Importance of non-reef and reef habitats for coral reef fish in Tanzania, Western Indian Ocean Monique G.G. Grol MSc – thesis Department of Animal Ecology & Ecophysiology Radboud University Nijmegen, The Netherlands Supervisors Ivan Nagelkerken & Martijn Dorenbosch October 2004 Index Abstract……………………………………………………………………………………………………….2 Introduction…………………………………………………………………………………………………..3 Seagrass beds and mangroves as nurseries for coral reef fishes……………………………3 Ontogenetic migrations…………………………………………………………………………....3 Factors determining suitability of seagrass beds and mangroves as a nursery habitat….…3 Alternative nursery habitats……………………………………………………………………….4 Effect of the presence of nursery habitats on community structure on the coral reef………4 Comparison Caribbean versus Indo-Pacific region…………………………………………….5 Anthropogenic pressure…………………………………………………………………………...5 Methods……………………………………………………………………………………………………....6 Study area…………………………………………………………………………………………..6 Field sampling & study design…………………………………………………………………....8 Data analysis……………………………………………………………………………………….9 Statistical analysis………………………………………………………………………………...10 Results…………………………………………………………………………………………………….…12 Discussion…………………………………………………………………………………………………...19 The use of different habitats by coral reef fish species........................................................19 Effect of the presence of nursery habitats on fish species and communities on the coral reef...................................................................................................................21 Conclusions………………………………………………………………………………………………….22 Acknowledgements…………………………………………………………………………………………23 Literature cited………………………………………………………………………………………………24 Appendix 1…………………………………………………………………………………………………..28 Abstract Seagrass beds and mangroves are distributed in tropical regions all over the world and considered important nurseries for coral reef fish species. Studies in the Caribbean have shown that seagrass beds and mangroves are the most important nurseries for juvenile coral reef fish, and when juveniles become adults they move to the coral reefs. Coral reef fish in the Indo-Pacific region do not show such a consistent pattern. In this region it is not clear to what extent coral reef fish use available habitats in a non-reef – coral reef gradient and what the influence is of the presence of non-reef habitats as a nursery on coral reef fish populations on coral reefs. The present study was carried out on four islands along the coast of Tanzania and on one island in the Comoros archipelago in the western Indian Ocean. Using a point count visual census method during the day, the length, abundance and distribution of 76 commercially important coral reef fish species were studied. In total, 2279 quadrats were surveyed at six different non-reef habitats, on coral reefs adjacent to seagrass beds and mangroves (reef sg-mg), on coral reefs at a greater distance to seagrass beds and mangroves (reef far) and on coral reefs in the Comoros archipelago completely lacking seagrass beds and mangroves (reef Comoros). Results showed that shallow seagrass beds did harbour much higher juvenile fish densities than mangroves, deep seagrass beds, intertidal flats, algal beds and coral reefs. This suggests that shallow seagrass beds are the most important nurseries for juvenile coral reef fish. Furthermore, all 76 observed species were categorised in six species groups: sg residents, sg-reef transients, generalists, reef generalists, reef residents and rare species, based on their juvenile and adult distribution patterns between non-reef and reef habitats. Two species groups, sg-reef transients and reef generalists, showed possible migrations between non-reef and reef habitats, but only sg-reef transients, consisted of two species, showed a possible ontogenetic migration from seagrass beds towards the coral reef adjacent to non-reef habitats. This suggests that the nursery hypothesis affected only two species and all other species can use alternative habitats. On reef Comoros the six species groups were also present, but the number of species found on reef Comoros was much lower than on reef far. On reef far three species groups and on reef Comoros four species groups showed significant higher adult densities on reef sg-mg than on reef far or reef Comoros. This suggests that on the scale of an entire island, non-reef habitats have an impact on species groups. The presence of non-reef habitats influences coral reef fish densities on reef sg-mg. At species level, 22 species of the 50 observed species on reef far and 5 species of the 19 observed species on reef Comoros showed significantly higher adult densities on reef sg-mg than on reef far or reef Comoros. This indicates that these species have an advantage of nearby non-reef habitats as a juvenile habitat, but they can sustain on reef far or reef Comoros. On reef far 28 species and on reef Comoros 14 species showed no effect of the absence of non-reef habitats on adult fish densities on the coral reef. It is suggested that these species are able to use the coral reef as an alternative nursery and do not depend on nonreef habitats. They can possibly spend their entire life cycle on the reef and are to some extent selfsustaining. Introduction Seagrass beds and mangroves as nurseries for coral reef fishes Coral reefs are distributed throughout the world and are often located adjacent to seagrass beds and mangroves. In various parts of the world shallow coastal areas, such as bays, lagoons and estuaries, containing these seagrass beds and mangroves are considered important nurseries for coral reef fish (Pollard 1984; Parrish 1989; Baelde 1990; Jackson et al. 2001; Cocheret de la Morinière et al. 2002 & 2003). These nursery habitats are characterized by a high abundance of juveniles of several coral reef fish species, while adult densities are very low (Austin 1971; Blaber & Blaber 1980; Parrish 1989; Beck et al. 2001; Laegdsgaard & Johnson 2001; Nagelkerken et al. 2000a,b & 2001; Adams & Ebersole 2002; Cocheret de la Morinière et al. 2002; Nagelkerken & van der Velde 2002 & 2003). Besides the high abundance of juvenile fish in seagrass beds and mangroves, Nagelkerken et al. (2000a,b & 2001) and Nagelkerken & van der Velde (2003) also suggested a high dependence on these habitats. According to Beck et al. (2001) a habitat functions as a nursery for a species if its contribution per unit area to the production of individuals that recruit to the adult population is larger, on average, than the production of other habitats in which juveniles occur. Some studies mention seagrass beds or mangroves as the most important nursery habitat, because they contain a large abundance of juvenile fish (Jackson et al. 2001, Clynick & Chapman 2002; Nagelkerken & van der Velde 2002), but most coral reef fish species can also occur in other habitat types. The number of studies comparing different nursery habitats is limited (Beck et al. 2001; Gillanders et al. 2003; Heck et al. 2003). So, there are discussions about the importance of seagrass beds and mangroves as a juvenile habitat. Ontogenetic migrations Pelagic fish larvae settle into seagrass beds and mangroves, and grow from juveniles to (sub) adults. Finally they become too large for optimal protection by seagrasses and mangroves, and migrate permanently from their nursery to their adult habitat: adjacent coral reefs or coral reefs at some distance away (Weinstein & Heck 1979; Schulman & Ogden 1987; Parrish 1989; Appeldoorn et al. 1997; Dahlgren & Eggleston 2000; Nagelkerken et al. 2000a,b & 2001; Cocheret de la Morinière et al. 2002; Nagelkerken & van der Velde 2003). Adult habitats are considered as areas where juveniles are not found or occur in very low densities (Gilliam & Fraser 1987; Dahlgren & Eggleston 2000). Although a number of studies have shown connectivity between juvenile and adult habitats, very few have determined the relative contribution of different juvenile habitats to the adult population (Gillanders 2002). Factors determining suitability of seagrass beds and mangroves a nursery habitat Several hypotheses have been proposed to explain the high abundance of juvenile fish in seagrass beds and mangroves: 1. these habitats provide a high abundance of food (Austin 1971; Lenanton 1977; Weinstein & Heck 1979; Blaber & Blaber 1980; Robertson & Duke 1987; Parrish 1989; Beck et al. 2001; Kathiresan & Bingham 2001; Laegdsgaard & Johnson 2001; Heck et al. 2003); 2. the structural complexity of these habitats and the relatively turbid water provide shelter and reduce predation risk (Austin 1971; Lenanton 1977; Weinstein & Heck 1979; Blaber & Blaber 1980; Lenanton 1982; Lenanton & Potter 1987; Robertson & Duke 1987; Parrish 1989; Beck et al. 2001; Kathiresan & Bingham 2001; Laegdsgaard & Johnson 2001; Nanami & Nishihira 2001; Heck et al. 2003; Nakamura and Sano 2004); 3. these habitats are often located at a distance from the coral reef or from offshore waters and are therefore less frequented by predators (Parrish 1989); 4. these habitats often cover extensive areas and due to their high structural complexity have higher interception rates of planktonic fish larvae than coral reefs (Parrish 1989). Alternative nursery habitats Heck et al. (2003) concluded that one of the most important factors in a nursery habitat is the presence of structure per se, rather than the type of structure. Cocheret de la Morinière et al. (2004) indicated that it is not the complexity or turbidity that reduces the predation pressure in a nursery habitat, but the presence of shade. So, the potential factors that may influence the nursery function could be found in several habitats, and juveniles are not restricted to seagrass beds and/or mangroves as a nursery. Although many coral reef fish species have great benefits from using seagrass beds and mangroves as nurseries, they are not completely ‗dependent‘ on these nursery habitats. Juveniles can also utilize ‗alternative‘ nursery