Content uploaded by Jochen Krauss
Author content
All content in this area was uploaded by Jochen Krauss
Content may be subject to copyright.
Effects of habitat area, isolation, and landscape
diversity on plant species richness of calcareous
grasslands
JOCHEN KRAUSS*, ALEXANDRA-MARIA KLEIN,
INGOLF STEFFAN-DEWENTER and TEJA TSCHARNTKE
Department of Agroecology, University of Go
¨
ttingen, Waldweg 26, D-37073 Go
¨
ttingen, Germany;
*Author for correspondence (e-mail: j.krauss@uaoe.gwdg.de; fax: +49-551-398806)
Received 18 November 2002; accepted in revised form 15 May 2003
Key words: Conservation, Generalists, Habitat fragmentation, Specialists, Species density, Species–area
relationships
Abstract. Calcareous grasslands harbour a high biodiversity, but are highly fragmented and endangered
in central Europe. We tested the relative importance of habitat area, habitat isolation, and landscape
diversity for species richness of vascular plants. Plants were recorded on 31 calcareous grasslands in the
vicinity of the city of Go
¨
ttingen (Germany) and were divided into habitat specialist and generalist
species. We expected that habitat specialists were more affected by area and isolation, and habitat
generalists more by landscape diversity. In multiple regression analysis, the species richness of habitat
specialists (n ¼ 66 species) and habitat generalists (n ¼ 242) increased with habitat area, while habitat
isolation or landscape diversity did not have significant effects. Contrary to predictions, habitat spe-
cialists were not more affected by reduced habitat area than generalists. This may have been caused by
delayed extinction of long-living plant specialists in small grasslands. Additionally, non-specialists may
profit more from high habitat heterogeneity in large grasslands compared to habitat specialists. Although
habitat isolation and landscape diversity revealed no significant effect on local plant diversity, only an
average of 54% of habitat specialists of the total species pool were found within one study site. In
conclusion, habitat area was important for plant species conservation, but regional variation between
habitats contributed also an important 46% of total species richness.
Introduction
Habitat loss and habitat fragmentation of natural and semi-natural habitats are
considered as major threats to biodiversity. Semi-natural calcareous grasslands
belong to the most species-rich plant and insect habitats in central Europe (Van
Swaay 2002; WallisDeVries et al. 2002). Habitat loss of calcareous grasslands in
Germany reached locally up to 60% in the last 100 years. Thereby, 1% of the
plant species restricted to this habitat type became extinct and 42% are threa-
tened in Germany (WallisDeVries et al. 2002). Calcareous grasslands are highly
fragmented, endangered, and nowadays protected by law in Germany (Riecken
et al. 1994). Interest in conservation is high (Beinlich and Plachter 1995;
WallisDeVries et al. 2002), leading to several studies on plant and insect com-
munity structure and extinction risks (e.g. Fischer and Sto
¨
cklin 1997; Bruun
2000; Zschokke et al. 2000; Krauss et al. 2003). However, a recent review
# 2004 Kluwer Academic Publishers. Printed in the Netherlands.
Biodiversity and Conservation 1427–1439, 2004.
13:
emphasises the lack of well-replicated community level studies on calcareous
grasslands (Steffan-Dewenter and Tscharntke 2002).
Habitat fragmentation is a major threat for plant species richness (e.g. Ouborg 1993;
Grashof-Bokdam 1997; Honnay et al. 1999; Bruun 2000). The influence of landscape
context in habitat fragmentation studies is often ignored (Wiens 1997; Hanski 1999),
although it is essential to understand plant community dynamics (De Blois et al. 2002).
Habitat isolation might be seen as a simple measure of landscape context, as it de-
scribes the distance to neighbouring habitats or the proportion of the habitats of a
certain type within a landscape (see Moilanen and Nieminen 2002). But landscape
context includes more factors that may affect local communities, as there is the pro-
portion of other habitat types or the diversity of habitats in the landscape matrix
(Steffan-Dewenter et al. 2002). Similar habitats and high habitat diversity in the sur-
roundings might enhance species richness of a local site. Landscape context might
affect gene flow via pollen and seed dispersal, the pressure of herbivory, and the
functional connectivity between habitat patches (Ricketts 2001). Recently published
studies started to focus on the complex effects of landscape context on local plant
species richness (Kollman and Schneider 1999; Metzger 2000; So
¨
derstro
¨
m et al. 2001).
