Ecology Letters, (2007) 10: 16–24

doi: 10.1111/j.1461-0248.2006.00993.x

LETTER

Scale dependence of the correlation between human population presence and vertebrate and plant species richness

Marco Pautasso Division of Biology, Imperial College London, Wye Campus, High Street, Wye, Kent TN25 5AH, UK *Correspondence: E-mail: [email protected]

1

Abstract Human presence is generally negatively related to species richness locally, but the relationship is positive at coarse scales. An increase in the strength of the latter correlation with increasing study resolution has been documented within studies, but it is not known whether such a scale dependence is present across different studies. We test this with data on the spatial co-occurrence of human beings and the species richness of plants and vertebrates from a continuum of scales. The correlation coefficient between human presence and species richness is positively related to study grain and extent. The correlation turns from positive to negative below a study grain of c. 1 km and below a study extent of c. 10 000 km2. The broad-scale positive correlation between human presence and species richness suggests that people have preferentially settled and generally flourished in areas of high biodiversity and/or have contributed to it with species introductions and habitat diversification. The scale dependency of the correlation between people and biodiversity’s presence emphasizes the importance of the preservation of green areas in densely populated regions. Keywords Landscape ecology, macroecology, plot size, reserve selection, sampling area, scaling, spatial covariance, species energy, study resolution, urban–rural gradients. Ecology Letters (2007) 10: 16–24

INTRODUCTION

As human beings, we are one of the most novel forces in the evolution of life. We have an ambiguous relationship with nature. Nature, when cultivated, sustains humanity. And humanity simultaneously degrades, studies and enhances nature. On the one hand, species introductions, pollution, land use and climate change cause species endangerment and extinctions. On the other hand, conservation activities document and preserve rare ecosystems, hot spots of biodiversity and species present in urban green areas. The magnitude of urbanization, and of its impacts on ecosystems, is increasing, if only because of the unprecedented numbers of people involved. Also because of this, the extent to which people and biodiversity spatially co-occur has been lately the object of an increased amount of attention in urban ecology, biogeography and macroecology (e.g. Pickett et al. 2001; Gaston 2005; Hansen et al. 2005). On the one hand, there are many reports of a positive correlation between human presence and the species  2006 Blackwell Publishing Ltd/CNRS

richness of various taxa at a coarse scale, with sample units ranging from province to country, and with resolution typically between quarter- and one-degree cells. Examples include studies in sub-Saharan Africa (Balmford et al. 2 2001; Chown et al. 2003), Asia (Nepal: Hunter & Yonzon 3 1993), Australia (Luck et al. 2004), Europe (e.g. Arau´jo 2003; Evans & Gaston 2005), North America (e.g. McKinney 2002; Va´zquez & Gaston 2006) and South America (Real et al. 2003; Diniz-Filho et al. 2006). Taxa thus analysed include plants, amphibians, reptiles, birds and mammals. In a nutshell, these findings imply that regions of high biodiversity and human population density basically coincide. Humans have become more numerous and culturally diverse in regions of milder climate, higher productivity and hence higher species richness, therefore competing with plants and wildlife for resources (e.g. Sutherland 2003; Gaston 2005; Maffi 2005). On the other hand, there are many findings of a negative correlation between increasing human population density and species richness in a number of studies at a local scale, again for a variety of taxa. Recent

