- Species richness in deciduous forests: Effects of species pool and environmental variables - 505 Journal of Vegetation Science 13: 505-516, 2002 �� IAVS Opulus Press Uppsala. Printed in Sweden Abstract. The study was conducted in deciduous forests of two Swedish regions, ��land and Uppland. It had two objec- tives: to (1) test the species pool hypothesis by examining if differences in small-scale species richness are related to dif- ferences in large-scale species richness and the size of the regional species pool, and (2) to examine the relationship between species richness and productivity and its scale- dependence. The first data set comprised 36 sites of moderate to high productivity. In each site, we recorded the presence of vascu- lar plant species in nested plots ranging from 0.001 to 1000 m2 and measured several environmental variables. Soil pH and Ellenberg site indicator scores for nitrogen were used as estimators of productivity. The second data set included 24 transects (each with 20 1-m2 plots) on ��land in sites with low to high productivity. Species number, soil pH and relative light intensity were determined in each plot. The forest sites on ��land were more species-rich than the Uppland sites on all spatial scales, although environmental conditions were simi- lar. Small-scale and large-scale species richness were posi- tively correlated. The results present evidence in favour of the species pool hypothesis. In the nested-plots data set, species number was negatively correlated with pH and nitrogen indi- cator scores, whereas a unimodal relationship between species number and pH was found for the transect data set. These results, as well as previously published data, support the hump-shaped relationship between species richness and pro- ductivity in Swedish deciduous forests. Two explanations for the higher species richness of the sites with moderate produc- tivity are given: first, these sites have a higher environmental heterogeneity and second, they have a larger ���habitat-specific��� species pool. Keywords: Ellenberg indicator values Habitat-specific species pool Hump-shaped model Regional species pool Soil pH Spatial heterogeneity Spatial scale. Nomenclature: Karlsson (1997). Introduction Variation in species richness has long attracted the interest of ecologists and a large number of theories of species diversity have been put forward (Tilman 1982 Huston 1994 Rosenzweig 1995 Gaston 2000). This paper deals with two topics that are currently under debate: (1) the species pool hypothesis and (2) the relationship between species number and productivity. These topics will be addressed by studying species richness patterns across geographical and environmen- tal gradients in deciduous forests in South Sweden. A recent approach to studying differences in species richness between communities has been the species pool hypothesis (Taylor et al. 1990 Eriksson 1993 P��rtel et al. 1996 Zobel 1997 Zobel et al. 1998). This hypothesis suggests that the number of species in plots of a certain community type is not only dependent on biotic mechanisms (the most important probably being competition) but is also related to the general species richness of that community type in a region. In other words, for a given habitat, the pool of available species or the large-scale species richness at least partly deter- mines the species richness on smaller scales. Thus, regional differences in species richness for a community type should be observed at all spatial scales. The con- cept of the species pool theory and the limits to its testing were recently discussed by Lep�� (2001) and Wilson & Anderson (2001). Species richness is correlated with productivity in most situations (Huston 1994 Rosenzweig 1995). The shape of this relationship may differ depending on the geographical or taxonomical context and the underlying mechanisms are still unclear (Rosenzweig & Abramsky 1993 Gaston 2000). Within regions, species richness has been hypothesised to show a hump-shaped or uni- modal response along the productivity gradient. Species richness rises from low to moderate levels of product- ivity to decline towards even higher productivity levels. Evidence for this pattern has been given for both plants and animals (Grime 1979 Garc��a et al. 1993 Rosenzweig Species richness in deciduous forests: Effects of species pools and environmental variables Dupr��, Cecilia1,2 Wessberg, Camilla2 & Diekmann, Martin1* 1Department of Ecology and Evolutionary Biology, Bremen University, FB 2, Leobener Str., DE-28359 Bremen, Germany 2Department of Plant Ecology, Evolutionary Biology Centre, Uppsala University, Villav��gen 14, SE-75236 Uppsala, Sweden *Corresponding author: Fax +494212187052 E-mail mdiekman@uni-bremen.de

506 Dupr��, C. et al. & Abramsky 1993 Tilman & Pacala 1993) and various explanations have been proposed (Rosenzweig & Abramsky 1993). One main hypothesis stresses the role of resource or habitat heterogeneity that is assumed to first increase and then decline along the productivity gradient. Low heterogeneity in limiting resources at high productivity may then cause increased competition (e.g. for light) and eventually competitive exclusion (Tilman 1982). Oksanen (1996), however, stated that the unimodal relationship between species richness and bio- mass might arise as an artefact of scale (but see e.g. Grime 1997 Zobel & Liira 1997). Abrams (1995) noted that species richness curves along the productivity gradient must not necessarily be unimodal, but may be monotonic, even if the entire gradient is considered. A comprehen- sive survey of the literature pointed at a large variation in the type of relationship between species number and productivity depending on, among others, gradient length, taxonomic unit, measure of productivity (Waide et al. 1999) and especially scale (Gaston 2000). Much of the theoretical framework with regard to the above- mentioned patterns has been developed in grasslands and other herb-dominated communities, whereas forests have received comparatively little attention. The hypotheses described above are concerned with different aspects of species richness patterns without being mutually exclusive. There may be a link between the two theories: the change in species richness along the productivity gradient in plots of a certain size may be accompanied, and perhaps explained, by the change in the number of species potentially able to grow at a par- ticular level of productivity in a given region. In other words, if the hump-shaped pattern holds true in the de- ciduous forests studied here, it may be partly the result of the small species pools capable of occurring in low- or high-productive forest sites. This leads us to question if the specific relationship between species richness and productivity can be observed at all spatial scales. Our study was conducted in deciduous forests of two separate Swedish regions, ��land and Uppland. We ad- dressed the following questions: - Are there differences in species number between the two regions and can these be observed at all spatial scales? Is there a general correlation between large- scale species richness and small-scale species richness i.e. can the above-mentioned differences be explained by differences in the regional species pool? - How does the small-scale (i.e. within-regional) relationship between species richness and productivity at different scales look? - Does the heterogeneity hypothesis offer a mecha- nistic explanation of the species richness-productivity relationship and can the observed relationship be ex- plained in terms of the species pool theory? Material and Methods Vegetation and data sets The study is based on two data sets. The first data set was sampled primarily to tackle the species pool hypothesis and consists of vegetation data collected in 1995-1996 from 36 deciduous forest sites located in two Swedish provinces (Fig. 1): the island of ��land at the southeastern fringe of the Boreo-nemoral zone (22 sites), and Uppland at its northern border close to the plant geographical boundary limes norrlandicus (14 sites). Deciduous forests on low-productive sites are absent from Uppland, as these sites are dominated by coniferous trees due to the prevailing climate (Diekmann 1994). As we did not want to distort the analysis by includ- ing two vegetation types different in structure and species composition, this data set only included sites of moderate to high productivity. The forest sites, ranging in size from 0.25 to 25 ha, were usually char- acterised by a mixture of deciduous trees and shrubs and a high species richness in the field layer (for a detailed description, see Diekmann 1994, 1999). Fig. 1. Map of Scandinavia. The two study regions are indi- cated in bold. Regions from which data were compiled for Table 6 and Fig. 5 are also outlined.