Not all species depend on habitat area, isolation and landscape context equally
(Tscharntke et al. 2002). (1) Habitat specialists are more affected by habitat loss
than generalists, (Warren et al. 2001). (2) The surrounding landscape is inhabitable
for habitat specialists, but at least partly habitable for generalists, supporting the
prediction that habitat isolation affects habitat specialists more than generalists (see
Jonsen and Fahrig 1997). (3) High landscape diversity in the surrounding matrix
provides more different habitat types for generalists or species with other habitat
preferences, supporting the prediction that landscape diversity enhances the number
of generalists, especially at edges, but hardly specialists (see Jonsen and Fahrig
1997). These predictions have rarely been tested for plant communities in the
context of spatial dynamics. True forest species showed a stronger response to
habitat area than other species, but no response to habitat isolation (Honnay et al.
1999), and shade tolerant species were more sensitive to habitat fragmentation than
shade intolerant species (Metzger 2000). Numerous functional groups, including
specialist grassland species, were positively related to habitat area and incon-
sistently to isolation (Bruun 2000).
Species–area relationships can be explained by two ecological hypotheses.
(1) The habitat heterogeneity hypothesis predicts higher species numbers because
of higher habitat heterogeneity. (2) The area per se or equilibrium hypothesis
considers colonisation–extinction dynamics to cause increasing species numbers
with increasing habitat area independent of habitat heterogeneity (Rosenzweig
1995). Habitat area and habitat heterogeneity are often closely correlated (Kohn
and Walsh 1994; Rosenzweig 1995; Ko
¨
chy and Rydin 1997). Equal sample sizes
should reduce the habitat heterogeneity effect, making it possible to test separately
for area per se effects (Kelly et al. 1989; Ko
¨
chy and Rydin 1997).
In this study we tested the impact of the three parameters habitat area, habitat
isolation, and landscape diversity on the species richness of habitat specialist and
generalist vascular plants on calcareous grasslands.
1428
We tested the following predictions:
(1) Plant species richness increases with increasing habitat area, decreasing habitat
isolation and increasing landscape diversity.
(2) Species richness of habitat specialists is more sensitive to habitat area and
isolation, whereas species richness of generalists is more sensitive to landscape
diversity.
(3) Species density, that is, plant species richness per area in samples of equal size
on all study sites, is related to habitat area, isolation and landscape diversity.
Materials and methods
Study region and study sites
Altogether 31 calcareous grasslands in the vicinity of the city of Go
¨
ttingen in
Lower Saxony (Germany) were studied. The grasslands were located in the Leine-
Weser Mountain lands (Gauss-Kru
¨
ger: R ¼ 5695, H ¼ 3550=R ¼ 5724, H ¼ 3579)
and belong to the plant association Gentiano-Koelerietum. Low impact manage-
ment to stop succession and to remove woody bushes was carried out on the study
sites mainly from late summer to winter. The average rainfall in the area around
Go
¨
ttingen is 635 mm per annum, with an average temperature of 6.8 8C (Deutscher
Wetterdienst 2001). The landscape is structurally rich with a mosaic of diverse
habitat types. Calcareous grasslands can be sharply delimited from the surrounding
landscape with little or no ambiguity and cover 0.26% of the study region.
Habitat area, isolation, landscape diversity
The area of the 31 calcareous grasslands was measured with a differential GPS
GEOmeter 12L (GEOsat GmbH 1998) and ranged from 314 to 51395 m
2
.Area
covered with shrubs was excluded from the measurement.
Habitat isolation was measured as an index (I) of each study site (i) from edge to
edge on the basis of all known calcareous grasslands in a radius of 8 km around our
study sites using the following formula:
I ¼ e
dij
A
j
where A
j
is the size (in m
2
) of neighbouring calcareous grasslands and dij the
distance (in km) from the neighbouring grassland j to the study site i. The formula
is based on Hanski et al. (1994). Larger values of I indicate lower isolation than
smaller values. Additionally, habitat isolation was measured as distance from edge
to edge to the nearest calcareous grassland (55–1894 m). This isolation distance and
the isolation index were always log
10
transformed. Both were correlated
(r ¼0.445, n ¼ 31, P ¼ 0.012) and showed similar, always non-significant results.
