was R. V. ORSOM II: she belonged to the Institute for Scientific Research in Madagascar. Before the investigations on land and at sea, two of the writers were able to fly over Mayotte, Glorieuses Islands, and the Geyser Bank, in an aircraft belonging to the French Navy and based at Diego-Suarez, Madagascar, and to take a number of photographs. Samples in the lagoon were obtained with a Mecabolier grab sampler, or with a Charcot dredge in other places; Berthois cones were used where the bottom was muddy. The location of samples is shown on figure 36. Echosounding lines were made across the lagoon, and were used to prepare the geomorphological map (fig. 5). Hydrological stations were occupied, and currents were measured, especially near or in passes across the barrier. Reefs were studied in many places at low spring tides (the tidal range in Mayotte reaches about 4 meters at spring tides); samples on reefs and beaches were collected by hand. 2° Mayotte in the Comoro Archipelago. The Comoro Archipelago (fig. 1) which consists exclusively of volcanic and coral rocks, is tilted from West to East, and the eruptions migrated progressively in the opposite direction. The following islands are successively found: the Great Comoro, an active volcano 2 361 m high, with only scattered fringing reefs; Moheli and Anjouan, two extinct volcanoes 790 and 1 595 m high, with extensive fringing reefs; Mayotte (fig. 2), 660 m high, consisting of a succession of lava flows and two other volcanic eruptions which have now come to an end, with embayed shorelines and a barrier reef (and even two barriers: see further) ; the Geyser Bank and Glorieuses Islands, where the basement is entirely concealed under the corals. This pattern resembles very much what is found in the Central and South Pacific, where the island chains are generally tilted from Southeast (volcanoes) to Northwest (atolls). Although the Geyser Bank and Glorieuses Islands were only seen from the air, a description of them is given since it seems that they were never studied before (photos 4 to 7). 3° General geomorphology of Mayotte lagoon and reefs. The maximum depth in the lagoon seems to be about 46 fathoms below lowest spring tides; 50 fathoms are possible, but not sure. The topography of the lagoon (fig. 5) falls into two types: flat-bottomed areas, and pinnacle areas. The first type is mostly found near Mayotte Island, although in the Northwest it extends comparatively far from it. The second type exists principally in the outer parts of the lagoon, that is, near the barrier. Intermediate areas (submarine plains with scattered pinnacles) are also represented in some places. The dredgings have shown that the pinnacles bear frequently living corals and other animals, or algae, even at depths exceeding 40 m, but the larger part or their surface appears to be now dead. As a rule, the lagoon slopes to the central island, so that the deepest parts of it are found near Mayotte and not near the barrier. But the slope may be uneven, especially in the Southwest where probable fault-scarps, facing the island, were discovered. In the west, the echosoundings point to a drowned karst topography; and, in several places, narrow submerged valleys are cut into the bottom. It is suggested that the karst and the valleys are evidences for a complete emersion or the lagoon during a Pleistocence low sea-level. The general or outer barrier extends over a length or 76 nautical miles. It is continuous only in the Northeast and East, and in the South; in its other sections, the barrier is breached by many passes, sometimes very wide, the most discontinuous section lying in the West. Pamanzi Island, which is embedded in the barrier, came certainly into existence at a time when the barrier was already built, since numerous corals are round among ashes and tuffs in cliffs or the east coast or this island; moreover, the eruptions at Pamanzi and around it are obviously younger than those in other parts or Mayotte (fig. 2). The shapes or the passes differ greatly from one another. Two profiles are given on figure 6. Longogori pass (fig. 28 and photo 12) has a fine meandering shape. The depth of M'Zamhourou pass, in the North, is as large as 47 fathoms, whereas some others are only drowned parts of the barrier, as Choazil pass, 3 or 4 fathoms deep. Mayotte shows a very rare feature in the coral seas: a double barrier in the Southwest (fig. 5 and 35). In this part of the lagoon, an inner barrier extends for 10 nautical miles, the depths between it and the central island exceeding generally 17 fathoms. It is supposed that the inner barrier was initially a fringing reef, which grew up to form a barrier as a consequence of a repetition of the subsidence especially in this part of Mayotte. Evidences for this explanation are the embayed shoreline, more developed in this area than elsewhere around Mayotte, and the probable fault-scarps, located in the same area as the inner harrier. The scarcity of the double harriers in the world, contrasting with the apparent frequency of repeated