Micro-Scale Heterogeneity of Spiders (Arachnida: Araneae) in the Soutpansberg, South Africa: a Comparative Survey and Inventory in Representative Habitats
Total Page：16
File Type：pdf, Size：1020Kb
Micro-scale heterogeneity of spiders (Arachnida: Araneae) in the Soutpansberg, South Africa: a comparative survey and inventory in representative habitats S.H. Foord1*, M.M. Mafadza1, A.S. Dippenaar-Schoeman2 & B.J. Van Rensburg3 1 Department of Biological Sciences, University of Venda, Thohoyandou, 0950 South Africa 2 ARC-Plant Protection Research Institute, Biosystematics Division, Private Bag X134, Queenswood, 0121 South Africa and Department of Zoology and Entomology, University of Pretoria, Pretoria, 0002 South Africa 3 Centre for Invasion Biology and Department of Zoology and Entomology, University of Pretoria, Pretoria, 0002 South Africa Received 5 October 2007. Accepted 11 June 2008 Coarse-scale studies that focus on species distributions and richness neglect heterogeneity that may be present at finer scales. Studies of arthropod assemblage structure at fine (1 × 1 km) scales are rare, but important, because these are the spatial levels at which real world applications are viable. Here we investigate fine-scale variation in spider assemblages, comparing five representative vegetation types in the western Soutpansberg, Limpopo Province, South Africa. We assess these vegetation types in terms of their family and species composition, as well as levels of endemicity, relating these differences with vegetation structure. We inventoried 297 species (49 families) in an area less than 450 ha, as part of South African National Survey of Arachnida. Analysis of the results suggests that endemic taxa are as- sociated with Tall Forest and, to a lesser extent, Woodland. The Woodland had the highest spe- cies diversity, and much of the variation observed in spider assemblage structure is explained by these two vegetation types. Based on vegetation structure variables that explained signifi- cant variation in spider assemblages, human influence through bush encroachment will result in a change of spider assemblages to that of Short Forest and Mosaic Woodland vegetation types, with implications for biodiversity maintenance and heterogeneity. Key words: endemicity, South Africa, spider, SANSA, conservation, Soutpansberg. INTRODUCTION heterogeneity in local biotas is therefore important If a country’s conservation goal is to retain as because: (i) such information will enhance area- much of the region’s biodiversity as possible, then selection techniques (which are currently mostly the delineation of spatial heterogeneity in com- conducted at coarse regional scales) to select units munities or assemblages at the local scale is an defined at the scale of land management units important step towards achieving this. That is, (Wiens 1989; Flather et al. 1997; Rodrigues et al. information on the extent to which local biotas 1999), and (ii) long-term conservation of the biota vary between and within habitat types is neces- that constitutes these heterogeneous communities sary for a better understanding of the underlying will be achieved (Rodrigues et al. 1999). processes explaining local community patterns The Soutpansberg is a major centre of plant (Ricklefs 1987; Cornell 1993). Priority conservation endemism and biodiversity, and has the highest areas identified at broad global or regional scales plant generic and family level diversity among the are often heterogeneous and cannot be translated 18 recognized Centres of Plant Endemism (CPEs) into effective local conservation strategies without for southern Africa (Van Wyk & Smith 2001). reference to local landscape and species distribu- Recent botanical studies conducted in the Sout- tion patterns. This is because local heterogeneity pansberg estimate a total of 3000 plant species can be present across scales and may significantly including 1066 genera (the world-renowned Cape complicate the development of effective regional Floral Kingdom has 1000 genera) (Hahn 2002). The conservation strategies (Flather et al. 1997; Rodrigues region also contains: (i) 33% of South Africa’s et al. 1999). The identification of fine-scale spatial reptiles, 116 species in total, equalling that of the *Author for correspondence. E-mail: [email protected] entire Kruger National Park (Berger et al. 2003); African Zoology 43(2): 156–174 (October 2008) Foord et al.: Micro-scale heterogeneity of spiders in the Soutpansberg, South Africa 157 (ii) 60% of South Africa’s mammal species, which reserve planning, is therefore not viable because of a represents more mammal species per unit area paucity of data. than seven of the eight most diverse biodiversity Here we attempt to use information on spiders hotspots of the world (Berger et al. 2003); and (iii) to estimate the relative conservation importance 75% of