Therefore we show only the results from the isolation index.
1429
Landscape diversity was analysed using modified digital thematic maps
(ATKIS
1
-DLM 25=1 Landesvermessung and Geobasisinformationen Niedersachsen
1991–1996, and ATKIS
1
-DLM 25/2 Hessisches Landesvermessungsamt 1996).
Eleven land-use types were defined, including arable land (42.15% of the study re-
gion), forest (36.80%), grassland (12.14%), built-up area (6.24%), other habitats
(1.48%), garden land (0.31%), hedgerow (0.30%), calcareous grassland (0.26%),
orchard meadow (0.20%), plantation (0.06%), and fen (0.05%). We used the Shannon–
Wiener index to calculate landscape diversity for each of the 31 grassland fragments
using a nested set of 12 circles with a radius ranging from 0.25 to 3.00 km in 0.25 km
steps:
Hs ¼p
i
ln p
i
where p
i
is the proportion of each of the 11 different land-use types (Krebs 1989).
Further we pooled the plant species-rich habitat types, grassland, garden land,
hedgerow, calcareous grassland, orchard meadow, and fen around the study sites.
We related the different plant species richness data with the proportion of this
pooled habitat type data set. We never found any positively significant relation
between both, and the pooled habitat types were highly correlated with landscape
diversity (250 m scale: r ¼ 0.685, P<0.0001). Therefore we only show results from
landscape diversity.
Due to some missing ATKIS data we could test the landscape data for a radius of
0.25 and 0.50 km for all 31 study sites, while 0.75–2.00 km scales were tested for
only 30, 2.25 km for 29, and 2.50–3.00 km for 28 sites. For each landscape analysis,
the habitat area of the central study site was excluded.
Plants
The complete survey of vascular plants (Spermatophyta plus Equisetum arvense L.)
for all of the 31 calcareous grasslands was compiled from four independent data
sets from 1996, 2000 and 2001 to achieve a total list of plant species per study site.
In 1996 and 2000 the vegetation was mapped in May=June and again in August in
randomised plots of 25 m
2
according to Braun-Blanquet. These plots were the same
for both records in 1996, while they changed in 2000. In 1996 one plot was mapped
for small (>1500 m
2
), two plots for medium (1500–5000 m
2
), three plots for
medium to large (5000 –1 0000 m
2
) and four plots for large grasslands (>10 000 m
2
)
(see Steffan-Dewenter and Tscharntke 2000). In 2000 one plot was mapped on
small grasslands (<1500 m
2
), two plots on medium grasslands (1500–10000 m
2
),
and three plots on large grasslands (>10000 m
2
). To complete these two Braun-
Blanquet surveys, in 2000 between April and August we mapped five times the
plant species in flower of each grassland on randomised transects. Each transect
covered from 11.4% (largest habitat) to 100.0% (small habitats) of the total habitat
area (average: 62.7 0.4%). In a fourth survey in May 2001, again plants were
mapped on the 31 grasslands to complete the species list.
1430
We used the total number of species per grassland for statistical analyses. To
compare equal sample sizes (species density) we also tested all correlations using
only one 25 m
2
plot (mapped in spring and summer) of the 2000 data set.
All vascular plant species recorded as species restricted to oligotrophic grasslands
in Von Drachenfels (1994) were defined as habitat specialists (n ¼ 66) for calcareous
grasslands (see Appendix 1). These species mainly occur on calcareous grasslands in
the study region, but might inhabit other habitat types in other regions of Germany
and Europe. All other species with no habitat preferences or preferences for other
habitats were defined as generalist plants (242 species). Species identification and
nomenclature follow Ba
¨
ssler et al. (1999).
Statistical analyses
The statistical analyses of the data were performed using the software ‘Statgraphics
Plus for Windows 3.0’ (Statgraphics 1995). All data were tested on whether they
satisfy the assumption of normality. We calculated simple and multiple regressions,
Pearson correlations, and comparisons of regression lines (Sokal and Rohlf 1995).
We chose backward selections for stepwise multiple regressions. The independent
variables habitat area and isolation were always log
10
transformed. Species num-
bers in regressions were also log
10
transformed to calculate scale-independent
slopes (z-values) for comparison with other studies. Arithmetic means one
standard error are given in the text.