Letter

examples include investigations of bryophytes in Wroclaw, Poland (Fudali 2001), the flora of Rome, Italy (Ricotta et al. 2001), tree species in Guangzhou City, China (Jim 2002), invertebrates in streams north of Washington, DC, USA (Moore & Palmer 2005), eusocial bees and wasps in Belo Horizonte, Brazil (Zanette et al. 2005), ants, e.g. in Coatepec, Mexico (Lopez-Moreno et al. 2003), birds, e.g. in Peninsular Malaysia (Soh et al. 2006) and in towns in Scandinavia, France and Italy (e.g. Clergeau et al. 2006a), and small mammals around Chernogolovka, Russia (Tikhonova et al. 2006). There are, however, many more examples of such a negative influence on species richness of increasing urbanization at local scales, a pattern that has been widely reported since the 1960s, when urban ecology started to become of pressing relevance due to escalating urban sprawl 4 (e.g. Erz 1964; Tomiałojc´ 1970; Beissinger & Osborne 1982; see also studies listed in Clergeau et al. 2001). Although there are exceptions [a positive correlation between people’s and biodiversity’s presence has been reported also at relatively fine resolutions: e.g. flora in a urban–agricultural comparison near Halle, Germany (Wania et al. 2006), birds in the Pampean region near Buenos Aires, Argentina (Leveau & Leveau 2005) and in the Czech Republic (Kubes & Fuchs 1998)], it seems therefore that a positive correlation between human population and species richness may be typical at coarse scales, whereas a negative one may be more common when studies take care to finely differentiate between natural and urbanized areas. At coarse scales, a high proportion of urbanized regions may consist of natural vegetation, whereas, at fine scales, the higher the level of urbanization, the lower the proportion of natural habitats. Such a scale dependence has been suggested by Manne (2003), restated by Va´zquez & Gaston (2006) and is confirmed by studies which analysed the spatial co-occurrence of people and of the number of species of the same taxon at different resolutions, reporting a stronger correlation for coarser resolutions. Such a pattern was found, for instance, by Chown et al. (2003) for birds of South Africa at quarter-, half- and one-degree resolutions. A prime example is also an investigation in the region around Manchester, UK, of butterfly species richness, which increases with urbanization level at the 4-km2 scale, but decreases at a resolution of 1 km2 and 1 ha (Hardy & Dennis 1999). There is thus some evidence that the direction and strength of the correlation between human population presence and species richness could be scale dependent within single studies. However, to the best of our knowledge, it is not known whether such a scale dependence may be present also across different studies. Here, we straightforwardly test such a hypothesis by compiling studies on the impact of urbanization on species richness at relatively fine resolutions and of the

The human–biodiversity correlation is scale dependent 17

co-occurrence of human population and species richness at coarser resolutions and by documenting that the correlation coefficient between these variables is positively related to grain and extent of studies. MATERIALS AND METHODS

Data on the correlation between human presence and species richness were compiled from the published literature. Relevant papers were obtained with systematic key word searches (e.g. Ôurban-rural AND species richnessÕ or Ôhuman population AND speciesÕ) in scientific data bases and with inspection of cited and citing references of relevant papers already found. There was no a priori decision as to which taxa to include in the analysis, but we decided to analyse only vascular plants and vertebrates, as insects may respond to urbanization at different scales. A further reason not to investigate insects in this analysis is that studies focusing on them have mostly been carried out at a local scale only (with the possible exception of Luck et al. 2004). As there are only nine data points for plants, we only present results for the whole of the data set. But results seem to be consistent for plants and vertebrates taken as independent groups, although this would need to be confirmed when more data become available. However, there is no evidence that the reported scale dependence of the correlation between human presence and species richness is dependent on the biology of the organisms studied, as it is not the case that, for instance, plant data come from studies at fine scales and report a negative correlation and vice versa for animals. Both data from botanical and zoological studies were obtained from a wide variety of study scales and show a wide variation in the correlation coefficient between human presence and species richness. As a rule, species richness consisted in studies analysed of both native and exotic species. Studies not reporting the size of the sampling units (grain) were not included. In all, 28 studies, with 37 data points (some studies reported results from different taxa or at differing resolution), were retrieved. If surveys were carried out in the form of point counts, the diameter of these was recorded as grain. In case of line transects, the width of the transect was recorded. For square sample areas, the side of the square was recorded. For each record, we roughly estimated also the extent of the region covered by the investigation. We recorded also the number of plots. Grain (in km), extent (in km2) and number of plots were log-transformed prior to analysis to conform to the assumptions of statistical tests. The correlation between human population presence and species richness was typically represented by the proportion of variance explained by the regression or correlation of species richness with a variable expressing the amount of urbanization or  2006 Blackwell Publishing Ltd/CNRS

18 M. Pautasso

human presence (population size, population density, housing density, etc.). Data used in analyses are investigations of the correlation between human presence and the species richness of birds (62%), plants (23%), mammals (8%), herptiles (5%) and amphibians (3%). This species richness is in all cases the number of species in sampled areas (alpha diversity), and not the species turnover or dissimilarity across sampled areas (beta diversity). Studies come from Europe (44%), North America (28%), Africa (23%) and Asia (5%). As studies performed in neighbouring regions may be spatially non-independent, mixed models were run (SAS 9.1; SAS Institute, Inc., Cary, NC, USA) to control for spatial autocorrelation within the four geographical regions listed above. An exponential covariance structure was used, after having ascertained its better fit (in terms of both Akaike’s and Bayesian information criteria) to the null model compared with spherical, Gaussian, linear, linear-logarithm and power structures. An additional parameter was added to the two parameters governing the converging process because of marked variations in the response variable at a small-scale hindering convergence (for further details of the methodology for controlling for spatial autocorrelation please see Pautasso & Gaston 2005). As results from nonspatial models and from those controlling for spatial autocorrelation (with one reported exception) are qualitatively similar, only the latter are given. However, the proportion of variance in the response variable explained by the dependent variables was assessed from linear regressions.

