Plant Ecology 147: 21–35, 2000.© 2000 Kluwer Academic Publishers. Printed in the Netherlands.21Vegetation and microclimatic edge effects in two mixed-mesophyticforest fragmentsSophia M. Gehlhausen1, Mark W. Schwartz2,∗& Carol K. Augspurger11Department of Plant Biology, University of Illinois, Urbana-Champaign, 265 Morrill Hall, 505 S. Goodwin Ave.,Urbana, IL 61801, USA;2Department of Environmental Science and Public Policy, 1 Shields Avenue, Universityof California, Davis, CA 95616, USA;∗Author for correspondence (e-mail: [email protected])Received 29 December 1997; accepted in revised form 25 October 1999Key words: Buffer zone, Conservation, Edge effects, Forest fragmentation, Forest microclimate, Herbaceous plantcommunityAbstractForest edges are known to consist of microenvironments that may provide habitat for a different suite of speciesthan forest interiors. Several abiotic attributes of the microenvironment may contribute to this change across theedge to center gradient (e.g., light, air temperature, soil moisture, humidity). Biotic components, such as seeddispersal, may also give rise to changes in species composition from forest edge to interior. We predicted thatabiotic and biotic measures would correlate with distance from forest edge and would differ among aspects. Totest these predictions, we measured abiotic and biotic variables on twelve 175 m transects in each of two 24 haforest fragments in east-central Illinois that have remained in continuous isolation for upwards of 100 years. Bothunivariate and multivariate techniques were used to best describe the complex relationships among abiotic factorsand between abiotic and biotic factors. Results indicate that microclimatic variables differ in the degree to anddistance over which they show an edge effect. Relative humidity shows the widest edge, while light and soilmoisture have the steepest gradients. Aspect influences are evidenced by the existence of more pronounced edgeeffects on south and west edges, except when these edges are protected by adjacent habitat. Edges bordered byagricultural fields have more extreme changes in microclimate than those bordered by trees. According to PCAresults, species richness correlates well with microclimatic variation, especially light and soil moisture; however,in many cases species richness had a differentdepth of edge influence than either of these variables. The herbaceousplant community is heavily dominated by three species. Distributions of individual species as well as changes inplant community composition, estimated with a similarity index, indicate that competition may be influencing theresponse of the vegetation to the edge to interior gradient. This study indicates that edge effects must be consideredwhen the size and potential buffering habitat of forest preserves are planned.IntroductionFragmented forest ecosystems change through time asa result of their isolation as well as other human andnatural disturbances. A reduction of species may oc-cur in fragments that are too small to support theiroriginal flora and fauna (Blake 1983; Blake & Karr1984; Harris 1984). While the reduction in speciesmay result directly from a decrease in forest area, it ismore likely due to the increased perimeter : area ratiothat results from fragmentation and the modificationof abiotic and biotic factors at the forest edge (Lau-rance & Yensen 1991). Edge effects are differences inbiotic or abiotic factors that exist along the borders ofa habitat fragment relative to the interior environment.Changes in microclimate at forest edges, for example,may favor a plant community differentfrom that foundin the forest interior (Noss & Cooperrider 1994; Harris1984).In order to understand the dynamics of the forestedge, ideally, we need to know how both the mi-22croclimate and the vegetation vary and co-vary withincreasing distance from the forest edge. Microclimateis made up of multiple variables, which can be studiedeither independently or in unison. Likewise, severalvariables indicative of changes in vegetation composi-tion, including distributions of individualspecies, maybe integrated through multivariate techniques. The useof multivariate techniques to illuminate interactionsbetween variables may give a more comprehensiveview of biotic and abiotic dynamics at the forest edgethan analyzing variables individually (Murcia 1995).The effectivenessof managementstrategies for for-est fragments depends on an understanding of edgewidth and the changes that occur along habitat edges(Palik & Murphy 1990). Laurance & Yensen (1991)argue that fragment size, shape and edge ‘penetra-tion distance’ are likely to affect the amount of corearea where native flora and fauna are able to per-sist. They conclude that the penetration distance variesdepending upon the variable measured.Documented changes in microclimate typical offorest edges are: higher light, air and soil tempera-tures, wind speed, and vapor pressure deficit (VPD);and lower relative humidity and soil or litter moisture(Brothers & Spingarn 1992; Brothers 1993; Matlack1993; Young & Mitchell 1994; Chen et al. 1995).The extent to which these factors vary with distancefrom edge may be influenced by aspect, or cardinaldirection. Edges facing south and west in the north-ern hemisphere are typically broader than those facingnorth and east (Matlack 1993; Fraver 1994).Typical vegetation responses to the edge environ-ment are: increased presence of exotic species (Lau-rance 1991; Brothers & Spingarn 1992; Fraver 1994),increased sapling and tree densities (Ranney et al.1981; Palik & Murphy 1990; Wales 1972; Broth-ers & Spingarn 1992; Chen et al. 1992; Brothers1993; Fraver 1994), increased shrub cover (Matlack1994), and higher species richness (Ranney et al.1981; Brothers & Spingarn 1992; Fraver 1994). Treespecies distributions may also vary with distance fromthe forest edge (Whitney & Runkle 1981; Brothers1993; Wales 1972), as may tree mortality from windthrow (Chen et al. 1992; Young & Mitchell 1994;Laurance et al. 1998). Evidence of aspect effects onvegetation include variation in exotic species coverand abundance, species richness, and tree species dis-tribution (Wales 1972; Brothers & Spingarn 1992;Fraver 1994).Previous studies have been limited by a lack ofreplication, a lack of continuity in sampling at sev-eral points from edge to center of the fragment, anda lack of consideration of the potential interactionsamong variables in the responses of