Results
Habitat area (0.03 – 5.14 ha, average: 0.90 0.23 ha) was not correlated with ha-
bitat isolation (index: 2051 – 85 978, average: 23 019 3368) (r ¼0.013,
P ¼ 0.944). Also landscape diversity (at a 250 m scale; Shannon–Wiener: 0.09–
1.56, average: 1.09 0.05) was not correlated with habitat isolation (r ¼0.015,
P ¼ 0.936), but with habitat area (r ¼ 0.375, P ¼ 0.038).
Altogether 308 plant species were identified on all 31 calcareous grasslands with
pooled data. This was an average of 89.0 3.7 species per habitat with a minimum of
56 species and a maximum of 138 species. In Braun-Blanquet plots we found 192
species in 1996 and 234 species in 2000, resulting in a total of 278 species. Additional
recordings due to the transect walks in 2000 and the survey in 2001 resulted in a total
of 308 species. Nine of 66 (13.6%) specialist species (see Appendix 1) and seven out
of 242 (2.9%) habitat generalist species (Achillea millefolium L., Campanula ro-
tundifolia L., Daucus carota L., Leontodon hispidus L., Lotus corniculatus L.,
Plantago lanceolata L., and Plantago media L.) were found on each of the 31
calcareous grasslands. A minimum of 33% and a maximum of 67% of habitat spe-
cialists of the 66 specialists were found within one study site. On average, per study
site 54% of all specialist species was found, emphasising that 46% of total species
richness was due to between-habitat variation (see Veech et al. (2002) for calculations
of b-diversity).
1431
Species richness increased with increasing habitat area for both habitat specialist
and generalist species, leading to an increase of all species (Figure 1, Table 1). The
z-value (slope of log–log regressions) for specialists (z-value: 0.08) was lower than
Figure 1. Relationship between the number of plant species and grassland area (n ¼ 31 fragments). (A)
Specialist plant species (66 species): y ¼ 13.64 þ 6.32 log
10
x. (B) Generalist plant species (242 species):
y ¼1.40 þ 15.44 log
10
x. For statistics see Table 1. Comparison of regression lines: slopes: F ¼ 5.34,
P ¼ 0.024.
1432
Table 1. Simple regressions for the habitat specialists among the plant species (66 species), generalist species (242 species) and all plant species (308 species).
Relations between species numbers and species density with the three independent factors habitat area, habitat isolation, and landscape diversity (n ¼ 31). Habitat area
and habitat isolation are log
10
transformed.
Habitat area Habitat isolation Landscape diversity (radius: 250 m)
Fr P Fr PFr
P
Total species numbers
Habitat specialists 23.49 0.669 <0.0001 0.39 0.116 0.536 3.81 0.341 0.061
Habitat generalists 17.18 0.610 0.0003
0.95 0.178 0.337 0.16 0.074 0.692
All plants 24.67 0.678 <0.0001 1.04 0.186 0.316 0.76 0.160 0.390
Species density
Habitat specialists 0.09 0.054 0.773
0.39 0.116 0.535 0.00 0.008 0.965
Habitat generalists 0.02 0.028 0.879 0.00 0.012 0.950 0.00 0.004 0.981
All plants 0.06 0.046 0.807 0.11 0.061 0.744 0.00 0.000 0.999
1433
for generalists (z-value: 0.13), but did not differ significantly (comparison of re-
gression lines F ¼ 2.16, P ¼ 0.148). The slope for not log-transformed species
numbers was even significantly steeper for generalists (Figure 1). For all plant
species the z-value was 0.11. Habitat isolation showed no significant effect on
species richness in simple regressions, while landscape diversity at the smallest
scale of 250 m was positively related to specialist species with marginal sig-
nificance (Table 1). We analysed the effects of landscape diversity at 12 nested
spatial scales. Correlations were highest at the smallest spatial scale of 250 m for
specialist, generalist and all plants (results for other spatial scales are not shown).
In multiple regressions with habitat area, isolation and landscape diversity
(250 m), habitat area was the only significant factor explaining 44.8% of variation
for specialists, 37.2% for generalists and 46.0% for all plant species.
Considering only equal sample size with one plot per grassland, we found al-
together 199 plant species. This was an average of 46.5 1.4 species per habitat
with a minimum of 30 species and a maximum of 63 species. Neither in simple
(Table 1) nor in multiple regressions we found evidence for an impact of habitat
area, isolation, and landscape diversity on plant species density for specialists,
generalists or all plants on the studied 31 calcareous grasslands.