Letter

variance explained is remarkable. The relationship is linear across at least four orders of magnitude of variation in study grain. The x-axis intercept of the regression line is approximately at (0, 0), which means that the correlation coefficient turns from positive into negative at below a study grain of c. 1 km.

(a)

(b)

RESULTS

The correlation coefficient between human population presence/urbanization level varies between )0.90 and +0.90. Its average is +0.08, its median is +0.03, and its SD is 0.44. Study grain varies between 0.1 and 250 km. Its average is 32 km, its median is 2 km, and its SD is 51 km. Study extent varies between 3 km2 and 20 · 106 km2. Its average is 3.7 · 106 km2, its median 3 · 103 km2 and its SD is 6.6 · 106 km2. The number of plots varies between 6 and 2995. The average number of plots is 852, the median is 204, and the SD is 972. The plots are located between 31 S and 59 N, with an average and median latitude of 28 N. The average absolute latitude is 38 N. There are no significant variations of any analysed variable with latitude, with the exception of the number of plots (the number of plots of studies increases with latitude in a marginally significant way). The correlation coefficient between human population presence/urbanization level and species richness significantly increases with the grain of the study (n ¼ 37, r2 ¼ 0.73, correlation coefficient ¼ )0.02 + 0.26 (log grain), slope SE ¼ 0.03, P < 0.0001; Fig. 1a). The proportion of  2006 Blackwell Publishing Ltd/CNRS

(c)

Figure 1 Regression of correlation coefficient between human

population/urbanization level and logarithmically transformed (a) study grain (km), (b) study extent (km2), and (c) number of plots. Data from botanical and zoological studies are represented by empty squares and filled diamonds respectively. Regression lines are based on the whole data set.

Letter

The correlation coefficient also increases significantly with the extent of the study (n ¼ 37, r2 ¼ 0.62, correlation coefficient ¼ )0.69 + 0.17 (log extent), SE ¼ 0.03, P < 0.0001; Fig. 1b). The proportion of variance explained is again substantial. The relationship appears to be linear across at least five orders of magnitude of variation in study extent. The x-axis intercept of the regression line is approximately at (4, 0), which means that the correlation coefficient turns from negative into positive at above a study extent of c. 10 000 km2 (this corresponds to a study grain of 1 km in the plot of study grain as a function of study extent, see below). In the spatial model, the correlation coefficient does not increase significantly with the number of plots (n ¼ 35, r2 ¼ 0.26, correlation coefficient ¼ )0.05 + 0.02 (log n plots), SE ¼0.07, P ¼ 0.82; Fig. 1c). However, in the non-spatial model the correlation coefficient increases significantly also with the number of plots. Study grain and extent are significantly positively related (n ¼ 37, r2 ¼ 0.79, log grain ¼ )2.33 + 0.57 (log extent), SE ¼ 0.07, P < 0.0001; Fig. 2a). The number of plots increases significantly with both study grain (n ¼ 35, r2 ¼ 0.48, n plots ¼ 2.29 + 0.37 (log grain), SE ¼ 0.08, P ¼ 0.0003; Fig. 2b) and extent (n ¼ 35, r2 ¼ 0.47, n plots ¼ 1.32 + 0.24 (log extent), SE ¼ 0.06, P ¼ 0.0008; Fig. 2c). These positive inter-relationships, together with the consistency of results using grain, extent or (to a lesser degree) number of plots, mean that study grain, extent and (possibly to a lesser degree) number of plots are on the whole interchangeable when discussing the effect of scale on the correlation between human presence and species richness.