habitats such as intertidal flats, sand flats, coral reefs and algal beds (Lenanton 1982; Nagelkerken et al. 2000b, 2001 & 2002; Clynick & Chapman 2002; Hajisamae 2003; Gillanders et al. 2003; Lazzari 2003). If there were no nursery areas, many species would survive, even though the overall numbers might decline (Lenanton & Potter 1987; Shulman & Ogden 1987; Weerts & Cyrus 2002). Such species would be best regarded as nursery opportunists rather than nursery dependents (Lenanton & Potter 1987; Shulman & Ogden 1987). So, also other habitats may contribute to a coral reef fish population. Although, the relative importance of alternative nurseries for juvenile coral reef fish species remains unclear (Cocheret de la Morinière et al. 2002). Effect of the presence of nursery habitats on community structure on the coral reef The community structure of fish on coral reefs in the Caribbean is strongly influenced by adjacent seagrass beds and mangroves (Mumby et al. 2004). Studies of Nagelkerken et al. (2000a,b & 2001) and Nagelkerken & van der Velde (2003) have shown this influence of adjacent seagrass beds and mangroves in the Caribbean. Here juveniles of at least 17 coral reef fish species were highly associated with bays containing seagrass beds and mangroves as nurseries and were almost completely absent from bays without seagrass beds and mangroves, and are seldom found on the coral reef. Adults of the nursery species are represented on coral reefs adjacent to bays with seagrass beds and mangroves (Nagelkerken et al. 2000a; Cocheret de la Morinière et al. 2002; Nagelkerken & van der Velde 2002). The absence of adjacent bays with seagrass beds and mangroves is related to the absence or low densities of adults of the nursery species on the coral reef (Robertson & Duke 1987; Nagelkerken et al. 2002). Comparison Caribbean versus Indo-Pacific region Studies in the Caribbean showed clear nursery functions of bays, lagoons and estuaries containing seagrass beds and/or mangroves for various coral reef fish species (Austin 1971; Weinstein & Heck 1979; Pollard 1984; Baelde 1990; Adams & Ebersole 2002; Cocheret de la Morinière et al. 2002; Nagelkerken et al. 2000b; Nagelkerken & van der Velde 2002; Mumby et al. 2004). However, in the Indo-Pacific region, the nursery function of these habitats is not as clear as in the Caribbean region. For some regions these seagrass and/or mangrove habitats are important (Jones & Chase 1975 (Guam); Robertson & Duke 1987 (Queensland, Australia); Blaber et al. 1989 (north-east Australia)), whereas in other regions these habitats are mentioned as not important (Blaber & Blaber 1980 (eastern Australia); Quinn & Koijs 1985 (Papua New Guinea); Lenanton & Potter 1987 (temperate western Australia); Thollot & Kulbicki 1988 (New Calendonia); Thollot 1992 (New Caledonia); Clynick & Chapman 2002 (Sydney harbour); Heck et al. 2003 (North America, Canada and Australia); Lazzari et al. 2003 (Maine, USA)). Anthropogenic pressure The large growth of the coastal population in Tanzania has led to the enormous degradation of coastal areas (Gullström et al. 2002; Jiddawi & Ohman 2002). At this moment 30 to 60% of all mangroves have already been lost, and seagrasses are declining in the same rate (Shepherd et al. 1989; Spalding 1998; Francis & Bryceson 2003). Due to increasing anthropogenic disturbances seagrass beds and mangroves disappear and coral reefs are damaged (Francis & Bryceson 2003). Continuous destruction of these habitats and increasing fishing activities of humans will have negative consequences for the ecosystem function and fish populations. In the western Indian Ocean, no studies exist that have investigated the fish community structure on a large scale. Therefore, it is important to obtain information on the nursery function of non-reef habitats in East Africa for coral reef fish. For future conservation and management is it necessary to provide a better insight into the function of seagrass beds and mangroves as a nursery in the Indo-Pacific region, and the function of alternative habitats for commercially important coral reef fish. The present study collected juvenile and adult information on density, abundance and distribution of a selection of 76 commercially important coral reef fish species. Three independent observers using the same survey methodology, a point count visual census method, surveyed nine different habitats on five different islands in the western Indian Ocean. The aim of the present study was to answer the following questions: (1) To what extent do coral reef fish use available non-reef and reef habitats as a juvenile and adult habitat? (2) Does the presence of non-reef habitats influence coral reef fish densities on adjacent coral reefs? (3) Are populations of coral reef fishes that depend on non-reef habitats as a juvenile habitat sustained on coral reefs situated further away of these nursery areas? Methods Study area The present study was carried out on the islands of Mafia, Mbudya, Pemba and Zanzibar along the coast of Tanzania and on the island Grande Comoros in the Comoros archipelago in the western Indian Ocean (Figure 1). The Tanzanian coastline runs approximately in a north-south direction and is 1424 km long (Francis et al. 2002). Mafia, Zanzibar, Mbudya and the mainland coast, are largely formed by raised limestone reef platforms dating from the Pleistocene, and form a low coastal plain (Spalding et al. 2001; Richmond 2002; Muhando & Francis, publishing date unknown). The continental shelf is relatively narrow, 8 -10 km wide, but extends to 40 kilometres around the islands Zanzibar and Mafia. Pemba Island also consists of limestone, but dates back to the Miocene. The 800 m deep Pemba Channel separates Pemba from the mainland continental shelf, until it approaches the coastline, and then begins a dramatic ascent to create sheer walls and drop offs. There are fringing reef systems along much of the mainland coast and the offshore islands, although 2 these are broken by shallow bays and estuaries. The total area of coral reef in Tanzania is 3.580 km (Spalding et al. 2001; Richmond 2002). The Grande Comoros is a volcanic island with rocky shores and cliffs of basalt. It is surrounded by very deep water and has steep walls direct along its coast. The total area of coral reef 2 is 430 km (Richmond 2002). In the study area, nine major habitat types were distinguished (Table 1): mangroves in the bay (Mg), shallow seagrass beds (depth < 5 m) situated in a bay with mangroves (Sg bay), shallow seagrass beds (depth < 5 m) in the bay adjacent to coral reefs (Sg reef), deep seagrass beds (depth > 5 m) on the edge of the bay adjacent to coral reefs (Sg reef deep), an intertidal flat containing patches of small coral, rubble, sand and seagrasses (Int flat), algal beds (Algae), coral reefs adjacent to areas containing seagrass beds and mangroves (Reef sg-mg), coral reefs situated far away from seagrass beds and mangroves (Reef far) and a fringing coral reef on the island Grande Comoros in the Comoros archipelago (Reef Comoros). On the islands of Mbudya and Grande Comoros, only coral reef habitats were sampled, on the islands of Mafia, Pemba and Zanzibar, mangroves, seagrass beds and coral reefs were sampled. Algal beds and intertidal flats were only sampled on Zanzibar. Mangroves were dominated by Sonneratia alba, all seagrass beds that were surveyed were dominated by Thalassodendron ciliatum (> 60% cover) and were sometimes mixed with Enhalus acoroides, Thalassia hemprichii or Cymodocea rotundata. Algal beds consisted out of a mix of the macro algae Turbinaria sp. and Sargassum sp. Coral reefs in front of sg-mg areas consisted predominantly out of patch reefs surrounded by sand, seagrass or algal beds. Coral reefs far consisted out of fringing and patch reefs situated around sandbanks. The Grande Comoros does not have extended areas of sg-mg, while Mafia, Pemba and Zanzibar were characterized by the presence of extended areas of sg-mg. The underwater visibility on coral reefs on Pemba and the Grande Comoros was in general high (> 20 m). The underwater visibility on Mafia, Mbudya and Zanzibar at coral reefs adjacent to sg-mg areas was in general small (4-12 m), while coral reefs situated further away from sg-mg areas were characterized by a larger visibility (10-30 m). Spring tidal range is approximately 3.6 m on the islands along the coast of Tanzania and 3.3 m in the Comoros archipelago (Richmond 2002). Table 1. Number of surveyed quadrats for each of the nine distinguished habitat types on the five sampled islands in the western Indian Ocean and the total number of surveyed quadrats per habitat type and per island. Number of Grande Habitat type Zanzibar Mafia Pemba Mbudya quadrats per Comoros habitat type Mangroves 9 36 37 0 0 82 Seagrass bay 123 103 0 0 0 226 Seagrass reef 121 74 31 0 0 226 Seagrass reef deep 65 0 0 0 0 65 Intertidal flat 262 0 0 0 0 262 Algal beds 125 0 0 0 0 125 Reef sg-mg 325 128 79 44 0 576 Reef far 411 68 101 40 0 620 0 0 0 0 97 97 1441 409 248 84 97 2279 Reef Comoros Number of quadrats per island Figure 1. East Africa and the location of the islands of Mafia, Mbudya, Pemba and Zanzibar along the coast of Tanzania and the island of Grande Comoros in the Comoros archipelago in the western Indian Ocean (1 to 5) and the locations of the nine selected habitat types (mg, sg bay, sg reef, sg reef deep, int flat, algae, reef sg-mg, reef far and reef Comoros). Field sampling & study design Seventy-six coral reef fish species were selected based on their commercial value for fisheries and their ease of identification in underwater visual census. All species belonging to the families Haemulidae (9), Lethrinidae (13), Lutjanidae (9), Monodactylidae (1), Mullidae (9), Nemipteridae (2), Plotosidae (1) and Scaridae (20) were included. Furthermore, two species of Acanthuridae (Acanthurus leucosternon and Naso unicornis), two Chaetodontidae (Chaetodon melannotus and C. auriga), one Gerreidae (Gerres oyena), four large Labridae (Cheilinus fasciatus, C. trilobatus C. undulatus and Cheilio inermis), two Siganidae (Siganus stellatus and S. sutor) and one Zanclidae (Zanclus cornutus) were also included. One species belonging to the family Scaridae (Scarus sp.) could not be identified underwater but was included. This species was small, had a green colour and only occurred in seagrass beds and was most probably a juvenile Calotomus carolinus or Chlorurus sordidus. Three independent observers quantified the selected fish species in all habitat types by underwater visual census using SCUBA gear and a point-count method (Polunin & Roberts 1993). Recent studies have shown that the point count method was more efficient than the belt transect method (Watson & Quinn 1997; Samoilys & Carlos 2002). Underwater visual census (UVC) has been used in many studies assessing coral reef fish communities and has several advantages: the method is quick and non-destructive, inexpensive, it can be used for all selected habitats in this study, it is repeatable over time and comparable to other studies (English et al. 1994). But there are also a few disadvantages using UVC: differences in accuracy and precision in estimation of numbers and sizes of fishes by the observers and fishes may be attracted or scared away by observers (Brock 1982; English et al. 1994). In mangrove habitats snorkelling gear was used instead of SCUBA gear, because of the shallow water level and the presence of mangrove branches. Square quadrats of 8 x 8 m were surveyed in habitats where visibility was > 8 m. In areas with visibility between 5 and 8 m, square quadrats of 5 x 5 m were surveyed. Depending on the quadrat size, a single rope of 5 or 8 m length was used as a reference for quadrat size. After placing the rope in a straight line on the bottom, the observer waited three minutes to minimize fish disturbance. All selected fish species within or passing through the quadrat were counted during ten minutes and their size was estimated. Fishes between 0 and 30 cm were estimated in size classes of 2.5 cm, and fishes larger than 30 cm were estimated in decimetres. During the first seven minutes the observer was on the edge of the quadrat; during the last three minutes the observer moved through the quadrat to search for small fish hiding behind corals or other structures. Care was taken not to count individuals or groups of fishes twice that regularly moved in and out the quadrat. In total, 2279 quadrats were surveyed in the nine different habitat types during the period September 2003 - February 2004. The surveys were conducted between 09:00h and 16:00h during daytime around high and low tide when tidal currents were low. Mangroves, shallow seagrass beds, intertidal flats, macro algae and the shallow reefs were surveyed around high tide. Species identification and quantification were first thoroughly and simultaneously practiced by the three observers, until results between observers were comparable. Estimation of size classes was trained underwater by repeatedly estimating the sizes of fifty pieces of electrical wire of known length tied randomly onto a rope, and representing all size fish classes. Training was continued until differences in size estimation between observers were minimal. Practicing quantification and sizeestimation was repeated regularly during the visual census period. In all habitats, replicate quadrats were randomly selected at each location. Mangroves, seagrass beds, intertidal flat and algal beds were sampled until a location was completely surveyed (when the area of habitat on a site was small) or a representative part was surveyed (when the area of habitat on a site was very large). Because coral reefs varied in depth zonations, quadrats were surveyed in all depth zones that could be distinguished at the coral reef (0-5 m, 5-10 m, 10-15 m 15-20 m). In Table 1 the number of surveyed quadrats is listed for each habitat per island. Since fish densities are often correlated to the degree of habitat complexity, maximum coral height, coral cover and coral rigidity were measured. Also maximum water depth of each surveyed quadrat was recorded (in meters) using a diving computer. Maximum height of corals was visually estimated in decimetres in each quadrat. To estimate coral cover, a quadrat was divided into four quarters. For each quarter coral cover in percentages was estimated separately and was averaged for rd th the whole quadrat afterwards. Coral rigidity of 1/3 of all reef sg-mg quadrats and of 1/4 of all reef far quadrats was measured. On reef Comoros coral rigidity was not measured. The rigidity was measured along 5 m of the single rope referring to the quadrat size, using a heavy 10 m chain. The structure of the corals was preciously followed with the chain. After 5 m the observer marked the place on the chain with a piece of electrical wire. The length of the chain needed to follow the coral structure for 5 m was estimated afterwards (in decimetres) at the research station and was used as a measure for reef rigidity. All measurements were conducted directly after the visual census of the quadrat. Data analysis For data analysis a distinction between juveniles and adults had to be made. Fish of each observed species were classified in size ranges of 0-10 cm, 0-20 cm or 0-30 cm. Based upon maturation size or maximum length obtained from FishBase World Wide Web (Froese & Pauly 2004), a subdivision between juveniles and adults was made. If maturation size of a species was known, the smallest possible size class most close to this maturation size was used to classify the juveniles; all other size classes were classified as adults. When maturation size was unknown, the maximum length rd of the species was used to determine juveniles and adults. Species < 1/3 of the maximum length rd were categorised as juveniles and species > 1/3 of the maximum length were categorised as adults (Nagelkerken & van der Velde 2002). For the complete fish community, counts of juveniles and adults were pooled and a comparison of mean total fish density and mean species richness between the selected habitat types was done. Because quadrat sizes were not always comparable, fish counts of different quadrats were 2 pooled to obtain one standard quadrat size: five quadrats of 5 x 5 m were summed (125 m ) or two 2 2 quadrats of 8 x 8 m were summed (128 m ). This standard quadrat size (average of 125 and 128 m ), was used to compare mean total fish density and mean species richness between habitat types. The 76 observed fish species were classified into species groups based on the distribution of juveniles and adults in the various habitat types, on the islands of Mafia, Pemba and Zanzibar. For each species per island, mean total juvenile and adult fish density was calculated per habitat type. Juvenile and adult fish densities per habitat type were used to classify the observed species into the following species groups: sg residents, sg-reef transients, generalists, reef generalists, reef residents and rare species separately for each island. Table 2 summarizes the criteria of the distribution of juveniles and adults that were used to classify these species groups. The pattern of each species per island was used to classify each species into a species group resembling the distribution pattern of juveniles and adults on the scale of all three islands (Table 3). To investigate the variation between the three islands, numbers of species per categorised species group were summed per island. Rare species were excluded from further analysis because of their absence or low densities on the islands. Statistical analysis For each of the 76 species, linear regressions were used to test the influence of the coral complexity factors maximum depth, maximum coral height, total coral cover and coral rigidity on the adult mean total fish density. The complexity factors were used as independent variables and mean total fish density was used as the dependent variable. To make comparisons between juvenile and adult densities in the selected habitat types for each species group, the mean total fish density calculated for juveniles and adults in each species group (all species pooled within the group) per habitat type was log transformed (log10 (1+x)) to normalize distributions and stabilize variances. The influence of the presence of non-reef habitats on the occurrence of adult coral reef fish on the coral reefs was investigated using General Linear Models (GLM‘s). For each species group (total adult density per species group, all species pooled within the group) and for each species separately, two models were run: in the first model fish density on reefs sg-mg was compared with fish density on reefs far, and in the second model, fish density on reefs sg-mg was compared with fish density on reefs Comoros. For each GLM, fish density was used as the dependent variable and the presence of a non-reef habitat was set as a fixed factor. Fish observations per quadrat were used as replicates. The maximum depth of each quadrat was set as a co-variable because regressions showed that depth might have a possible influence. Depth was log transformed (log10 (1+x)) and fish density was square root transformed (SQRT (½+x)) to normalize distributions and stabilize variances. No comparisons between reef sg-mg and reef Comoros could be made using GLM‘s for sg residents and sg-reef transients, because sg residents were not observed on reef Comoros and densities of sg-reef transients were too low to be tested (resulting in non-equal variances). For these species groups, a non-parametric Mann Whitney U-test was used. For this test data were not transformed. GLM‘s could also not be computed for some species separately (resulting in non-equal variances), because of the absence or low densities of these species on the reefs. All statistical tests were done using SPSS version 11.5. Table 2. Criteria used to distinguish species groups on the occurrence of juveniles and adults on the islands Mafia, Pemba