Discussion
Landscape diversity is known to increase generalist insect species numbers (Jonsen
and Fahrig 1997; Krauss et al. 2003), but did not affect the number of generalist
plants in our study. Increasing landscape diversity even tended to increase specialist
plant species numbers, but due to correlation with habitat area this effect was
eliminated in multiple regressions. We also could not detect an effect of landscape
on total plant species richness. This might be explained by the relatively complex
landscapes that surrounded our calcareous grasslands. In multiple tests for different
landscape parameters and numerous functional groups, inconsistent results of
correlations of plant species richness with landscape diversity were reported
(Kollman and Schneider 1999; Metzger 2000). To reduce multiple testing and
intercorrelations of landscape factors, Steffan-Dewenter et al. (2002) suggest fo-
cusing on landscape diversity and proportion of habitats in the surrounding matrix.
So
¨
derstro
¨
m et al. (2001) found lower plant species numbers on semi-natural pas-
tures with increasing proportion of arable fields in the surrounding landscape, but
landscape diversity was not studied. In general, landscape studies on plant species
composition are still rare, but essential to fully understand community dynamics
(De Blois et al. 2002).
Species numbers increased significantly with increasing habitat area for both ha-
bitat specialist and habitat generalist plant species. Our results confirm the general
validity of species–area relationships, as shown before for plant communities (e.g.
Ouborg 1993; Grashof-Bokdam 1997; Honnay et al. 1999; Bruun 2000). Our high
sampling effort appeared to guarantee almost complete recordings. In addition, our
numbers of plant species on 27 of the grasslands were very similar to a complete
1434
survey made between 1980 and 1986 (Eggers 1986, unpublished data; B. Preuschoff,
personal communication). We did not measure habitat heterogeneity, but species
density, that is, species numbers measured with equal sample sizes on all grasslands,
was not affected by habitat area in our study, giving indirect evidence for the habitat
heterogeneity hypothesis as an explanation for species–area relations (Kelly et al.
1989; Holt et al. 1999). Habitat heterogeneity is assumed to be the main predictor for
plant species richness (Ko
¨
chy and Rydin 1997), as neither Ko
¨
chy and Rydin (1997)
nor Lawesson et al. (1998) found positive species–area relations for area-corrected
sample sizes.
Surprisingly, habitat specialists were not more affected by habitat area than gen-
eralists. This is in contrast to previous findings for specialist and generalist plant
species, where the z-value was higher for true forest plant species (z ¼ 0.45) than for
species of edges and clearings (z ¼ 0.39) and for woody species and lianas (z ¼ 0.17)
in Belgium forests (Honnay et al. 1999). Also habitat specialist butterflies on the
same calcareous grasslands were significantly more affected by habitat loss than
generalists (Krauss et al. 2003). Large areas with high habitat heterogeneity can be
expected to offer a diverse mosaic of microhabitat types for species that are non-
specialists. This may explain the steeper species–area relationships for generalists in
our study. Habitat specialists are expected to have higher extinction rates than gen-
eralists, as shown for plants and butterflies (Fischer and Sto
¨
cklin 1997; J. Krauss,
unpublished data). However, when these specialists are perennial plants, they may
persist for several years in small populations, thereby delaying their extinction
(Oostermeijer et al. 1994). An extinction debt for these species in small habitats keeps
the z-value lower than in equilibrium situations (Tilman et al. 1994; Gonzales 2000).
Hypotheses developed for mobile birds and butterflies with short life cycles might be
not applicable to sessile plants with long life cycles (Eriksson 1996).
Contrary to our predictions, we could not find effects of habitat isolation, neither
for specialist nor for generalist plant species. Isolation effects on plant species are
mainly found for plant species with low dispersal abilities (Van Ruremonde and
Kalkhoven 1991; Grashof-Bokdam 1997), while other species groups generally
show no or weak isolation effects (Ouborg 1993; Ko
¨
chy and Rydin 1997; Honnay
et al. 1999; Bruun 2000). Also for the more mobile butterfly species no isolation
effects could be reported on the same calcareous grasslands (Steffan-Dewenter and
Tscharntke 2000; Krauss et al. 2003). As calcareous grasslands were better con-
nected decades ago (WallisDeVries et al. 2002), we might have found patterns that
show previous situations because of extinction debt of plants, and therefore no
isolation effect.
