Epigeic spider communities at the timberline of Alp Flix (Switzerland: Grisons) Diplomarbeit der Philosophisch-naturwissenschaftlichen Fakultät der Universität Bern vorgelegt von Holger Frick 2005 Leiter der Arbeit: Prof. Dr. Wolfgang Nentwig, Zoologisches Institut der Universität Bern Ich danke Prof. Wolfgang Nentwig herzlich für die Leitung meiner Diplomarbeit und dafür dass er immer Zeit für mich hatte. Ein ganz spezieller Dank gilt dem Betreuer meiner Diplomarbeit Dr. Christian Kropf, der mir stets wichtige wissenschaftliche und persöhnliche Ratschläge gab und mir immer mit motivierenden Worten zur Seite stand. Die Diskussionen mit ihm, sein Wissen und Engagement haben mich in vielerlei Hinsicht weitergebracht. Bei der Belegschaft des Naturhistorischen Museums Bern bedanke ich mich für das tolle Arbeitsklima und der Hilfe bei allen möglichen Problemen. Schliesslich möchte ich meiner Familie für ihre Unterstützung während meines ganzen Studiums recht herzlich Danke sagen. 2 Table of contents Part 1: Influence of stand-alone trees on epigeic spiders (Araneae) at the alpine timberline Titel……………………………………………………………………………………..5 Abstract…………………………………………………………………………………6 Introduction…………………………………………………………………..…………7 Methods…………………………………………………………………………………7 Study area……………………………………………………………………….7 Sampling of spiders…………………………………………..…………………8 Parameters………………………………………………………………………8 Data analysis….…………………………………………..….…………………9 Results…………………………………………………………………………………11 Community composition……….……………………………………..……….11 Activity densities of selected species …………………..……………………..12 Discussion…………………………………………………………..…………………14 Community composition ……………………..…………………….…………14 Activity densities of selected species……………………………….…………15 Acknowledgements……………………………………………………………………18 References……………………………………………………..………………………18 Figures…………………………………………………………………………………23 Tables…………………………………….……………………………………………29 3 Part 2: Spiders (Arachnida: Araneae) of the timberline in the Swiss Central Alps (Alp Flix, Grisons) Title……………………………………………………………….…………...………31 Abstract…………………………………………………………..……………………32 Introduction……………………………………………………………………………33 Methods……………………………………………………………..…………………33 Study site……………………………………………………..………………..33 Study period………………………………………………….….…………….34 Sampling methods for spiders on trees…………………………….………….34 Sampling methods for ground dwelling spiders……………….....………..….35 Determination…………………………………………………….…………...36 Results and Discussion…………………….………………………………….………36 List of species…………………………………………………………………36 Comments on first records, endemic and rare species and their phenology…..36 Acknowledgements……………………………………………………………………44 References……………………………………………………………..………………44 Figures…………………………………………………………………………………53 Tables………………………………………………………………………………….56 4 Influence of stand-alone trees on epigeic spiders (Araneae) at the alpine timberline Holger Frick, Wolfgang Nentwig and Christian Kropf Holger Frick ([email protected]), Zoological Institute, University of Bern, Baltzerstrasse 6, CH-3012 Bern, Switzerland, and Natural History Museum Bern, Department of Invertebrates, Bernastrasse 15, CH-3005 Bern. Wolfgang Nentwig ([email protected]), Zoological Institute, University of Bern, Baltzerstrasse 6, CH-3012 Bern, Switzerland. Corresponding author: Christian Kropf ([email protected]), Natural History Museum Bern, Department of Invertebrates, Bernastrasse 15, CH-3005 Bern, Switzerland. 5 Abstract We investigated epigeic spiders (Araneae) under stand-alone Norway spruce trees (Picea abies) at the alpine timberline using pitfall traps. These were positioned in three different ranges, i.e. an inner range (close to the tree trunk), a median range, and an outer range (at the outer limit of branch cover). We studied community composition and activity densities with respect to three significantly interrelated parameters: (1) Relative distance of a trap from the tree trunk, (2) branch cover of a trap, and (3) vegetation cover. Community composition: Species numbers correlated negatively with the relative distance from the tree trunk in linyphiids, and positively in lycosids and gnaphosids. Branch cover correlated positively with linyphiid species numbers, and negatively with lycosids. Linyphiidae showed higher activity densities in the inner ranges, Lycosidae and Gnaphosidae in the outer ranges. A Chi2-test revealed also a significant difference for the total number of specimens per range that increases from the inner to the outer range. Activity densities of selected species: 11 out of 14 species were significantly correlated to the relative distance to the tree trunk, 10 to the branch cover, and 2 to vegetation cover. Open- land species preferred the outer range, forest species the inner range. One species, the linyphiid Caracladus avicula, possibly can be classified as a habitat specialist of the alpine timberline. The data show that stand-alone spruce trees are an important structural element of the timberline that offers habitats for a wide variety of species with different habitat preferences. Thus, the alpine timberline could play a key role in future conservation efforts. 