and Zanzibar separately. Species group Occurrence of juveniles Occurrence of adults Reef residents > 70% of total density observed in reef habitats > 70% of total density observed in nonreef habitats > 70% of total density observed in nonreef habitats Density in non-reef habitats as well as in reef habitats between 30% and 70% of total density Density in non-reef habitats as well as in reef habitats between 30% and 70% of total density -2 Density is very low (< 0,05/100 m ) and species were absent on one or more islands > 70% of total density observed in reef habitats > 70% of total density observed in nonreef habitats > 70% of total density observed in reef habitats Density in non-reef habitats as well as in reef habitats between 30% and 70% of total density > 70% of total density observed in reef habitats Sg residents Sg—reef transients Generalists Reef generalists Rare species -2 Density is very low (< 0,05/100 m ) and species were absent on one or more islands Results The nine habitat types displayed differences in total fish densities and species richness (Figure 2). The total fish density for juveniles and adults calculated for each quadrat and pooled per habitat type (Figure 2a) showed that juveniles and adults occurred in all nine habitat types. Juveniles showed their highest density in shallow seagrass beds (sg bay and sg reef). Adults were mainly associated with reef habitats, and showed the highest density on reef sg-mg, followed by reef far and sg reef. The total mean fish density and species richness of standardised quadrats showed almost the same pattern as the total mean fish density for juveniles and adults separated (Figure 2b & 2c). In the non-reef habitats, the shallow seagrass beds sg bay and sg reef contained a high fish density and species richness. Of the reef habitats, reef sg-mg and reef far showed higher densities and species richness than reef Comoros, with the highest density and species richness on reef sg-mg. Reef sg-mg showed the highest species richness of all nine habitats, while sg reef showed the highest density of all nine habitats. Linear regressions for 25 species did show significant influences for the factor maximum depth on adult fish densities, and for the factors maximum coral height, total coral cover and coral rigidity only a few showed a significant influence on adult fish density. For these significant coral complexity 2 2 factors the R values were low (R < 0.05). For seven species (Gerres oyena, Lethrinus olivaceus, Monodactylus argenteus, Plectorhinchus gibbosus, P. orientalis, Plotosus lineatus and Scarus sp.) regressions could not be computed. In total, 76 fish species were observed and on each island separately species were categorised in the same six species groups based on the density of juveniles and adults in the nine habitat types. The number of species per species group did show large variation between the three islands and the final classification (Figure 3). On Zanzibar more sg residents and generalists were found compared to the other islands. The number of sg-reef transients found on all islands was low. Variation in the number of generalist species was largest: Mafia (n=11), Pemba (n=4) and Zanzibar (n=20). Reef generalist species showed highest species richness in Mafia. Reef residents showed the highest species richness of all species groups, and little variation was present between islands. Rare species were most abundant on Pemba. a) densities not corrected for quadrat size Mean total fish density + SE (100 m-²) 160 Juveniles Adults 140 120 100 80 60 40 20 0 Mg Sg bay Sg reef Sg reef deep Int flat Algae Reef sg- Reef far Reef mg Comoros c) adult species richness corrected for quadrat size -2 Mean species richness + SE (100 m ) 250 200 150 100 50 0 Mg Sg bay Sg reef Sg reef deep Int flat Algae Reef sgmg Reef far Reef Comoros 18 16 14 12 10 8 6 4 2 0 Mg Sg bay Sg reef Sg reef deep Int flat Algae Reef sg- Reef far Reef mg Comoros Figure 2. (a) Total mean fish density + SE (100 m 2) of juveniles and adults in the nine selected habitat types, not corrected for the quadrat size. (b) Total mean fish density + SE (per 100 m 2) and (c) mean species richness + SE (per 100 m 2) of the complete fish community in the nine selected habitat types, corrected for the quadrat size by summing five 5 x 5 m quadrats or two 8 x 8 m quadrats. 35 Final classification Mafia 30 Pemba Zanzibar 25 Number of species Mean total fish density + SE (100 m-2) b) adult density corrected for quadrat size 20 15 10 5 0 sg resident sg-reef transient generalist reef generalist reef resident rare Figure 3. Number of species per species group for the final classification (+ SE) and the number of species per species group for the islands Mafia, Pemba and Zanzibar. Based on the classification on the scale of all three islands pooled, the same six species groups could be classified. Table 3 shows the mean juvenile and adult fish density for each species within a species group. The first group were sg residents and consisted of five species. Both juveniles and adults of this group showed their highest densities in seagrass beds. Both adult and juvenile densities were low in all other habitats and sg residents did not occur in mangroves and reef Comoros. The second species group were sg-reef transients and consisted of two species for which juveniles showed their highest density in sg bay, while adult densities were higher on reef sg-mg. All other habitats harboured lower densities. A third species group was that of generalists and consisted of 15 species. Juveniles showed highest densities in all habitats, except for the int flat. Adults showed highest densities in all habitats, except for mg and int flat. The fourth classified species group was that of reef generalists and also consisted of 15 species. Juveniles within this group were most abundant in sg bay and sg reef, but were also found on the reef. Adults were most abundant on reef sg-mg. The fifth species group consisted of 26 species and was classified as reef residents. Here, juveniles and adults showed their highest densities on coral reefs. The sixth species group, rare species, consisted of 13 species, of which both juveniles and adults showed very low densities in the nine selected habitat types or only occurred on one of the islands. Figure 4 shows the mean juvenile and adult fish density per species group. Juvenile sg residents were mainly found in seagrass beds, and showed very low densities in the other non-reef habitats and on the coral reefs (Figure 4a). In contrast, juvenile reef residents showed their highest densities on the coral reefs, and the lowest densities in the non-reef habitats. Adults showed almost the same pattern for both species groups (Figure 4b). Adult sg residents were well represented in sg reef and sg reef deep and a lower density in sg bay. Very low densities were found in the other nonreef habitats and on coral reefs. Non-reef habitats showed very low densities of adult reef residents, while coral reefs showed much higher densities. Juvenile sg-reef transients had their highest density in sg bay. Adult sg-reef transients showed their highest density on reef sg-mg. Densities of juvenile and adult generalists and reef generalists were commonly found in all nine habitat types. Juvenile generalists showed the highest densities in all seagrass habitats, while juvenile reef generalists only showed highest densities in sg bay and sg reef. Adult generalists showed their highest density in sg reef, reef sg-mg and reef far, and adult reef generalists on reef sg-mg and reef far. These two species groups also showed higher densities in mg, int flat and algae compared to all other groups. Of the five species groups of the adult coral reef fish, four species groups showed significant differences between reef sg-mg and reef far or reef Comoros (Figure 4b). Sg residents, sg-reef transients and reef generalists showed a significantly higher density on reefs sg-mg compared to reefs far. Sg residents, sg-reef transients, generalists and reef generalists showed a significantly higher density on reefs sg-mg compared to reefs Comoros. Adults of reef residents did not show any significant differences between the densities on the coral reefs. At species level, adults of 29 of the 76 observed species showed a significant difference between reef sg-mg and reef far and 12 adults between reef sg-mg and reef Comoros. On reef sg-mg, a significant higher density was found for 22 species compared to reefs far and for five species compared to reef Comoros. A lower significant density on reef sg-mg was found for seven species compared to reef far and also for seven species compared to reef Comoros. No significant difference between reef sg-mg and reef far was found for 21 species and between reef sg-mg and reef Comoros for seven species. For 26 species, GLM‘s could not be computed between reef sg-mg and reef far and for 57 species between reef sg-mg and reef Comoros because of the absence or low densities of species on the reefs. Depth as a co-variable, showed no significant effect. On reef far in each species group, a higher relative species richness of adults was observed (e.g. all 76 fish species minus the number of fish species that were not observed on the coral reef) compared to reef Comoros (Figure 5). In each species group, more species on reef far showed a significant lower density on reef far than on reef sg-mg compared to the number of species on reef Comoros that showed a significant lower density on reef Comoros than on reef sg-mg. More species within each species group on reef far showed no difference or a higher density on reef far than on reef sg-mg, compared to the number of species on reef Comoros that showed no difference or a higher density on reef Comoros than on reef sg-mg. Mangroves Seagrass bay Seagrass reef Seagrass reef deep Intertidal flat Macro algae a) juveniles 2,2 Mean fish density + SE ( 10log (100 m-2)) 2 1,8 1,6 1,4 Reef sg-mg Reef far Reef Comoros 1,2 1 0,8 0,6 0,4 0,2 0 sg resident 1,8 sg-reef transient generalist reef generalist reef resident b) adults Mean fish density + SE ( 10log (100 m-2)) 1,6 1,4 1,2 AB 1 0,8 B 0,6 0,4 0,2 A AB B AB A B B 0 sg resident B A sg-reef transient generalist reef generalist reef resident Figure 4. Total mean fish density + SE (10log(per 100 m2)) of (a) juveniles and (b) adults of the complete fish community for each species group based on the final classification of the nine selected habitat types. Significant differences of adult fish densities between reef sg-mg, reef far and reef Comoros are indicated with the letters A and B on the bars, where A means a significantly higher density on reef sg-mg than on reef far and B means a significantly higher density on reef sg-mg than on reef Comoros (P < 0.05). 