In conclusion, habitat area was the only predictor explaining vascular plant com-
munity structure for habitat specialists, generalists, and all plant species. Per study site
an average of 54% of all specialist species was found, indicating that 46% of total
species richness was due to between study site variation. Nevertheless, habitat iso-
lation and landscape diversity had no effect in our study region. These results stress the
point that habitat area and increasing habitat heterogeneity are the most important
basis of plant diversity. Several populations of specialist plant species on small habitat
fragments may be prone to extinction in the near future, showing the so-called
1435
extinction debt. Sheep flock migration might contribute to a better dispersal of seeds
(Poschlod and WallisDeVries 2002), to reduce potentially increasing specialist ex-
tinction in small and isolated habitats. For conservation we suggest to protect (a) the
largest grasslands with low extinction risk of habitat specialists, and (b) a series of
habitat fragments covering a sufficient range of geographical area to maximise re-
gional diversity.
Acknowledgements
We thank David Kleijn, Olivier Honnay, Sabine Gu
¨
sewell and one anonymous
referee for helpful comments on the manuscript and Bertram Preuschoff
(Untere Naturschutzbeho
¨
rde des Landkreis Go
¨
ttingen) for his expert advice. This
work was financially supported by the German Science Foundation (Deutsche
Forschungsgemeinschaft).
Appendix 1
Habitat specialist plants on 31 calcareous grasslands in Southern Lower Saxony.
Taxon Number
of
occupied
habitats
Taxon Number
of
occupied
habitats
Agrimonia eupatoria L. 30 Koelaeria pyramidata (Lamk.) P.B. 29
Anemone sylvestris L. 3 Medicago falcata L. 7
Anthyllis vulneraria L. 20 M. lupulina L. 30
Astragalus glycyphyllos L. 10 Melampyrum arvense L. 3
Avenula pratensis (L.) Dum. 16 M. nemorosum L. 2
A. pubescens (HUDS.) Dum. 11 Onobrychis viciifolia Scop. 7
Brachypodium pinnatum (L.) P.B. 31 Ononis spinosa L. 24
Briza media L. 30 Ophris apifera Huds. 1
Bromus erectus Huds. 19 Ophrys insectifera L. em L. 16
Campanula rapunculoides L. 24 Orchis mascula (L.) L. 13
Carex caryophyllea Latourr. 15 O. militaris L. 3
C. flacca Schreber 30 O. purpurea Huds. 5
C. ornithopoda Willd. 3 Orchis tridentata Scop. 3
Carlina vulgaris L. 25 Origanum vulgare L. 3
Centaurea scabiosa L. 28 Pimpinella saxifraga L. 31
Cerastium arvense L. 11 Platanthera bifolia (L.) L.C. Richard 2
Cirsium acaule SCOP. 31 Polygala comosa Schkuhr 27
Clinopodium vulgare L. 15 Potentilla neumanniana Rchb. 31
Euphorbia cyparissias L. 14 Primula veris L. 30
Euphrasia stricta Wolff ex Lehm 2 Prunella grandiflora (L.) Scholler 11
Festuca ovina L. 31 Ranunculus bulbosus L. 31
Filipendula vulgaris Moench 1 Salvia pratensis L. 4
Fragaria viridis (Duchesne) Weston 18 Sanguisorba minor Scop. 31
Galium pumilum Murray 22 Scabiosa columbaria L. 31
1436
References
Ba
¨
ssler M., Ja
¨
ger E.J. and Werner K. 1999. Rothmaler, Exkursionsflora von Deutschland, Bd. 2,
Gefa
¨
sspflanzen: Grundband. Spektrum Akademischer Verlag, Heidelberg, Berlin, Germany.
Beinlich B. and Plachter H. 1995. Ein Naturschutzkonzept fu
¨
r Kalkmagerrasen der Mittleren
Schwa
¨
bischen Alb (Baden-Wu
¨
rttemberg): Schutz, Nutzung und Entwicklung. Beihefte zu den
Vero
¨
ffentlichungen fu
¨
r Naturschutz und Landschaftspflege in Baden–Wu
¨
rttemberg 83: 1–520.
