Habitats of Snakes in the Rainforests of Eket (Akwa-Ibom State, South-Eastern Nigeria) (Squamata: Serpentes)
Total Page:16
File Type:pdf, Size:1020Kb
©Österreichische Gesellschaft für Herpetologie e.V., Wien, Austria, download unter www.biologiezentrum.at HERPETOZOA 11 (3/4): 99-107 Wien, 26. Febniar 1999 Habitats of snakes in the rainforests of Eket (Akwa-Ibom State, south-eastern Nigeria) (Squamata: Serpentes) Habitate von Schlangen im Regenwald von Eket (Provinz Akwa-Ibom, Südost-Nigeria) (Squamata: Serpentes) LUCA LUISELLI & GODFREY e. AKANI KURZFASSUNG Wir untersuchten den Lebensraum von Schlangen in einem Regenwaldgebiet Südost-Nigerias in der Umge- bung von Eket (Provinz Akwa-Ibom). Das Untersuchungsgebiet umfaßte ein Mosaik von Sumpfregenwald, nicht sumpfigem Regenwald, landwirtschaftlich genutzten Flächen und Siedlungsraum; letzterer erstreckt sich besonders entlang der Hauptstraßen. Insgesamt wurden während der Untersuchung (Juni bis September 1997; 109 Felderhe- bungstage) 237 Exemplare von Schlangen aus 24 Arten beobachtet. Mit einem UPGMA-Dendrogramm auf Grund- lage der Ähnlichkeit der Lebensräume lassen sich die Schlangenarten zu sieben Clustem bündeln: (i) Calabaha rein- hardtii, Naja nigricollis, Thelotornis kirtlandii; (ü) Psammophis phillipsi, Gastropyxis smaragdina; (iii) Python regius, Athens squamiger, Bitis gabonica; (iv) Python sebae, Dendroaspis jamesoni; (v) Bitis nasicornis, Naja melanoleuca, Grayia smythii und andere typische Sumpfwaldbewohner, (vi) Aparallactus modestus, Mehelya crossi, Mehelya poensis; und (vii) Dasypeltis sp. Von den drei Haupttypen von Lebensräumen des Untersuchungs- gebietes beherbergte der Sumpfregenwald die artenreichsten und sehr wahrscheinlich auch dichtesten Schlangenpopu- lationen. Betrachtungen über die Schlangengesellschaften der drei Haupttypen von Lebensräumen im Untersuchungs- gebiet werden angestellt. ABSTRACT The habitat of snakes was studied in a rainforest area of south-eastern Nigeria, situated in the surroundings of Eket (Akwa-Ibom State). Patches of both swamp rainforest and dryland rainforest were available in the territory, which was however characterized by a mosaic of farmlands, cultivations and suburbia, especially near the main roads. A total of 237 snake specimens, composed of 24 species, was observed throughout the research period (June to Sep- tember 1997; 109 days of field work). A UPGMA dendrogam based on habitat similarities between species allows to define seven clusters of snakes: (i) Calabaha reinhardtii, Naja nigricollis, Thelotornis kirtlandii; (ii) Psammophis phillipsi, Gastropyxis smaragdina; (iii) Python regius, Atheris squamiger, Bitis gabonica; (iv) Python sebae, Den- droaspis jamesoni; (v) Bitis nasicornis, Naja melanoleuca, Grayia smythii, and other typical swamp rainforest spe- cies; (vi) Aparallactus modestus, Mehelya crossi, Mehelya poensis; and (vii) Dasypeltis sp. Among the three main habitat types prevailing in the study area, the swamp rainforest patches housed the highest snake species diversity, and very probably the most abundant snake populations. Some considerations on the snake assemblages in the main habitat types of the study area are given. KEY WORDS Reptilia, Squamata, Serpentes; Afrotropical snakes, habitat, ecology, community structure, rainforests, Eket, Nigeria. INTRODUCTION Rainforests of south-eastern Nigeria the Niger Delta (Port Harcourt, Rivers State) house a variety of remarkable vertebrate to the Cross River bordering the Cameroon species, which are still largely unknown (Calabar, Cross River State): Procolobus (POWELL 1993; IsouN & al. 1996). Even badius epieni and Cercopithecus sclateri among primates, "which represent the best (OATES & al. 1992, 1994; TOOZE 1995). known order of mammals and possibly of Our knowledge of Nigerian reptiles in gen- vertebrates, two remarkable new taxa have eral, and snakes in particular, is still recently been described from the south- scarce, in spite of several more or less de- eastern Nigeria region which extends from tailed contributions and records (e.g., ©Österreichische Gesellschaft für Herpetologie e.V., Wien, Austria, download unter www.biologiezentrum.at 100 LUCA LUISELLI & GODFREY e. AKANI ROMER 1953; DUNGER 1971a, 1971b, servations on the habitats of snake assem- 1972a, 1972b; JOGER 1981, 1982; BUTLER blages inhabiting the forests of Eket (Ak- & REID 1986, 1990; REID 1989; SCHMITT wa-Ibom State), a town situated in south- 1996, LUISELLI & al. 1998 a, 1998 b; eastern Nigeria, between the cities of Port AKANI & al. 1998, 1999; LUISELLI & AKANI Harcourt (Rivers State) and Calabar (Cross 1999 a, 1999 b). River State). With regard to habitat preferences of Compared with the study of BUTLER Nigerian snakes, we can only avail our- & REID (1986), the present one is based on selves of a contribution by BUTLER & REID a different