6 1. Introduction Structural heterogeneity of a biotope is one of the most important factors influencing animal communities and species diversity (Hatley and MacMahon 1980, Horvath et al. 2000, Nentwig et al. 2004, Niemelä et al. 1996, Robinson 1981, Rosenzweig and Abramsky 1993). The alpine timberline represents one of the most heterogeneous and biologically diverse biotopes of Central Europe (Thaler 1989). Despite of this, studies on small-scale distribution patterns of animal communities at the timberline are scarce and our understanding of the importance of certain structural elements on these patterns is still unsatisfying. Spiders (Araneae) are an important, abundant and “megadiverse” (Coddington et al. 1996) group of epigeic arthropods with elaborate habitat preferences of species (Bauchhenss 1990, Foelix 1996, Samu et al. 1999, Wise 1993 and others). Thus, they are suitable indicative organisms for studying field-ecological problems. We investigated the small-scale distribution patterns of epigeic spiders around stand-alone trees at the timberline in the Swiss Central Alps. The epigeic fauna around these single trees deserves special interest as the trees presumably offer microhabitats for forest species and represent a sharp boundary against the open dwarf shrub heath and alpine meadow. So, as a working hypothesis, we assume that the mosaic-like structured timberline harbors not only habitat specialists, but also a broad range of small-scale distributed forest and open-land species. However, to our knowledge, these stand-alone trees at the alpine timberline have never been studied before in this respect. 2. Methods: 2.1 Study area Alp Flix (GPS: 769 400 / 154 350; WGS84 coordinates: N: 46° 31' 8.41'', E: 9° 38' 47.12'') is situated in the Central Swiss Alps in the Canton Grisons and belongs to the village of Sur (Fig. 1). It is an approximately 1000 m broad and south-west exposed terrace at 1950 m above sea level with adjacent high mountain tops (ca. 3000 m a.s.l.) and a 400 m deep valley. In the south-west, the study site is bordered by the forest edge with stand alone Norway spruce trees 7 (Picea abies), that we chose as study objects. The trees are surrounded by dwarf shrub heath, dominated by Juniperus communis and Rhododendron ferrugineum combined with alpine meadow patches (Hänggi and Müller 2001). Occasional cattle-grazing occurs throughout the vegetation period. 2.2 Sampling of spiders We selected four Norway spruce trees of similar age and size that were at least 100 m apart (ca. 8 m in diameter and ca. 12 m height) and used pitfall traps for assessment of activity densities (Luff 1975). Nine pitfall traps were placed around each tree, in three directions with three traps each. We defined three ranges: an inner range, from the tree trunk up to about 1 m away; a median range, from ca. 1 m to ca. 2.5 m from the tree trunk and an outer range, from ca. 2.5 m from the tree trunk until the outer border of the branch cover at approx. 4 m from the tree trunk (Fig. 2). The traps consisted of white plastic cups with an upper diameter of 6.9 cm and a depth of 7.5 cm. We filled them with 4 % formaldehyde and 0.05 % SDS as tenside until 3 cm under the upper edge and fixed a quadrangular transparent plastic cover (15 x 15 cm) with 3 wooden rods 5 cm above the traps. The traps were emptied once a month between 15th May 2003 and 28th October 2003 (snow-free time) and a last time on 24th May 2004, when the snow started to melt. 2.3 Parameters: We recorded three parameters: 1) Relative distance from tree trunk: Shape and outline of the trees were not exactly equal. Therefore, we calculated the position of traps between tree trunk and outer limit of branch cover as a number between 0 (trap at tree trunk) and 1 (trap at outer limit of branch cover). The distance between the tree trunk and the outer limit of the branch cover was measured in eight different directions: N, NE, E, SE, S, SW, W and NW (Fig. 2). In this way, eight triangles can be defined (Fig. 3). The outer tips of branches (D) were considered to lie on c, the distance between NE and E (Fig. 3). The relative position (r) of each pitfall trap (P) between the tree trunk (T) and the interpolated outer tip of the branches (D) was calculated by using the following formulas: 8 t (1) r = d a⋅ b ⋅sin()γ (2) d = c ⋅sin() 180 −δ − β ⎛ b ⎞ (3) β = a sin⎜ sin γ ⋅ ⎟ ⎝ c ⎠ (4) c = a2+ b 2 −2 ⋅ a ⋅ b ⋅ cos()γ With