Bruun H.H. 2000. Patterns of species richness in dry grassland patches in an agricultural landscape.
Ecography 23: 641–650.
De Blois S., Domon G. and Bouchard A. 2002. Landscape issues in plant ecology. Ecography 25: 244–256.
Deutscher Wetterdienst 2001. Klimadaten der Wetterstation Go
¨
ttingen. In: Stadt Go
¨
ttingen, Amt fu
¨
r
Statistik und Stadtforschung (eds), Statistikjournal 1/2001.
Eriksson O. 1996. Regional dynamics of plants: a review of evidence for remnant, source-sink and
metapopulations. Oikos 77: 248–258.
Fischer M. and Sto
¨
cklin J. 1997. Local extinctions of plants in remnants of extensively used calcareous
grasslands 1950–1985. Conservation Biology 11: 727–737.
GEOsat GmbH 1998. GEOlink Version 1. 46. Wuppertal, Germany.
Gonzales A. 2000. Community relaxation in fragmented landscapes: the relation between species
richness, area and age. Ecology Letters 3: 441–448.
Grashof-Bokdam C. 1997. Forest species in an agricultural landscape in the Netherlands: effects of
habitat fragmentation. Journal of Vegetation Science 8: 21–28.
Hanski I. 1999. Metapopulation Ecology. Oxford University Press, Oxford, UK.
Hanski I., Kuussaari M. and Nieminen M. 1994. Metapopulation structure and migration in the butterfly
Melitaea cinxia. Ecology 75: 747–762.
Holt R.D., Lawton J.H., Polis G.A. and Martinez N.D. 1999. Trophic rank and species–area relationship.
Ecology 80: 1495–1504.
Honnay O., Endels P., Vereecken H. and Hermy M. 1999. The role of patch area and habitat diversity in
explaining native plant species richness in disturbed suburban forest patches in northern Belgium.
Diversity and Distribution 5: 129–141.
Jonsen I.D. and Fahrig L. 1997. Response of generalist and specialist insect herbivores to landscape
spatial structure. Landscape Ecology 12: 185–197.
Kelly B.J., Bastow W.J. and Mark A.F. 1989. Causes of the species–area relation: a study of islands in
lake Manapouri, New Zealand. Journal of Ecology 77: 1021–1028.
Ko
¨
chy M. and Rydin H. 1997. Biogeography of vascular plants on habitat islands, peninsulas and mainlands
in an east-central Swedish agricultural landscape. Nordic Journal of Botany 17: 215–223.
Appendix 1. (continued)
Taxon Number
of
occupied
habitats
Taxon Number
of
occupied
habitats
G. verum L. 25 Sedum sexangulare L. 6
Gentianella ciliata (L.) Borkh. 10 Silene nutans L. 3
G. germanica (Willd.) Bo
¨
rner 9 Thymus pulegioides L. 31
Gymnadenia conopsea (L.) R.Br. 26 Trifolium campestre Schreber 7
Helianthemum nummularium (L) 20 T. medium L. 18
Hieracium pilosella L. 29 T. montanum L. 2
Hippocrepis comosa L. 25 Veronica teucrium L. 17
Hypericum perforatum L. 30 Vincetoxicum hirundinaria Med. 3
Inula conyzae (Griesselich) Meikle 12 Viola hirta L. 26
1437
Kohn D.D. and Walsh D.W. 1994. Plant species richness – the effect of island size and habitat diversity.
Journal of Ecology 82: 367–377.
Kollmann J. and Schneider B. 1999. Landscape structure and diversity of fleshy-fruited species at forest
edges. Plant Ecology 144: 37–48.
Krauss J., Steffan-Dewenter I. and Tscharntke T. 2003. How does landscape context contribute to effects
of habitat fragmentation on diversity and population density of butterflies? Journal of Biogeography
30: 889–900.
Krebs C.J. 1989. Ecological Methodology. Harper Collins Publishers, New York.
Lawesson J.E., de Blust G., Grashof C., Firbank L., Honnay O., Hermy M., Hobitz P. and Jensen L.M.
1998. Species diversity and area-relationships in Danish beech forests. Forest Ecology and Manage-
ment 106: 235–245.
Metzger J.P. 2000. Tree functional group richness and landscape structure in a Brazilian tropical frag-
mented landscape. Ecological Applications 10: 1147–1161.