1) not observed 2) low density 3) << than reef sg-mg 4) no difference 5) >> than reef sg-mg Relative species richness of adults (%) . 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% Sg residents Sg-reef transients Generalists Reef genenralists Reef Comoros Reef far Reef Comoros Reef far Reef Comoros Reef far Reef Comoros Reef far Reef Comoros Reef far 0% Reef residents Figure 5. Relative species richness of adults expressed in percentages of the total number of species within each species group for all habitat types, based on the statistical comparison between reef sg-mg and reef far or reef Comoros (P < 0.05). Adult species are classified as: (1) number of species not observed, (2) number of species occur in low densities on the coral reefs and statistical analysis cannot be performed, (3) number of species with a significantly lower density on reef far or reef Comoros than on reef sg-mg, (4) number of species which show no significant differences between the three reefs, and (5) number of species which show a significantly higher density on reef far or reef Comoros than on reef sg-mg. Discussion The use of different habitats by coral reef fish species In most studies, surveys were often limited to one habitat type (Gillanders et al. 2003). In the present study nine different habitats were compared on five different islands in the western Indian Ocean. Pooled densities for each habitat separately showed that juveniles and adults of the 76 observed fish species were common in all habitats. However, shallow seagrass beds contained the highest densities of juvenile coral reef fish, and adults showed their highest densities on reef sg-mg, reef far and sg reef. This suggests that shallow seagrass beds are the most important nursery areas for juveniles, in contrast to other studies where mangroves are considered to be the most important nurseries (Robertson & Duke 1987; Laegdsgaard & Johnson 2001; Clynick & Chapman 2002; Nagelkerken & van der Velde 2003). Following the nursery hypothesis we suspected higher densities of adults on coral reefs than in non-reef habitats, because of migration from nursery habitats to coral reefs. The high adult fish density in sg reef could be explained by the fact that these seagrass beds are adjacent to the coral reef and some fish species utilize seagrass beds as foraging habitats. Studies of Baelde (1990) found large fishes on seagrass beds nearby coral reefs as a result of diurnal feeding migrations from coral reefs to seagrass beds. Mean total fish densities and species richness of standardised quadrats showed the same pattern as the pooled juvenile and adult densities: in non-reef habitats the highest density and species richness in seagrass beds, and in reef-habitats the highest density and species richness on reef sgmg. This suggests that shallow seagrass beds and reef sg-mg were the most important habitats for coral reef fish species. In this study and in studies of Birkeland & Amesbury (1987), a higher number of species was recorded on coral reefs than in seagrass beds and lower densities and species richness were observed on isolated reefs compared to reefs adjacent to seagrass or mangroves. The most likely explanation for this low density and species richness on reef Comoros is the complete absence of non-reef habitats that are utilized as nurseries by several coral reef fish species. For this reason the highest density and species richness could be found on reef sg-mg, where seagrass beds and mangroves are located adjacent. Juveniles and adults of the 76 observed species were classified into six species groups on a scale of all three islands. Juvenile and adult sg residents showed a high preference for seagrass beds, because both juveniles and adults are limited to seagrass beds. These species probably have a high dependence on seagrass habitats. It might be possible that these species shift from shallow seagrass beds deep into the bay to seagrass beds further outside the bay or deeper seagrass beds on the edge of the bay. Juvenile sg-reef transients showed their highest densities on sg bay, and adults on reef sgmg. This may suggest a shift from juvenile sg bay habitats to adult reef sg-mg habitats. For the two species of sg-reef transients, fish densities on reef sg-mg appeared to be a function of the presence of nearby seagrass beds that function as nursery areas. Juvenile generalists and reef generalists were not restricted to non-reef habitats as a nursery and were capable of using the reef as an alternative nursery. In contrast to adult generalists, adult reef generalists did show a preference for coral reefs. According to the juvenile and adult preferences, generalists showed no migration pattern, while reef generalists showed a shift from non-reef habitats and coral reefs to coral reefs, especially to reef sgmg. Both juvenile and adult reef residents showed a preference for coral reefs. Species within this species group probably shift between the coral reefs. For the species group reef residents, the absence of nursery habitats showed no effect on the adult fish densities on the reef. Nevertheless, some species did occur in non-reef habitats, but did not appear to depend on these habitats. Rare species showed low densities and occurred on only one or two islands. These species did not show any preferences for a habitat type or a clear shift between habitats. The different species groups showed that juveniles and adults occurred in all habitat types, and that some species prefer non-reef habitats and others coral reefs. Between these habitats only sg-reef transients and reef residents did show possible ontogenetic migrations. Based on the distribution of juveniles and adults in the various habitat types on each island separately, the 76 observed fish species showed comparable distribution patterns and were categorized in the same six species groups on each island. However, the number of species in each species group on each island showed some variation. On all islands a very low number of sg-reef transients was counted, which use seagrass beds as a nursery area and migrate to the coral reef when they reach adulthood. This indicates that the nursery hypothesis was true for a few of the 76 observed fish species, and that all other species can utilize alternative nurseries. Since juvenile reef generalists do not have a clear habitat preference, and adults prefer coral reefs, the high number of reef generalists in Mafia may be explained by the fact that the surveyed bay in Mafia was very large, and was connected to a continuous coral reef that runs across the length of the island. The bay contained various habitat types adjacent to each other, and fish could easily migrate through the bay to their adult habitat on the reef. Other explanations for the differences in number of species per species group between the islands could be the differences in reef morphology and environmental variables, the site of the island, time of the year or a combination of these factors. In this study, juvenile and adult habitats have been sampled and like many other studies, assumed that once juveniles are no longer found in their habitats, they have moved to adult habitats. Gillanders et al. (2003) have discussed in one of their studies that this does not provide strong evidence that organisms have successfully moved to adult habitats or that juvenile habitats may contribute the most individuals to the adult population. A direct connection between these different habitats or intermediate habitats is difficult to show, and further studies, such as using otolith microchemistry have to confirm the existence of these life-cycle migrations for several coral reef fish. Some studies have correlated total fish densities and/or species richness to several coral complexity factors (Luckhurst & Luckhurst 1978; Birkeland & Amesbury 1987; Parrish 1989; Nagelkerken et al. 2000a). However, in the present study no strong correlations were found. Furthermore, depth showed no significant effect on adult fish densities on reef sg-mg, reef far and reef Comoros per species and per species group. Dorenbosch et al. (2004) have also found no correlations between fish density and coral cover, and Bergman et al. (2000) found that rigidity had no influence on juvenile and adult fish abundance. Other factors such as salinity, tide, water temperature, currents, season and turbidity were not quantified here and it is not clear whether these factors had an influence on adult fish densities or not. Effect of the presence of nursery habitats on fish species and communities on the coral reef Adults sg residents, sg-reef transients and reef generalists showed significantly higher densities on reef sg-mg than on reef far, and adults of sg residents, sg-reef transients, generalists and reef generalists showed significantly higher densities on reef sg-mg than on reef Comoros. These higher densities on reef sg-mg than on reef far or reef Comoros could be caused by the presence of adjacent non-reef habitats. This indicates that juveniles of these four different species groups have an advantage of nearby non-reef habitats. Generalists did not show significant differences between reef sg-mg and reef