approach, since it does not make (1986), which however presents merely use of an extensive search across a whole qualitative considerations, without any region, but focuses on the habitat prefer- quantitative and statistical approach. In ences of snakes in a specific rainforest area. this paper, we report some quantitative ob- STUDY AREA AND METHODS Study area monthly maximum temperatures ranged between 27°C and 34°C, while the minima All observations presented here are varied between 22°C and 24°C. The Eket based on data collected from June to Sep- region is one of the wettest of the world, tember 1997 in the surroundings of Eket with an average yearly rainfall of more (04°50'N; 07° 59'E), one of the main than 3000 mm (source: Department of Ge- towns of the Akwa-Ibom State (south- ography, University of Calabar). eastern Nigeria). The area is heavily popu- lated, with several villages and small towns Methods in the surroundings of Eket. Several terrestrial macrohabitat types, Most observations were carried out in potentially available to snakes, were sam- ten sample stations, 10 ha in size each, pled throughout the present study. Great which are well representative of the various parts of the territory are occupied by farm- macrohabitat types available to snakes in lands and cultivations (cassava, yam, oil the area. These study sites were not far palm, plantain, banana, pineapple, etc), from Ibeno. A total of 109 days of research with wide patches of "bush" interspersed were spent in the field. Field trips were amongst the various villages. The "bush" made on both sunny and rainy days. The was partly of the dryland rainforest type, everyday field work was done regularly, and partly of the freshwater swamp-forest approximately from 08.00 a.m. to 06.00 type (moist lowland forest). Dryland rain- p.m. Because of potential security risks due forests were (i) primary, (ii) secondary, or to groups of criminals operating in the re- (iii) bushy forests. Freshwater swamp- gion, we were usually unable to conduct forests were permanently or temporarily field trips during twilight and nocturnal flooded. For practical reasons, in the fol- hours. Random routes for locating snakes lowing analyses, we assigned these habitats were set up throughout all habitat types to three1 main categories: (i) farmlands and available in the study area. When seen, suburbia (including cultivations, compounds, snakes were captured by hand (see PHELPS and gardens), (ii) dryland rainforest, and 1981, 1989, for the description of a method (iii) swamp (moist lowland) rainforest. to handle large African venomous snakes), The climate of the study area is tropi- sexed by analysing the morphology of the cal sub-Saharan, with well-marked dry and tail (a method which is easy to apply to wet seasons and relatively little monthly most snakes but psammophine colubrids), fluctuations in maximum and minimum and measured for snout-vent length (SVL) temperatures (GRIFFITHS 1972). The dry and tail length (TL, both to the nearest ± season extends from November to April, 0.5 cm), and then individually marked by the wet season from May to October, with ventral scale-clipping. However, data con- the highest rainfall peak during July. Mean cerning body sizes, food habits, and other ©Österreichische Gesellschaft für Herpetologie e.V., Wien, Austria, download unter www.biologiezentrum.at Habitat characteristics of Nigerian snakes 101 ecological matter are reported elsewhere Snakes found dead (squashed by cars (LUISELLI & AKANI 1999 b). A white num- along the road, macheted by farmers, or ber was painted to the back of each cap- offered in bush-meat markets of local tured snake to facilitate later identification tribes) were initially preserved in 10% of the specimen from a distance without formalin and stored in our private herpeto- further disturbing it. Meteorological condi- logical collections or in the collections of tions and time of observation were also re- the Zoological Museum of the Rivers State corded. Pitfall traps and drift fences were University of Science and Technology, Port used in each habitat type. Moreover, sev- Harcourt (Rivers State, Nigeria). eral snakes were sampled under flat objects Species diversity (D) was calculated which were placed on the ground and using Simpson's (1949) diversity index. checked each day (VITT & VANGILDER Species dominance (d) was assessed by 1983). Specimens captured by local people means of the Berger-Parker index (MAGUR- were also examined, although in some RAN 1988): cases it was difficult to determine the habitat type where the snake was found. d = Nmax/N Some individuals of Bitis gabonica (DUMÈ- where N is the total number of individuals RIL& BlBRON, 1845), B. nasicornis (SHAW, observed and N^ is the total number of 1802), Dendroaspis jamesoni (TRAILL, individuals of the most abundant species.