Redalyc.Vertical Zonation in the Rocky Intertidal at Cocos Island (Isla Del
Total Page:16
File Type:pdf, Size:1020Kb
Revista de Biología Tropical ISSN: 0034-7744 [email protected] Universidad de Costa Rica Costa Rica Sibaja-Cordero, Jeffrey Alejandro Vertical zonation in the rocky intertidal at Cocos Island (Isla del Coco), Costa Rica: A comparison with other tropical locations Revista de Biología Tropical, vol. 56, núm. 2, agosto, 2008, pp. 171-187 Universidad de Costa Rica San Pedro de Montes de Oca, Costa Rica Available in: http://www.redalyc.org/articulo.oa?id=44920231013 How to cite Complete issue Scientific Information System More information about this article Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal Journal's homepage in redalyc.org Non-profit academic project, developed under the open access initiative Vertical zonation in the rocky intertidal at Cocos Island (Isla del Coco), Costa Rica: A comparison with other tropical locations Jeffrey Alejandro Sibaja-Cordero Centro de Investigación en Ciencias del Mar y Limnología (CIMAR), Ciudad de la Investigación, Universidad de Costa Rica, San Pedro, San José, 2060, Costa Rica; [email protected] Received 21-IX-2007. Corrected 18-IV-2008. Accepted 11-VI-2008. Abstract: This study presents a description of the tropical intertidal shores of Cocos Island (Isla del Coco), Pacific, Costa Rica. The intertidal zones were surveyed photographically with 116 digital images of quadrats (25 x 25 cm), and 130 of the cliffs in five days. The percent of cover and abundance of species in the intertidal zones were estimated by digital image analysis. Twenty-three taxa were found, Mollusca was the most spe- cious (12 species). Littorinid and neritid mollusks were dominant in the higher littoral area. The predator snails Plicopurpura patula pansa, and Thais brevidentata, high densities of the pulmonate limpet Siphonaria gigas and some patches with low cover of the barnacle Tetraclita stalactifera were present in the mid-littoral zone. The encrusting algae dominate the rock surface at mid and low littoral and erect-frondose forms were only found in sheltered boulder shores. A high similarity in species identity was found between mainland and insular territories of the eastern tropical Pacific. Panamanian islands were most similar in terms of the algae assemblage of Cocos Island. The Galápagos Islands differ from Cocos Island in their higher erect-frondose and crustose algal cover, and lower densities of S. gigas. Grazing may be an important factor in structuring the intertidal community of Cocos Island and other regions of the eastern tropical Pacific. The main grazer in the cliffs and boulders was S. gigas which was larger in size (5.77 cm ± 1.00) than those of mainland and island sites in Costa Rica and Panama. Boulders and cliffs presented changes in the zonation and densities between sheltered or wave exposed areas. The position of organisms on the intertidal cliffs increased at high tidal level in more exposed sections of the coast. Moreover, topographic characteristics of boulders and cliffs influenced the densities of some gastro- pods around the island. Shores with ignimbrites, aa lavas or basaltic walls with a slight inclination towards the sea presented high densities of gastropods. The difference in the assemblages at Cocos Island from those of the continental and insular shores indicates high dynamics within intertidal tropical ecosystems. Rev. Biol. Trop. 56 (Suppl. 2): 171-187. Epub 2008 August 29. Key words: Vertical zonation, Isla del Coco, cliffs, tropical rocky shores, boulders, pulmonate limpets, encrust- ing algae, digital image analysis, Cocos Island, Costa Rica. The variation in species richness of terres- present the same problems in colonization trial and marine species between mainlands and that terrestrial habitat. For example, Benedetti- islands is due to the degree of isolation and size Cecchi et al. (2003) found a difference between of the islands. This topic has been well studied rocky shores of mainland and islands at the and models such as McArthur and Wilson´s Tuscany Archipelago, Italy (20-150 km apart of (1967) “equilibrium” hypothesis have been mainland), in the composition and high cover used to explain differences in assemblages of erect-frondose algae (mainland), and the (Begon et al. 1996). The intertidal organisms low cover of encrusting algae and barnacles have larvae or propagules with short life time (island). Moreover, the difference may have and settlement in specific height range in the been the result of human effects on the main- shore; in this way intertidal shores of island land shores. The islands in the Bay of Panama Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 56 (Suppl. 