Moilanen A. and Nieminen M. 2002. Simple connectivity measures in spatial ecology. Ecology 83:
1131–1145.
Oostermeijer J.G.B., Vant ’t Veer R. and Den Nijs J.C.M. 1994. Population structure of the rare, long-
lived perennial Gentiana pneumonanthe in relation to vegetation and management in the Netherlands.
Journal of Applied Ecology 31: 428–438.
Ouborg N.J. 1993. Isolation, population size and extinction: the classical and metapopulation approaches
applied to vascular plants along the Dutch Rhine-system. Oikos 66: 298–308.
Poschlod P. and WallisDeVries M.F. 2002. The historical and socioeconomic perspective of calcareous
grasslands – lessons from the distant and recent past. Biology Conservation 104: 361–376.
Ricketts T.H. 2001. The matrix matters: effective isolation in fragmented landscapes. The American
Nationalist 158: 87–99.
Riecken U., Ries U. and Ssymank A. 1994. Rote Liste der gefa
¨
hrdeten Biotoptypen der Bundesrepublik
Deutschland. Kilda-Verlag, Greven, Germany.
Rosenzweig M.L. 1995. Species diversity in space and time. Cambridge University Press, Cambridge,
UK.
So
¨
derstro
¨
m B., Svensson B., Vessby K. and Glimska
¨
r A. 2001. Plants, insects and birds in semi-natural
pastures in relation to local habitat and landscape factors. Biodiversity and Conservation 10: 1839–1863.
Sokal R.R. and Rohlf F. 1995. Biometry. Freeman, New York.
Statgraphics 1995. Statgraphics plus for Windows, Version 3.0. Manugistics, Inc, Rockville, Mary land.
Steffan-Dewenter I. and Tscharntke T. 2000. Butterfly community structure in fragmented habitats.
Ecology Letters 3: 449–456.
Steffan-Dewenter I. and Tscharntke T. 2002. Insect communities and biotic interactions on fragmented
calcareous grasslands – a mini review. Biological Conservation 104: 275–284.
Steffan-Dewenter I., Mu
¨
nzenberg U., Bu
¨
rger C., Thies C. and Tscharntke T. 2002. Scale-dependent
effects of landscape context on three pollinator guilds. Ecology 83: 1421–1432.
Tilman D., May R.M., Lehman C.L. and Nowak M.A. 1994. Habitat destruction and the extinction debt.
Nature 371: 65–66.
Tscharntke T., Steffan-Dewenter I., Kruess A. and Thies C. 2002. Characteristics of insect populations on
habitat fragments: a mini review. Ecological Research 17: 229–239.
Van Ruremonde R.H.A.C. and Kalkhoven J.T.R. 1991. Effects of woodlot isolation on the dispersion of
plants with fleshy fruits. Journal of Vegetation Science 2: 377–384.
Van Swaay C.A.M. 2002. The importance of calcareous grasslands for butterflies in Europe. Biological
Conservation 104: 315–318.
Veech J.A., Summerville K.S., Crist T.O. and Gering J.C. 2002. The additive partitioning of species
diversity: recent revival of an old idea. Oikos 99: 3–9.
Von Drachenfels O. 1994. Kartierschlu
¨
ssel fu
¨
r Biotoptypen in Niedersachsen. Naturschutz und Land-
schaftspflege Niedersachsen 4: 1–192.
WallisDeVries M.F., Poschlod P. and Willems J.H. 2002. Challenges for the conservation of calcareous
grasslands in northwestern Europe: integrating the requirements of flora and fauna. Biological
Conservation 104: 265–273.
1438
Warren M.S., Hill J.K., Thomas J.A., Asher J., Fox R., Huntley B. et al. 2001. Rapid responses of British
butterflies to opposing forces of climate and habitat change. Nature 414: 65–69.
Wiens J.A. 1997. Metapopulation dynamics and landscape ecology. In: Hanski I. andGilpin M.E. (eds)
Metapopulation Biology: Ecology, Genetics, and Evolution. Academic Press, London, pp. 43–62.
Zschokke S., Dolt C., Rusterholz H.-P., Oggier P., Braschler B., Thommen G.H. et al. 2000. Short-term
responses of plants and invertebrates to experimental small-scale grassland fragmentation. Oecologia
125: 559–572.
1439