subsidence, was related by Davis to the fact that, in lagoon waters, the growth of corals is enfeebled by protection from surf by the outer barrier; but here, the exception may be due to the unusually high tidal range ( 4 rn at spring tides), which allows the swell to enter the lagoon at high tide. An important feature is that the outer harrier fades in the Northwest: this fact might be ascribed to a subsidence less pronounced in that area, where the lagoon merges into an extensive bank, less than 50 fathoms deep. Therefore, Mayotte evolution can be summarized as follows. After the two first volcanic eruptions (basaltic, and phonolitic), which built the bulk of the island, fringing reefs grew around it; a subsidence occured, and a barrier reef was formed. After a period of stability, the subsidence went on again in the Southwest, and an inner barrier came into being. The last eruptions poured across the Eastern section of the outer barrier and created Pamanzi Island. During the Pleistocene eustatic oscillations of sea-level, the lagoon was dried up, a karst topography developed in its calcareous deposits, and submarine valleys, ending to gaps (now passes) in the outer harrier, were carved into it. The postglacial transgression drowned the lagoon, in which marine sedimentation processes were allowed to go on again, without obscuring all the former features, owing to lack of time. 4° Climate and oceanography. Mayotte lies in the Southern trade wind belt during the Winter of the Southern Hemisphere, and is affected by the NNW monsoon during the Summer (see tables 1 and 2, frequency of winds per directions, velocities, and months). Hurricanes occur from time to time in the Summer, but ordinary-time winds are moderate. Mean temperatures vary frorn 27.7° C (December) to 23.6° C (July). Diurnal oscillation is 6.0° C. Rainfall does not exceed 40 inches at Dzaoudzi, East coast, But is certainly much higher in the mountains; the larger part of the rain, by far, falls from December to March (table 3). The tidal range, as stated before, is rather high (4 m at spring tides). The currents were measured during this expedition in several places (fig. 7 to 14). They seem to be mostly related to the tide, but many anomalies were ohserved in velocities, in space as weIl as in time. Important factors in anomalies must be the wind and the swell, which can accumulate water over the barrier into the lagoon, and, therefore, increase the velocity of the outcoming currents in passes. The greatest velocities observed during these measurements are 0,92 m/sec. in M'Zambourou pass (incoming current, flood, spring tide), and 1,02 m/sec. in Middle Saziley pass (outcoming current, ebb, neap tide, SE wind bringing water over the barrier). Much stronger currents in passes are possible, although not so strong than across barriers and atoll rims in which passes are not so wide and numerous than they are here. Water transparency in the lagoon is very good everywhere (fig. 15), even when the bottorn is muddy, because the depth is generally too large to allow the waves to stir the sediments. Salinity, temperature and turbidity are given in table 4 for stations inside of the lagoon, and in table 5 for stations outside of the lagoon. The water of the lagoon has the same properties as the upper part of the water in the open sea, a fact which shows that the communication between the lagoon and the ocean is fairly good, as expected from the number and width of the passes. The results of pH measurements are given in table 6; Calcium, Magnesium and Potassium contents are shown in table 7. The distribution of these properties in relation to chlorinity are indicated in figures 21 and 22, and the relations between Ca, Mg and K will be found in table 8 and figure 23. 5° Detailed reef morphology .The outer slope of the barrier into the ocean depths, as shown by echosounding profiles (fig. 24), is very steep from 50 to 400 m, and more or less concave farther down; the lower slope includes large spurs and intervening ravines, which might be old subaerial features, drowned by a combination of volcanic subsidence and eustatism. As many other reefs in the world, the outer barrier bears on its outer edge conspicuous spurs and grooves, rather evenly spaced (table 9 and photos 14,17). Other spurs and grooves, longer and farther apart, are found in shallow passes (table 10 and photo 15) ; and smaller ones were observed in two places on the inner side of the outer barrier, exposed to a comparatively large fetch (photo 16). All of them are an adjustment of the reefs to wave attack, with a combination of coral growth and erosion by particles stirred by waves: modifications in direction of wave approach result in modifications in spur-and-groove pattern (photo 17). Striation on reef-flats, which is common (photo 14), is also related to waves. Contrary to Central and South Pacific, no pink algal ridge exists on Mayotte barriers.On the most continuous parts of the outer barrier, a striking contrast can be seen between the outer and the inner side, the former