2): 171-187, August 2008 171 showed difference in cover and abundance from MATERIALS AND METHODS that of the mainland possibly due to differences in predation and wave exposure (Levings & The rocky intertidal shores of Wafer, Garrity 1984, Lubchenco et al. 1984). Yglesias and Chatham bays (Fig. 1) were sam- Rocky shores present high differences in pled at low neap tide on 12, 14 and 17 January vertical distribution of organisms. The verti- 2007, respectively. A total of 116 digital pho- cal zonation patterns at local and regional tographs (4 mega-pixels) were taken of a 25 x scales are caused by structuring factors such as 25 cm quadrats distributed between the differ- temperature, predation, inter-specific competi- ent vertical zonation bands found at each site. tion, wave action, mean tidal variation, salin- Voucher specimens of epifauna were deposited ity and topography (Doty 1957, Dayton 1971, in the Museo de Zoología, Escuela de Biología, Connell 1972, Menge & Sutherland 1976). Universidad de Costa Rica. Algal species were Differences in abundance and frequencies of collected for their identification. species between artificial sea walls (lowest val- Additionally, the rock walls of the cliffs ues) and horizontal or vertical natural intertidal around Cocos Island and its islets were sur- shores were found by Chapman and Bulleri veyed photographically during low neap tide (2003) in Sydney Harbour, Australia. Mettam from a boat on 15 and 18 January 2007. A total (1994) used photographic survey data taken of 130 digital photographs were taking amount- from an aircraft to survey the epibenthos of the ing to a survey of approximately 1074 m of rocky cliffs of Bristol Channel and the Servern coast line (Fig. 1). Estuary and described a change in the com- From the photographs, the percent cover of munities of estuaries and marine areas. Similar sessile species were estimated by digital image variations between and within sites are found analysis using the UTHSCSA Image Tool, in São Miguel Island with sixteen biotopes developed by the University of Texas Health (Wallestein & Neto 2006). Science Center, San Antonio, Texas (ftp:// In the case of Cocos Island (Isla del Coco), maxrad6.uthscsa.edu). Estimates of percent Pacific Costa Rica with 23 km2, the basaltic cover were obtained using three methods. The cliffs and boulder beaches differing in wave first was a direct measure of the area of the exposure and with a tidal range of up to 4 m. The image occupied by each organism. The second island lies 495 km from the mainland of Costa procedure was the manual segmentation on Rica, and more than 630 km of Malpelo and the gray scale to create a binary image (black and Galápagos Islands (Lizano 2001). It was cre- white pixels). The percent of pixels of the cat- ated by volcanic processes (Garrison 2005) and egory of interest represented in the image was has a shoreline similar to that of other islands estimate by the “Count Black/White pixels” in the eastern Pacific. Cocos Island represented command. The third percent cover analysis an interesting scenario because its distance or was used for organisms with hard to distinguish proximity to other locations, for comparisons coloration on the gray scale. Color saturation of the intertidal communities in islands and was changed (Adobe Photoshop) in each spe- mainland in the eastern tropical Pacific. cies category and a background subtraction was This study presents a description of the done between the original and edited image, for vertical zonation of organisms in the intertidal each species category prior to cover estimation. boulders and cliffs at Cocos Island, and com- The abundance of mobile species in quadrats parisons of faunal and algal assemblages and and cliffs images was determined with the limpet density and size with other locations “point” command. in mainland and island in the eastern tropical The length and width of Siphonaria gigas Pacific. were measured in the quadrat photographs of 172 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 56 (Suppl. 2): 171-187, August 2008 Fig. 1. Study sites of the intertidal rocky shore at Cocos Island. January 2007. Open circles are quadrat sampled sites. Black point and lines are cliffs surveyed photographically. the three bays. Twenty individuals were mea- Warwick 1994) to find intertidal level and site sured in situ at Chatham Bay to obtain a size differences in species composition. The value base for calibrating the spatial scale of cliffs of R determines the level of difference, if R=1 images and convert the x-y pixels coordinates the groups are completely different from each to a metric scale. In this way the height above other. The results of the ANOSIM were showed the sea level of each species was estimate. in a two PCA axes of transformed data that pre- serve the euclidean distances among attributes Data analysis: The statistical analysis (Legendre & Gallagher 2001). Finally, the were carried out with PRIMER 4 (Clarke & mean of the transformed data of each species Warwick 1994) and the free software PAST by stratum/site was estimated and expressed (Hammer et al. 2001) (http://folk.uio.no/oham- back to the original scale by the inverse trans- mer/past).