Annales Zoologici Fennici 39: 131-149
Total Page:16
File Type:pdf, Size:1020Kb
ANN. ZOOL. FENNICI Vol. 39 • Carabids in thinned and clear-cut stands 131 Ann. Zool. Fennici 39: 131–149 ISSN 0003-455X Helsinki 14 June 2002 © Finnish Zoological and Botanical Publishing Board 2002 Boreal carabid-beetle (Coleoptera, Carabidae) assemblages in thinned uneven-aged and clear-cut spruce stands Matti Koivula Department of Ecology and Systematics, Division of Population Biology, P.O. Box 65, FIN-00014 University of Helsinki, Finland (e-mail: matti. koivula@helsinki.fi ) Received 7 February 2001, accepted 17 May 2001 Koivula, M. 2002: Boreal carabid-beetle (Coleoptera, Carabidae) assem- blages in thinned uneven-aged and clear-cut spruce stands. — Ann. Zool. Fennici 39: 131–149. Forestry has altered the boreal fl ora and fauna strongly during the 1900s. At present, logging methods other than clear-cutting are often applied but the ecological effects of these modifi cations are poorly studied. I collected carabid beetles in 8 uncut, 8 thinned (10%–30% of trees removed, with the aim of generating an uneven age structure) and 8 clear-cut, spruce-dominated stands, by using pitfall traps in central Finland during 1995–1998. The carabid species fell into three distinctive ecological groups in the multivariate analyses: forest, open-habitat and Sphagnum bog species. The forest species further formed a continuum from forest specialists to canopy- closure generalists. Logging affected the forest species slightly, while generalists and open-habitat species benefi tted from clear-cutting. Thinning maintained the forest-fl oor carabid assemblage well. Site characteristics, such as the amount of trees and bottom and fi eld-layer vegetation, were important determinants of carabid assemblages. Certain within-stand habitat types (e.g. spruce mires) were shown to be important for carabid diversity and should be managed with methods other than clear-cutting, in order to avoid extinctions of local populations. Increasing distance to the nearest source habitat had a negative effect on the abundance and distribution of carabids. Therefore, landscape-level forestry planning is also important for the maintenance of forest-species assemblages. Introduction mainly of managed forests of different succes- sional stages (Hansson 1992). Over 90% of The Fennoscandian boreal landscape consists Finnish forests are managed (Sevola 1999) and 132 Koivula • ANN. ZOOL. FENNICI Vol. 39 usually cut at an age of 90–120 years (Kuusela tially be found among carabid beetles (Coleop- 1990). Old-growth forests (for defi nition, see tera, Carabidae). Boreal carabids can be roughly Aksenov et al. 1999) on the other hand, are divided into open-habitat, (habitat-type) gener- extremely scarce in the southern parts of Fen- alist and forest species (e.g. Lindroth 1985, noscandia; protected forests cover 1% of the 1986, Kinnunen 1999). Bortmann (1996) divided forested area in southern Finland and only 5% German beech-forest species further into for- of this represents old-growth forests (Virkkala et est-centre, forest-margin and clear-cut species. al. 2000). This is a consequence of the fragmen- Niemelä et al. (1988, 1993a) made a similar divi- tation process that has lasted several hundreds sion for boreal coniferous-forest assemblages, of years, but the most drastic effects of forestry except that there were habitat generalists instead on forest biota have taken place only during of margin species. the last 50 years (Niemelä 1999). Fragmentation Most forest species require certain elements has strong effects on forest species; for example, of forests rather than “average forest” (Niemelä in an experimentally fragmented tropical rain et al. 1996). Forest-fl oor environmental factors forest in Amazonia, Didham et al. (1998) showed create a heterogeneous mosaic of microsites and that the abundances of 15 out of 32 abundant the resulting site characteristics may appear dif- beetle species were signifi cantly affected by ferent for different species. Consequently, some forest fragmentation. Also extinctions occur (e.g. species with strict microhabitat demands occur Saunders et al. 1991, Harrison & Bruna 1999) in metapopulations (Hanski 1999), whereas for and over one-third of the Finnish red-listed spe- some other species with wider tolerance, the cies are threatened because of forestry (Rassi environment may be divided into source and sink et al. 2000). The inadequacy of southern Finn- habitats (Pulliam 1988) and some may be able ish old-growth forests for the maintenance of to utilize several habitat types. Carabids are dis- old-growth forest specialists (e.g. Heliövaara tributed non-randomly among stands of different & Väisänen 1984, Heikkinen et al. 2000) has quality and also within the stands (Niemelä et al. led to demands to increase the amount of pro- 1992a). Factors such as moisture, temperature, tected forests, restore mature managed forests food abundance and occurrence of red wood and improve the quality of the surrounding man- ants (Formica rufa group), affect the distribu- aged landscape matrix (Niemelä 1997, 1999, tions and abundances of carabids and other Nilsson 1997). The latter two are thought to be invertebrates (Thiele 1977, Niemelä et al. 1986, best achievable by including natural processes 1990, 1992, Koivula et al. 1999, Laakso & and structural and compositional elements into Setälä 2000). Increased vegetational richness the managed stands (Fries et al. 1997). (Siemann et al. 1998) and herb-layer cover The long-term persistence of many forest (Bortmann 1996) also affect invertebrate diver- species is dependent on populations which live sity. Logging affects the boreal carabids, for in managed forests (Lindenmeyer & Franklin example, through changes in the above factors. 