far, but did show a significant higher density on reef sg-mg than on reef Comoros. This suggests that juveniles of this group can occur on reefs far away from non-reef habitats, but cannot occur on reef Comoros where non-reef habitats are completely lacking. Thus on island scale, non-reef habitats also have an effect on generalists. Adults of reef residents, on the other hand, did not show significant differences in densities between reef sg-mg and reef far or reef Comoros, which suggests that this species group does not depend on non-reef habitats. Juveniles can apparently survive on coral reefs in absence of such nursery areas. Although reef residents had the highest species richness, the high importance of nonreef habitats for fish communities, which include some of the most commercially important coral reef fish species, should not be underestimated. Nagelkerken et al. (2002) compared Caribbean islands with bays containing seagrass beds and mangroves with islands completely lacking these habitats. They found that adults of many nursery species were either absent or present in very low densities on the coral reef of islands lacking bays with seagrass beds and mangroves. In the present study, more species were observed on reef far in each species group than on reef Comoros, and compared to reef Comoros, more species in each species group on reef far showed a significant higher density on reef sg-mg. On reef Comoros more species in each species group were observed in very low densities than on reef far. These differences between reef far and reef Comoros could be explained by the fact that reefs far are situated 20-25 km away from non-reef habitats, and the reef and the non-reef habitats are not separated by very deep water, making some migration of fish possible. This is not possible on reef Comoros, where the nearest bay with non-reef habitats is located at a distance of > 70 km and is separated from the reef by very deep water. Migration here is very unlikely. Non-reef habitats on the Grande Comoros were completely lacking, and species found on reef Comoros must therefore be self-sustaining. Appeldoorn et al. (1997 & 2003) suggested in their studies that adult migrations will be short at oceanic islands with a narrow shelf. Thus, species within species groups that do not occur on reef far or reef Comoros are to some extent depending on non-reef habitats, like sg residents that were not observed on reef Comoros. At species level, only 19 species on reef Comoros could be tested on significance between adult densities on reef sg-mg and reef Comoros, while fifty species could be tested on significance between reef sg-mg and reef far. All species that could not be tested were absent or their adult densities were too low. Out of the 50 species, for 22 species a significant higher adult density was found on reef sg-mg than on reef far and for five species a significant higher adult density was found on reef sg-mg than on reef Comoros. Several other studies (Birkeland & Amesbury 1987; Nagelkerken et al. 2000a; Cocheret de la Morinière et al. 2002; Nagelkerken & van der Velde 2002; Dorenbosch et al. 2004 and Mumby et al. 2004) have also indicated connectivity between nursery habitats and coral reefs for coral reef fish species. Regarding the species with significantly higher densities on reef sg-mg than on reef far or reef Comoros, these species had reduced densities on reefs lacking non-reef habitats, although they did occur there. This indicates that these species highly depend on non-reef habitats as a juvenile habitat, but they can sustain populations on coral reefs situated further away. Species that showed significantly higher adult densities on reef far (n=7) and reef Comoros (n=7) or showed no significant difference between the adult densities on reef sg-mg and reef far (n=21) or reef Comoros (n=7) are most likely not dependent on non-reef habitats. Based on these observations in this study and according to other studies (Thollot & Kulbicki 1988; Thollot et al. 1991; Thollot 1992; Nagelkerken et al. 2000b&c; Dorenbosch et al. 2004), recruitment of some species on the reef could contribute to fish communities on coral reefs. For this reason fish species on reefs lacking non-reef habitats are possibly to some extent self-sustaining and can sustain the fish community on distant coral reefs. Most of the 76 selected fish species in the western Indian Ocean are of high commercial importance for the human coastal population of Tanzania. Anthropogenic impacts on seagrass beds, mangroves and also coral reefs result in the loss of these areas, which have been shown to be very important for many coral reef fish species. In this study, most fish have either a dependence or preference for those nursery habitats and these habitats must therefore be conserved to ensure sustainable fishery in the western Indian Ocean. Conclusions Juveniles and adults of the 76 observed species are present in all non-reef and reef habitats. Shallow seagrass beds showed the highest total mean fish densities, juvenile densities and species richness of the non-reef habitats, while reef sg-mg showed the highest total mean fish density, adult density and species richness of the reef habitats. This suggests that shallow seagrass beds are the most important nurseries of the six surveyed non-reef habitats. The higher density on reef sg-mg suggests that coral reef fish are to a certain extent dependent on non-reef habitats. Coral reef fish probably do show a migration pattern from a seagrass beds to the coral reef. On Mafia, Pemba and Zanzibar the same six species groups were used. These species groups showed that juveniles and adults occur in all habitat types. The different species groups on a scale of all three islands showed that some species prefer non-reef habitats and others coral reefs. Sg residents depend on seagrass beds, in contrast to reef residents, which spend their entire life cycle on coral reefs. Generalists show no preference for either a non-reef or reef habitat. Species within this species group are able to use alternative nurseries. Only the species groups sg-reef transients and reef generalists did show possible ontogenetic migrations. Juveniles of sg-reef transients showed a high dependence on seagrass beds. Adults of sg-reef transients were present on reef sg-mg. This suggests that sg-reef transients migrate from their nursery to nearby coral reefs. Juveniles of reef generalists showed no clear preference for either a non-reef or reef habitat. Adults of reef generalists, however, showed a preference for coral reefs. This suggests that reef generalists may migrate from non-reef and reef habitats towards reef habitats, especially to reef sg-mg. At species-level, the nursery hypothesis was only true for two species of sg-reef transients. All other species could utilize alternative nurseries. Higher adult fish densities on reef sg-mg than on reef far and reef Comoros of species within the species groups sg residents, sg-reef transients and reef generalists appear to be a function of the presence of adjacent non-reef habitats. So, the presence of non-reef habitats has an influence on coral reef fish densities on adjacent coral reefs. Generalists are present on reef far, but cannot sustain populations on reefs completely lacking non-reef habitats. This suggests that non-reef habitats also have an influence on generalists. For reef residents no effect of the absence of non-reef habitats was found for adult fish densities on the coral reefs. Both generalists and reef residents can utilize alternative nurseries. At species level, adult densities of 22 species on reef far and five species on reef Comoros are significantly related to the presence of non-reef habitats. These species showed significantly higher densities on reef sg-mg and reduced densities on reef far or reef Comoros. This indicates that these species have an advantage of nearby non-reef habitats as a juvenile habitat, although they can sustain populations on coral reefs situated further away. For 28 species on reef far and 14 species on reef Comoros, no effect of the absence of nursery habitats was found for adult fish densities on the coral reef. These fish species can possibly spend their entire life cycle on the reef and are to some extent self-sustaining. They can sustain a coral reef fish population in complete absence of nursery habitats. Acknowledgements This study was part of the PhD thesis of drs. M. Dorenbosch at the Radboud University Nijmegen, The Netherlands, and was funded by the Schure-Beijerinck-Popping Foundation, ENVIRONS, PADI Project Aware Foundation and Sensation Divers Zanzibar. Additional funding for students was received from Stichting Nijmeegs Universitair Fonds (SNUF) at the Raboud University Nijmegen, The Netherlands. The staff, personnel and dr. N. Jiddawi of the Institute of Marine Science (IMS) Zanzibar are thanked for the use of their facilities and for their support. Staff, rangers and Jason Rubens of Mafia Marine Park are thanked for the use of a compressor and diving tanks, and their support. The complete crew and Peter J. Minchin of Sensation Divers Zanzibar are thanked for the use of their facilities and their support and help during several weeks of diving at Nungwi. Local fishermen are thanked for bringing us everyday to the right location by boat. Furthermore, I would like to thank dr. I.A. Nagelkerken for supervising me in Holland and drs. M. Dorenbosch for supervising and supporting me during my stay in Tanzania as well as the proceeding work in Holland. Literature cited Abdullah, Ali, Hamad, Ali Said, Ali, Ali Mbarouk & Wild R.G. (2000) Misali Island, Tanzania – An Open Access Resource redefined.Constituting the Commons: Crafting Sustainable Commons in the New Millennium. 