South Africa’s avifauna (Berger et al. 2003). of the dominant vegetation types in the western Although the area was not rated as an urgent Soutpansberg. This area is climatically and topo- priority for conservation by Reyers (2003), the very graphically the most varied part of the Soutpans- high levels of biological diversity of the Soutpans- berg, and recent ad hoc collections recorded 47 berg gives it a high long-term priority for conser- spider families (74% of the known families from vation (Van Wyk & Smith 2001). It is also the site of South Africa) in an area less than 450 ha. We a proposed Biosphere Reserve (UNESCO 2005) initiated a quantitative small-scale survey in this and a process has been initiated for the proposed same area, which afforded the following opportu- Soutpansberg to Limpopo Biosphere Reserve nities: (i) contributing to SANSA’s database by (Hahn 2002). This process includes the identifica- inventorying spiders and measuring species rich- tion of conservation hotspots that would form the ness, abundance and endemicity in major vegeta- basis for the selection of core conservation areas. tion types of the Soutpansberg, (ii) determining Spiders are a ubiquitous component of inverte- whether spider assemblage structure differed brate assemblages and important generalist between the major vegetation types at a fine scale predators in ecosystems (Wise 2005) with the and how restricted taxa are to these habitats, and potential to complement existing large-scale area (iii) establishing which, and to what extent, vege- selection activities based on vertebrates (Jocqué & tation structure variables are related to these Dippenaar-Schoeman 2006). They are known to be differences. sensitive towards fine-scale changes in environ- mental factors, are relatively easy to identify based METHODS on external genitalia and have the potential to act as indicators of habitat quality and conservation Study area value (Jocqué 1981). Spiders are a ‘hyperdiverse’ Fieldwork was conducted in the western parts of taxon, and because of a general awareness of the Soutpansberg mountain range situated near spiders within the public realm, have the potential the northern border of South Africa with Zimbabwe. to act as a focal taxon for conservation assessments The mountain forms a geographic unit with the (New 1999). Recent developments in South Africa Makgabeng Plateau, Blouberg Mountain to the have added weight to this suggestion with the west and the Waterberg to the south. inclusion of spiders in conservation assessments Transects were set out on the farm Lajuma and red listing of species (Dippenaar-Schoeman & (23°1.49’S, 29°25.74’E) that has a surface area of Haddad 2006). The South African National Survey 430 ha. The highest point of the mountain range, of Arachnida (SANSA) and its recent collaboration Letjume (1747 m a.s.l), is on Lajuma and is c. 800 m with the South African National Biodiversity Insti- above the surrounding plains (Gaigher 2006). tute will provide the critical information needed Average annual rainfall at Lajuma is 730 mm, vary- for this process (Dippenaar-Schoeman & Craemer ing considerably from year to year (Gaigher 2006). 2000). Commercial farming in the western Soutpansberg Contemporary research has shown that, at least has resulted in vegetation that mainly consists of for certain invertebrate groups such as spiders, thickets and scrublands, vegetation with patches the eastern parts of South Africa have comparable of grass and fragmentary groundwater forests levels of endemism, at both the generic and specific below cliffs. Higher altitudes are characterized by levels, to that of the Cape Floral Kingdom and areas with an interesting mosaic vegetation type Succulent Karoo (Huber 2003). Although levels of consisting of short grassland and ‘islands’ of closed endemicity seem to be high, very little is known woody cover. about the levels of invertebrate diversity in the Soutpansberg, and the region in general, has been Sampling and analytical procedures poorly surveyed with only 13 survey sites (and Five representative vegetation types, based on an average of 13 records per site). Determining broad-scale structural classification (Edwards 1983) conservation priorities for spiders, namely core of the western Soutpansberg were identified, areas and buffer zones as set out in biosphere namely Tall Forest (TF) or Evergreen Northern 158 African Zoology Vol. 43, No. 2, October 2008 Mistbelt Forest (Mucina et al. 2005), 10–20 m high of three parts 70% ethyl alcohol and one part and characterized by plant species such as Rhus glycerol. The samples were emptied on a weekly chirindensis, Gymnosporia havianus, and Eugenia basis and remained active for 14 days. Finally, leaf natalitia with a weakly developed understorey; litter, in three (1×1m)randomly selected qua- Short Forest (SF) or Semi-deciduous