1997). However, managed forests host less thre- They are often more numerous and form richer atened forest-specialist species than old-growth assemblages in open habitats than in forests forests do (e.g. Esseen et al. 1997) and also (e.g. Niemelä & Halme 1992, Beaudry et al. many commoner species occur there in low 1997, Kinnunen 1999) though some species abundances. Examples are e.g. epiphytic lichens may suffer from clear-cutting (e.g. Langor et al. (Kuusinen 1994, 1996, Dettki & Esseen 1998), 1994, Spence et al. 1996). In Finland, regenera- bryophytes (Andersson & Hytteborn 1991), beet- tion clear-cutting makes up 40%–50% of the les living in dead wood (Väisänen et al. 1993, forest area of all annual logging, while thinning Siitonen & Martikainen 1994) and Mycetophili- is almost equally commonly used with only a dae insects (Økland 1994). Therefore, for the slight increase between 1970 and 1998 (Västilä purposes of ecological studies, species with inter- & Herrala-Ylinen 1999). Although the ecological mediate commonness and relatively strict micro- effects of clear-cutting are well documented (e.g. habitat requirements may be useful indicators of Huhta 1976, Niemelä et al. 1993a, 1993b, Pet- environmental change. Such species can poten- tersson 1996, Davies & Margules 1998, Abild- ANN. ZOOL. FENNICI Vol. 39 • Carabids in thinned and clear-cut stands 133 snes & Tømmerås 2000), the effects of other left intact (no thinning or other management harvesting methods are much less studied. Stud- activities) for at least 10 years before the exper- ies concerning thinning have mostly focused on iment. Prior to logging, the herb layer was its effects on tree growth and studies concerning dominated by Vaccinium vitis-idaea and V. myr- the importance of uneven age structure of trees tillus dwarf shrubs, and Dicranum, Pleurozium for invertebrates are lacking. and Hylocomium mosses covered most of the In the present paper, I focus on four ques- bottom layer. The surroundings of the study tions concerning carabid beetles in managed stands varied from recently cut stands to mature stands. (1) Is the species division into open- (90–150 year-old) spruce forests. The treated habitat, generalist and forest species relevant in area within each study stand was a one-hectare forest ecology? (2) How much site variation is square and its immediate surroundings. The dis- there in the study stands, and do the carabid tance to the nearest study stand (with different catches refl ect the microhabitat characteristics treatment) within a given area varied between and logging? (3) Do the studied logging meth- 50 and 1500 m. Within the study areas, a road ods affect the carabid catches? (4) Does the and/or different stand type (dissimilar in forest quality of adjacent habitat matrix explain the type, age and species distribution of dominant abundance patterns of carabids? These ques- trees) always separated the study stands. Thus tions are investigated using pitfall trappings for carabids they were independent from each over a short time scale, i.e. two years after other and one treatment was randomly assigned logging. Species responses to habitat variability to each stand. The randomisation was done and natural and anthropogenic processes may be simultaneously with the establishment of the determined by different factors acting at differ- study areas in spring 1995, before the experi- ent spatial levels (Addicott et al. 1987, Wiens ment was started. 1989). It is therefore important to study the The study design follows a Before/After effects of forestry at several spatial scales (Haila with Control/Impact design (BACI, Underwood & Kouki 1994, Niemelä 1999). Thus, logging 1991). I thus had data from both before and is examined at within stand (site) and landscape after the logging and control (not cut) and (between study stands within an area) levels in impact (clear-cut and thinned) stands. The study this paper. In Finland, there are no red-listed began in 1995 (pre-treatment study year) and (threatened) forest carabids (Rassi et al. 2000), the stands to be cut were felled the following but in this paper carabids are used as tools, winter (1995–1996). In each study area, there rather than examples of endangered species, to was a control stand, a clear-cut stand, and a study the effects of logging. stand where trees were cut with the aim of developing an uneven age structure of trees (hereafter referred to as thinned stands).