8th Biennial Conference of the International Association for the Study of Common Property (IASCP). Adams A.J. & Ebersole J.P. (2002) Use of back-reef and lagoon habitats by coral reef fishes. Marine Ecology Progress Series 228: 213-226 Adams A.J. & Ebersole J.P. (2002) Use of back-reef and lagoon habitats by coral reef fishes. Marine Ecology Progress Series 228: 213-226 Appeldoorn R.S., Friedlander A., Sladek Nowlis J., Usseglio P. & Mitchell-Chui A. (2003) Habitat connectivity in reef fish communities and marine reserve design in old Providence-Santa Catalina, Colombia. Gulf and Caribbean Research 14: 61-77 Appeldoorn R.S., Recksiek C.W., Hill R.L., Pagan F.E. & Dennis G.D. (1997) Marine protected areas and reef fish movements: the role of habitat in controlling ontogenetic migration. Proceedings of the 8th International Coral Reef Symposium 2: 1917-1922 Ault T.R. & Johnson C.R. (1998) Spatial variation in fish species richnees on coral reefs: habitat fragmentation and stochastic structuring processes. Oikos 82: 354-364 Austin H.M. (1971) A survey of the ichtyofauna of the mangroves of western Puerto Rico during December, 1967-August, 1968. Caribbean Journal of Science 11: 27-39 Baelde P. (1990) Differences in the structures of fish assemblages iin Thalassia testudinum beds in Guadeloupe, French west Indies, and their ecological significance. Marine Biology 105: 163-173 Beck M.W., Heck K.L., Jr., Able K.W., Childers D.L., Eggleston D.B., Gillanders B.M., Halpern B., Hays C.G., Hoshino K., Minello T.J. Orth R.J., Sheridan P.F. & Weinstein M.P. (2001) The identification, conservation, and management of estuarine and marine nurseries for fish and invertebrates. BioScience 51: 633-641 Bergman K.C., Ohman M.C. & Svensson S. (2000) Influence of habitat structure on Pomacenturs sulfureus, a western Indian Ocean reef fish. Environmental Biology of Fisheries 59: 243-252 Birkeland C. & Amesbury S.S. (1987) Fish-transect surveys to determine the influences of neighboring habitats on fish community structure in the tropical Pacific. UNEP Regional Seas Reports and Studies 97: 195-202 Blaber S.J.M, Brewer D.T. & Salini J.P. (1989) Species composition an biomasses of fishes in different habitats of a tropical nothern Australian estuary: their occurrence in the adjoining sea and estuarine dependence. Estuarine Coastal and Shelf Science 29: 509-531 Blaber S.J.M. & Blaber T.G. (1980) Factors affecting the distribution of juvenile estuarine and inshore fish. Journal of Fish Biology 17: 143-162 Brock R.E. (1982) A critique of the visual census method for assessing coral reef fish populations. Bulletin of Marine Science 32: 269-276 Clynick B. & Chapman M.G. (2002) Assemblages of small fish in patchy mangrove forests in Sydney Harbour. Marine and Freshwater Research 53: 669-677 Cocheret de la Morinière E.C., Nagelkerken I., van der Meij H. & van der Velde G. (2004) What attracts juvenile coral reef fish to mangroves: habitat complexity or shade? Marine Biology 144: 139-145 Cocheret de la Morinière E.C., Pollux B.J.A., Nagelkerken I. & van der Velde G. (2002) Postsettlement life cycle migration patterns and habitat preference of coral reef fish tha use seagrass a and mangrove habitats as nurseries. Estuarine Coastal and Shelf Science 55: 309-321 Cocheret de la Morinière E.C., Pollux B.J.A., Nagelkerken I., Hemminga M.A., Huiskes A.H.L. & van der Velde G. (2003) Ontogenetic dietary changes of coral reef fishes in the mangrove-seagrass-reef continuum: stable isotopes and gut-content analysis. Marine Ecology Progress Series 246: 279-289 Dahlgren C.P. & Eggleston D.B. (2000) Ecological processes underlying ontogenetic habitat shifts in a coral reef fish. Ecology 81: 2227-2240 Dahodouh-Guebas F., Coppejans E. & van Speybroeck D. (1999) Remote sensing and zonation of seagrasses and algae along the Kenyan coast. Hydrobiologica 400: 63-73 Darwell W.R.T. & Dulvy N.K. (1996) An evaluation of the suitability of non-specialist volunteer researchers for coral reef fish surveys. Mafia Island, Tanzania – a case study. Biological conservation 78: 223-231 Dorenbosch M., van Riel M.C., Nagelkerken I. & van der Velde G (2004) The relationship of reef fish densities to the proximity of mangrove and seagrass nurseries. Estuarine, Coastal and Shelf Science 60: 37-48 English S., Wilkonson C. & Baker V. (1994) Survey manual for tropical marine resources. ASEANAustralia marine science project: living coastal resources. Australian Institute of marine Science, Townsville, pp. 68-80 Francis J. & Bryceson I. (2003) Lessons Learned: Case Studies in Sustainable Use.Chapter 4: Tanzanian Coastal and Marine Resources: Some Examples Illustrating Questions of Sustainable Use. pp 76-102 Francis J., Wagner G.M., Mvungi A., Ngwale J. & Salema R. (2002) GEF MSP Sub-Saharan Africa Project (GF/6010-0016): ―Development and Protection of the Coastal and Marine Environment in Sub-Saharan Africa‖. Tanzania National Report 040302.doc. Phase 1: Integrated Problem Analysis. The content of this document represents the position of the authors and does not necessarily reflect the views or official policies of the Government of Tanzania, ACOPS, IOC/UNESCO or UNEP Froese R. & Pauly D. Editors (2004) FishBase. World Wide Web electronic publication; version March 2004 http://www.fishbase.org Gillanders B.M., Able K.W., Brown J.A., Eggleston D.B. & Sheridan P.F. (2003) Evidence of connectivity between juvenile and adult habitats for mobile marine fauna: an important component of nurseries. Marine Ecology Progress Series 247: 281-295 Gillanders B.M. (2002) Connectivity between juvenile and adult fish populations: do adults remain near their recruitment estuaries? Marine Ecology Progress Series 240: 215-223 Gullström M., de la Torre Castro M., Bandeira S. O., Björk M. Dahlberg M., Kautsky N., Rönnbäck P. & Öhman C. (2002) Seagrass Ecosystems in the Western Indian Ocean. Ambio 31: 588-596 Hajisamae S. & Chou L.M. (2003) Do shallow water habitats of an impacted coastal strait serve as nursery grounds for fish? Estuarine Coastal and Shelf Science 56: 281-290 Heck K.L., Hays G. & Orth R.J. (2003) Critical evaluation of the nursery role hypothesis for seagrass meadows. Marine Ecology Progress Series 253: 123-136 Jackson E.L., Rowden A.A., Attrill M.J., Bossey S.J. & Jones M.B. (2001) The importance of seagrass beds as a habitat for fishery species. Oceanography and Marine Biology 39: 269-303 Jiddawi N. S. & Ohman M.C. (2002) Marine Fisheries in Tanzania. Ambio 31: 518-527 Johnstone R.W., Muhando C.A. & Francis J. (1998) The status of the coral reefs of Zanzibar. One example of a regional predicament. Ambio 27: 700-707 Jones R.S. & Chase J.A. (1975) Community structure and distribution of fishes in an enclosed high island lagoon in Guam. Micronesica 11: 127-148 Kathiresan K. & Bingham B.L. (2001) Biology of Mangroves and Mangrove Ecosystems. Advances in Marine Biology 40: 81-251 Laegdsgaard P. & Johnson C. (2001) Why do juvenile fish utilize mangrove habitats? Journal of Experimental Marine Biology and Ecology 257: 229-253 Lazzari M.A. & Tupper B. (2002) Importance of shallow water habitats for demersal fishes and decapod crustaceans in Penobscot Bay, Maine. Environmental Biology of Fishes 63: 57-66 Lazzari M.A., Sherman S. & Kanwit J.K. (2003) Nursery use of shallow habitats by epibenthic fishes in Maine nearshore waters. Estuarine Coastal and Shelf Science 56: 73-84 Lenanton R.C.J. (1982) Alternative non-estuarine nursery habitats for some commercially and recreationally important fish species of south-western Australia. Australian Journal of Marine and Freshwater Research 33: 881-900 Lenanton R.C.J. & Potter I.C. (1987) Contribution of Estuaries to Commercial Fisheries in Temperate Western Australia and the Concept of Estuarine Dependence. Estuaries 10: 28-35 Luckhurst B.E. & Luckhurst K. (1978) Analysis of the influence of substrate variables on coral reef fish communities. Marine Biology 49: 317-324 Moran S.M., Jenkins G.P., Keough M.J. & Hindell J.S. (2003) Role of physical disturbance in structuring fish assemblages in seagrass beds in Port Phillip Bay, Australia. Marine Ecology Progress Series 251: 127-139 Muhando C.A. & Francis J. (publishing date unknown) The status of coral reefs in the Dar es Salaam Marine Reserves System and the state of reefs in other Marine Protected Areas of Tanzania (http://www.unep.org/eaf/Docs/Coral/DMR%20Report.doc) Mumby P.J., Edwards A.J., Arias-González E., Lindeman K.C., Blackwell P.G., Gall A.,. Gorczynska M.I., Harborne A.R. Pescod C.L., Renken H., Wabnitz C.C.C. & Llewellyn G. (2004) Mangroves enhance the biomass of coral reef fish communities in the Caribbean. Nature 427: 533-536 Nagelkerken I. & van der Velde G. (2002) Do non-Estuarine mangroves harbour higher densities of juvenile fish than adjacent shallow-water and coral reef habitats in Curacao ( Netherland Antilles)? Marine Ecology Progress Series 245: 191-204 Nagelkerken I. & van der Velde G. (2003) Connectivity between coastal habitats of two oceanic Caribbean islands as inferred from ontogenetic shifts by coral reef fishes. Gulf and Caribbean Research 14: 43-59 Nagelkerken I., Dorenbosch M., Verbeek W.C.E.P., Cocheret de la Morinière E. & van der Velde G. (2000a) Importance of shallow-water biotopes of a Caribbean bay for juvenile coral reef fishes: patterns in biotope association, community structure and spatial distribution. Marine Ecology Progress Series 202: 175-192 Nagelkerken I., Kleijnen S., Klop T., van den Brand R.A.C.J., Cocheret de la Morinière E. & van der Velde G. (2001) Dependence of Caribbean reef fishes on mangroves and seagrass beds as nursery habitats: a comparison of fish faunas between bays with and without mangroves/seagrass beds. Marine Ecology Progress Series 214: 225-235 Nagelkerken I., Roberts C.M., van der Velde G., Dorenbosch M., van Riel M.C., Cocheret de la Morinière E. & Nienhuis P. H. (2002) How important are mangroves and seagrass beds for coralreef fish? The nursery hypothesis tested on an island scale. Marine Ecology Progress Series 244: 299-305 Nagelkerken I., van der Velde G., Gorissen M.W., Meijer G.J., van ‗t Hof T. & den Hartog C. (2000b) Importance of mangroves, seagrass beds and the shallow coral reef as a nursery for important coral reef fishes, using a visual census tehcnique. Estuarine, Coastal and Shelf Science 51: 31-44 Nakamura Y., Horinouchi M., Nakai T. & Sano M. (2003) Food habits of fishes in a seagrass bed on a fringing coral reef at Iriomote Island, southern Japan. Ichthyological Research 50: 15-22 Nakamura Y. & Sano M. (2004) Comparison between community structures of fishes in Enhalus acoroides- and Thalassia hemprichii-dominated seagrass beds on fringing coral reefs in the Ryukyu Islands, Japan. Ichthyological Research 51: 38-45 Nanami A. & Nishihira A. (2001) Survival rates of juvenile coral reef fishes differ between pathcy and continuous habitats. Bulletin of Marine Science 69: 1209-1221 Parrish J.D. (1989) Fish communities of interacting shallow-wter habitats in tropical oceanic regions. Marine Ecology Progress Series 58: 143-160 Pollard D.A. (1984) A review of ecological studies on seagrass-fish communities, with particular reference to recent studies in Australia. Aquatic Botany 18: 3-42 Polunin N.V.C. & Roberts C.M. (1993) Greater biomass and value of target coral-reef fishes in two small Caribbean marine reserves. Marine Ecology Progress Series 100: 167-176 Quinn N.J. & Kojis B.J. (1985) Does the presence of coral reefs in proximity to a tropical estuary affect the estuarine fish assemblage? Proceedings of the Fifth International Coral Reef Congress, Tahiti. 