consisting of spurs and grooves, whereas the latter is full of coral patches growing on a sandy bottom (photos 14 and 18). This is also a consequence of differences in wave action. Faros, or circular reefs enclosing small lagoons inside of the outer barrier, exist in several places. The best faro around Mayotte is Boeni Faro (lig. 25 and photo 19), the others being located in the NNE (lig. 26), near Longogori Pass (lig. 28), in the North (lig. 29), and in the Northwest (lig. 31, upper part). It appears that these faros are related to wave refraction, which recurves the sediments on one or two sides of passes, thus enclosing progressively a water body behind the arcuate reef. Different stages in development are exemplified here, the final stage being the filling up of the lagoon and its complete incorporation into the reef flat. The Mayotte barrier bears no sand cay which would be permanently emerged and able to receive a cover of bushes or coconut trees; it bears only four cays submerged at high tides (or, at least, high spring tides), which are located close to passes, verifying the principle expressed by Steers and by Stoddart (fig. 33 and 31). No beach-rock is found in them. They may be defined as immature cays, contrasting greatly with the large, permanently emerged, and vegetated motus occuring on barriers of the Society Islands in the South Pacific. Fringing reefs around Pamanzi and M'Zambourou Islands are described; the latter are shown on figure 34 and photo 23. Other fringing reefs are described around the mainland, where turbid waters can cause the reefs to take the shape of micro-atolls, to escape frorn sedimentation and burial by terrigenous red clays (photo 25). The inner barrier was investigated in two places (fig. 35). The coral growth is flourishing on it, leading sometimes to enclosures of small lagoons. Small negro-heads scattered on its surface testify that the surf can be very strong here during hurricanes, because of the tidal range as previously said. 6° Lagoon and reef sedimentology. For location of samples, see figure 36. The calcium carbonate content measurements (fig. 37) show that the organic sedimentation by skeletal or coral particles has an overwhelming preponderance: in the median and outer parts of the lagoon, and on reefs, the content is comprised between 98 and 100 per cent. In the lagoon, the area where the content falls below 90 per cent does not generally exceed 5 kilometres in width in the West, and is narrower in the East, North and South. In the Northwest, the lagoon is almost exclusively calcareous in its all parts. Therefore, the influence of land on sedimentation is very small, in spite of deeply weathered soils on basalts which receive heavy rains during the Summer. The highly calcareous sediments consist mostly of sand, with an usually small fraction of gravel (shells and sometimes Halimeda particles). These sandy sediments can be classified into several groups. On sand cays and leeward tails of islands (fig. 38), the sorting is best, as expected, but it decreases in lower parts of the beaches. On beaches along volcanic islands (mainland, and smaller islands), the sorting is still very high, but the sediments are finer (fig. 39). In seaweed meadows (fig. 40), the sorting is considerably poorer (Trask index between 1.49 and 2.73, whereas it ranges from 1.15 to 1.25 on cays and leeward tails ), because the sediments are entangled in roots. The sorting is better on reef-flats (fig. 41), since the particles are more easily moved by waves. All these classes belong to the intertidal zone. In passes (fig. 46 and 47), the results of grain-size analyses are highly variable, probably because the topographic conditions differ greatly from place to place. It cannot be said that the sediments in passes are well sorted, although the currents can be strong. On submarine slopes close to reefs, large differences are also found between individual curves (fig. 48). But these sediments are, as a whole, much finer than those lying on reef-flats (compare figs. 48 and 41). When they include almost exclusively sand without gravel (fig. 49, right, and 50), they must be supplied for a very large part by fishes eating corals in surrounding reefs. The sorting is slightly worse than on reef-flats (Trask index ranging from 1.29 to 2.09). In the lagoon between Mayotte and the outer barrier, the closest resemblances between individual curves are found in the North and Northwest (fig. 52 to 54), where the sorting is again rather good (Trask index between 1.55 and 2.13). On the Northwest bank outside of M'Zambourou, the sediments are much the same with only slight differences (fig. 55). In the South part of the lagoon (fig. 58 to 61), the sediments are generally coarser than in the two preceding submarine groups, in spite of the sheltered position behind a barrier devoid of passes: this is probably due to a difference in supply of sediments, the Halimeda thriving in the Southwest part of the lagoon whereas they are unfrequent in the Northwest and North (fig. 85). In the West, the