5: 445-450 Richmond M.D. (ed.) (2002) A field guide ot the seashores of eastern Africa and the western Indian ocean islands. Sida/SAREC – UDSM pp.461 ISBN 91-586-8783-1 Robertson A.I. & Duke N.C. (1987) Mangroves as nursery sites: comparisons of the abundance and species composition of fish and crustaceans in mangroves and other near shore habitats in tropical Australia. Marine Biology 96: 193-205 Samoilys M.A. & Carlos G. (2000) Determining methods of underwater visual census for estimating the abundance of coral reef fish. Environment Biology of Fishes 57: 289-304 Sano M. (2001) Short-term responses of fishes to macroalgal overgrowth on coral rubble on a degraded reef at Iriomote Island, Japan. Bulletin of Marine Science 68: 543-556 Shulman M.J. & Ogden J.C. (1987) What controls tropical reef fish populations: recruitment or benthic mortality? AN example in the Caribbean reef fish Haemulon flavolineatum. Marine Ecology and Progress Series 39: 233-242 Spalding M.D., Ravilious C. & Green E.P. (2001) Word Atlas of Coral Reefs. Prepared at the UNEP World Conservation Monitoring Centre. University of California Press, Berkeley, USA. Chapter 7: Western Indian Ocean – Tanzania. pp 186-189 Thollot P. (1992) Importance of mangroves for Pacific reef species, myth or reality? Preceedings of the 7th International Coral Reef Symposium, Guam. 2: 934-941 Thollot P. & Kulbicki M. (1988) Overlap between the fish fauna inventories of coral reefs, soft bottoms and mangroves in Saint-Vincent Bay (New Caledonia). Proceedings of the 6th International Coral Reef Symposium, Australia. 2: 613-618 Watson R.A. & Quinn T.J. (1997) Performance of transect and point count underwater visual census methods. Ecological Modelling 104: 103-112 Weerts S.P. & Cyrus D.P. (2002) Occurrence of young and small-sized fishes in different habitats within a subtropical South African estuary and adjacent harbour. Marine and Freshwater Research 53: 447-456 Weinstein M.P. & Heck K.L., Jr. (1979) Ichthyofauna of seagrass meadows along the Caribbean coast of Panama and in the Gulf of Mexico: composition, structure and community ecology. Marine Biology 50: 97-107 Appendix 1. Standard error of mean per species of juvenile and adult densities (ind. per 100 m 2) for all observed fish species, categorised in nine different habitat types according to the text. Fish species are categorised in six groups: sg residents, sgreef transients, generalists, reef generalists, reef residents and rare species, depending on their occurrence as juveniles and adults on the islands of Mafia, Pemba and Zanzibar separately. SEM (100 m-²) Species Sg resident Calotomus spinidens Cheilio inermis Leptoscarus vaigiensis Lethrinus variegatus Scarus sp. Sg-reef transient Cheilinus undulatus Hipposcarus harid Generalist Calotomus carolinus Chlorurus sordidus Gerres oyena Lethrinus harak Lethrinus mahsena Lutjanus argentimaculatus Lutjanus ehrenbergi Mulloidichthys flavolineatus Parupeneus bifasciatus Parupeneus macronema Parupeneus pleurostigma Scarus ghobban Scarus scaber Scolopsis ghanam Siganus sutor Reef generalist Chaetodon auriga Chaetodon melannotus Cheilinus trilobatus Chlorurus strongylocephalus Lethrinus lentjan Lethrinus nebulosus Lethrinus obsoletus Lutjanus fulviflamma Lutjanus gibbus Lutjanus monostigma Parupeneus barberinus Parupeneus rubescens Plectorhinchus flavomaculatus Scarus psittacus Siganus stellatus Juveniles Mg Sg bay Sg reef Sg reef deep Int flat Algae Reef sg-mg Reef far Reef Comoros 0.0 0.0 0.0 0.0 0.0 0.4 0.4 3.7 0.2 1.8 0.3 1.3 2.0 1.9 6.7 0.4 1.4 3.2 0.7 0.0 0.0 0.1 0.1 0.1 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.1 0.1 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 1.1 0.0 0.3 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.6 5.1 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.3 0.2 0.9 0.0 1.9 0.2 0.1 0.0 0.0 0.0 0.5 0.0 3.0 0.1 0.3 2.3 0.3 1.8 0.0 0.3 0.4 0.0 0.0 0.2 0.0 0.9 0.1 1.8 0.0 1.3 1.8 0.1 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.0 0.5 0.1 0.0 0.0 0.0 4.1 0.1 0.2 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.5 0.1 0.4 0.0 0.1 0.2 0.3 0.4 0.0 0.0 0.3 0.0 0.0 0.0 0.0 0.9 0.0 0.1 0.0 0.0 0.6 0.1 0.7 0.0 0.1 0.3 0.0 0.0 0.0 0.0 0.2 0.1 0.2 0.0 0.1 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.5 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.2 0.3 1.0 0.0 0.1 1.2 0.1 0.0 0.7 0.3 0.1 1.1 0.1 0.0 0.1 0.5 0.2 1.5 0.0 0.1 1.4 0.8 0.0 0.7 0.2 0.0 2.3 0.5 0.1 0.0 0.4 0.0 0.3 0.6 0.0 0.0 0.1 0.0 0.3 0.0 0.0 0.5 0.0 0.2 0.1 0.0 0.0 0.1 0.0 0.0 0.7 0.0 0.0 0.2 0.2 0.0 0.4 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.0 0.1 0.1 0.0 0.3 0.0 0.0 0.6 0.0 0.1 0.1 0.2 0.1 0.4 0.1 0.0 1.7 0.7 0.1 0.2 0.1 0.0 0.2 0.1 Adults Mg Sg bay Sg reef Sg reef deep Int flat Algae Reef sg-mg Reef far Reef Comoros 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.5 0.1 0.3 0.5 1.0 1.0 0.5 1.8 0.1 2.0 0.9 4.3 0.0 0.0 0.1 0.1 0.0 0.0 0.1 0.2 0.0 0.0 0.0 0.1 0.1 0.1 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.1 0.0 0.2 0.0 0.1 0.0 0.0 0.2 0.3 0.0 0.0 0.0 0.0 0.0 1.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.5 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.2 0.0 1.2 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.4 0.0 0.0 0.0 0.2 0.0 0.2 0.2 0.1 0.0 0.3 0.1 0.7 0.0 0.6 0.5 0.1 3.1 0.5 0.3 0.1 0.0 0.0 1.4 0.0 1.5 0.2 0.3 0.2 0.8 0.3 0.1 0.2 0.0 0.4 0.3 0.0 0.0 0.0 0.0 1.0 0.4 0.1 0.0 0.1 0.3 0.1 0.5 0.0 0.4 0.0 0.0 0.0 0.5 0.1 0.8 0.1 0.2 0.0 0.4 0.3 0.2 0.9 0.0 0.1 0.0 0.0 0.0 0.1 0.2 1.4 0.0 0.0 0.0 0.3 0.0 0.1 0.2 0.0 0.2 0.0 0.1 0.1 0.3 0.0 0.4 0.1 0.3 0.0 1.0 0.3 0.0 0.3 0.0 0.1 0.0 0.0 0.1 0.8 0.0 0.3 0.1 0.1 0.0 0.6 1.6 0.1 0.3 0.0 0.0 0.0 0.0 0.0 1.3 0.1 0.3 0.1 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.1 0.9 0.3 0.0 1.1 0.2 0.0 0.1 0.0 0.0 0.6 0.0 0.1 0.0 0.0 0.1 0.0 0.0 0.0 0.1 4.6 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.1 0.0 0.0 0.0 0.0 0.4 0.0 0.0 0.2 0.2 0.0 0.0 0.0 0.1 0.1 0.3 0.0 0.0 0.0 0.0 4.9 0.0 0.0 0.2 0.0 0.0 0.2 0.0 0.0 0.0 0.1 0.0 0.0 0.5 0.5 0.0 0.0 0.0 0.9 0.0 0.0 0.1 0.0 0.3 0.1 0.0 0.0 0.1 0.0 0.0 0.5 0.0 0.0 0.3 0.0 0.0 0.5 0.0 0.1 0.1 0.2 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.9 0.0 0.1 0.1 0.1 0.1 0.5 0.3 0.1 4.8 0.1 0.1 0.2 0.0 0.0 0.4 0.1 0.1 0.1 0.1 0.1 0.1 0.0 0.1 2.7 0.3 0.0 0.1 0.0 0.0 0.7 0.0 0.1 0.1 0.0 0.3 0.0 0.2 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.1 0.1 Appendix 1 contiunued SEM (100 m-²) Species Reef resident Acanthurus leucosternon Cetoscarus bicolor Cheilinus fasciatus Chlorurus atrilunula Diagramma pictum Gnathodentex aurolineatus Gymnocranius grandoculus Lethrinus erythracanthus Lethrinus microdon Lutjanus bohar Lutjanus k asmira Lutjanus lutjanus Macolor niger Monotaxis grandoculus Mulloidichthys vanicolensis Naso unicornis Parupeneus cyclostomus Plectorhinchus gaterinus Plotosus lineatus Scarus frenatus Scarus niger Scarus rubroviolaceus Scarus russelli Scarus tricolor Scolopsis bimaculatus Zanclus cornutus Rare Bolbometopon muricatum Chlorurus japanensis Lethrinus olivaceus Lethrinus xanthochilus Monodactylus argenteus Parupeneus indicus Plectorhinchus gibbosus Plectorhinchus orientalis Plectorhinchus picus Plectorhinchus plagiodesmus Plectorhinchus playfairi Plectorhinchus schotaf Scarus viridifucatus Juveniles Mg Sg bay Sg reef Sg reef deep Int flat Algae Reef sg-mg Reef far Reef Comoros 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 1.3 0.1 0.0 0.1 0.1 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.2 0.0 0.0 0.0 0.1 0.1 0.3 0.0 0.2 0.1 0.0 0.1 0.1 0.4 3.0 0.1 0.1 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 1.4 0.0 0.0 0.0 0.0 0.0 0.0 2.3 0.0 0.2 0.1 0.0 0.0 0.1 0.0 0.0 0.1 0.0 0.1 0.1 0.1 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.4 0.0 0.0 0.0 0.1 0.4 0.0 0.1 0.3 0.0 0.1 0.1 0.0 0.0 0.0 0.0 0.6 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Adults Mg Sg bay Sg reef Sg reef deep Int flat Algae Reef sg-mg Reef far Reef Comoros 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.1 0.0 0.1 0.2 0.0 0.0 0.0 0.1 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.2 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.3 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.2 0.1 0.0 0.1 0.1 0.1 0.5 0.0 0.0 0.0 0.0 0.2 4.5 0.1 0.3 0.5 0.0 0.0 0.6 0.0 0.0 0.1 0.0 0.0 0.1 0.2 0.2 0.1 0.0 0.0 0.1 0.1 0.6 0.0 0.0 0.0 0.2 0.7 3.1 0.1 0.1 1.0 0.1 0.0 0.3 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.1 0.5 0.2 0.0 0.0 0.0 4.8 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.7 0.9 0.0 0.3 0.0 0.0 0.0 0.1 1.0 0.0 0.1 0.0 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 1.4 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0