samples are not well sorted and the curves are different from one another (fig. 62 to 64) ; fine sands and coarse particles occur at the same time in many samples; here again the coarse fraction is related to Halimeda for a large part. AIso in the East and Southeast, large differences are found between curves (fig. 65). In the Northeast (fig. 66 to 69),the differences are even larger, although the shelter by the barrier is effective, and the Trask index is very variable (from 1.30 to 2.51). Behind continuous barriers, many local environments can occur, especially when pinnacles are scattered over the bottom. The sediments which include a fraction finer than 60 microns (silt and clay) have been analysed by densimetry as to their fine fraction. The global curves are given on figures 72 to 79, according to Doeglas' method of figuration. These curves include the sand, silt and clay fractions (gravel was found in only one sample). The distribution of the sediments including a fine fraction (fig. 71) is the same that the distribution of samples in which the calcium carbonate content is not very high: this shows that the fine fraction has a terrigenous origin, and that terrigenous particles do not settle far from the mainland. During the dry season, the very high transparency of water (fig. 15) indicates that practically no terrigenous sediments are carried to the lagoon. The contrary occurs certainly during the rainy season, probably as a result of heavy rains washing the weathering products from the basaltic areas and bringing muddy waters to tbe lagoon where the solid discharge setties quickly. Such a process can explain composite sediments, in which the fine fraction is deposited as it has just been said, while the sandy fraction comes from coral-eating fishes or is carried by rather swift currents. Detailed descriptions of the results of grain-size analyses seem to corroborate this interpretation and lead to distinguish different types of sediments related to local differences in deposition, especially where the bottom is uneven. Ten thermal differential analyses have been made in order to compare fine sediments from the lagoon to soils from the mainland (location on fig. 80). The results are given in table 11 and on figure 81. Mayotte soils does not seem to be typically lateritic, so far as a few analyses can allow a conclusion, but what is most interesting is that the clay fraction in the lagoon and in subaerial soils has much the same composition. Thus, these analyses confirm the conclusions deriving from other methods of investigation. Complete chemical analyses were also made for samples of different kinds (table 12, location on fig. 80). Again appears a distinction between a calcareous, organic group, and a calcaro-siliceous, terrigenous group. Nitrogen and Carbon were also titrated for a number of samples (tables 14 and 15 and fig. 82 and 83). 7° Detailed description of organic and mineral composition of sediments. The larger part, by far, or the samples from the lagoon and the reefs were studied in their constitution, after division into grain-size fractions. Results are given in tables included in chapter 7. The two colums on the right concern the roundness. The distribution of detrital Halimeda is shown on figure 85. 8° Foraminifera. Many samples from the lagoon and the Northwest Bank include a wealth of Foraminifera, which were studied by Yolande Le Calvez. Seven types of location are recognized (outer stations, coastal stations, rugged topography stations, flat topography stations, stations close to the Southwestern fault-scarp, stations near passes, stations in seaweeds or on beaches). A distribution map for the principal types is given (fig. 86). The distribution and frequency of families is described in chapter 8. See also photos 27 to 30. All Foraminifera which are found around Mayotte are already weIl known, except for one species which is described here as Operculina mayottana Le CaJvez. The microfauna collected at Mayotte consists mostly of present-time forms (241) typical of indo-pacific and west-atlantic areas. The others (36) are Foraminifera known so far as fossils, which came to us without noticeable modifications. Thus, Mayotte microfauna includes relict species which are found as fossils in other areas in the world, especially in Europe. This shows that environmental conditions now existing in Mayotte existed probably in Europe in various periods of the Tertiary Era. For example, the conditions of the Lutetian Sea, Paris Basin, still exist now in the Comoro Archipelago. The distribution of Foraminifera in different parts of the lagoon seems to be inversely related to currents. A comparatively quiet water is necessary to allow them to thrive. Such an environment is mostly found in places where numerous coral pinnarles create shelters, whereas, in samples collected near passes, the foraminiferal content is usually poor, owing to swift currents which carry Foraminifera to other more quiet areas.

All data in IMIS is subject to the VLIZ privacy policy Top | Authors