The human–biodiversity correlation is scale dependent 19

(a)

(b)

(c)

DISCUSSION

When studying urbanization, as for other processes, different groups of ecologists may focus their investigations over distinct hierarchical levels of ecological functioning (Clergeau et al. 2006b): (i) the habitat or local level (at an absolute scale varying from meters to hundreds of meters; this is usually how far an individual can see from the ground in a town), which features elements with contrasting niche opportunities such as individual trees, green space, private gardens and buildings; (ii) the landscape level (with study units of size around 1–10 km; this order of magnitude represents the daily range of action of most human beings), with patches of fragmented vegetation in a matrix of residential, recreational, and industrial areas and a gradient between the city centre(s) and the edge of suburbs; (iii) a regional or subregional level (c. 100 km, depending on the continent), which includes greenways linking districts within metropolitan areas along urban–rural gradients, and can be best grasped from airplanes flying at mid-altitudes; and (iv) a

Figure 2 Correlation plots of logarithmically transformed (a) study

grain (km) and study extent (km2), (b) number of plots and study grain (km), and (c) number of plots and study extent (km2). See Fig. 1 for symbols and conventions.

biogeographical or macroecological level (> 1000 km), which deals with, for example, differences in urbanization patterns across latitudinal gradients or between different continents and involves remote sensing tools and data obtained from satellites. Having recognized that urbanization is a process operating across a continuum of different scales, and having acknowledged that urbanization’s effects on biodiversity are often reaching areas far away from densely inhabited regions, not only because of urban sprawl, the main result of this analysis is that the correlation between human population presence and species richness is scale dependent across independent studies. At coarse scales, there is a  2006 Blackwell Publishing Ltd/CNRS

20 M. Pautasso

spatial co-occurrence of human presence and biodiversity. At fine scales, the pattern is reversed, with higher species richness in plots where human population is less densely present. This is an unambiguous and compelling instance of the influence of scale on ecological patterns and processes (e.g. Levin 1992; Ricklefs 2004; Wu 2004). The boundary between these two regions is at a study grain of c. 1 km, which, for the studies analysed, corresponds to a square region c. 100 km along its sides. Other things being equal, if studies are carried out at scales coarser than these, they are likely to document an increase in species richness with increasing human population size or density. This means that, at broad scales, human beings are competing for living space with biodiversity (e.g. Balmford et al. 2001; Arau´jo 2003; Ding et al. 2006). Both tend to be supported by regions of high productivity. It has long been a troublesome conservation biology problem that national parks have been located in regions where human beings are not present in high numbers, and that these regions are also relatively poor in numbers of species, whereas conservation measures would actually be needed where human densities are highest. Reserve selection exercises and population viability analyses are needed to assess whether conservation and development can coexist in areas of high population density and biodiversity (Diniz-Filho et al. 2006). More research is also required both in the developing and developed world to evaluate the regional conservation value of patches of natural vegetation in urbanized settings when these contain high proportions of exotic or generalist species. The current unprecedented levels in worldwide travel and trade are associated with an increase in accidental and deliberate species introductions, particularly in urbanized landscapes. An opposite correlation between people’s presence and biodiversity is found at fine resolutions. When ecologists study nature in cityscapes with sample units smaller than roughly 1 km2, they tend to report a decrease in species richness with increasing level of urbanization. At this level, the presence of a higher proportion of buildings, concrete, tarmac or other paved surfaces is generally found to correlate negatively with the diversity of living organisms. Town centres are so densely built and so frequently trampled that they often end up in resembling biological deserts. Although the basic mechanism driving this pattern is evident (densely inhabited areas provide no physical space for plants and trees, the absence of which – also in terms of decaying deadwood – in turn makes the habitat less suitable for consumers), it is difficult to disentangle the contributions of different factors such as area, human disturbance and habitat homogeneity (McKinney 2006). A fine scale of enquiry allows urban ecologists to delimit green patches of towns from areas where vegetation is nearly absent. But are densely built towns species poorer than more extended  2006 Blackwell Publishing Ltd/CNRS

Letter

urbanized environments because they cover a smaller size, because they are subject to a higher cumulative impact by humans, or because they contain a less diverse suite of habitats? It is possible that all these factors are operating at the same time but at different scales. At a sufficiently large scale, urbanized ecosystems may be species richer than their surroundings because they exhibit a richer mosaic of habitats and microenvironments: allotments, backyards, brownfield sites, flower beds, private, public and botanical gardens, golf courses, grasslands, graveyards, hedges, orchards, parks, railway land, relict agricultural land, roadsides, roofs, roundabouts, ruderal sites, sport facilities, squares, terraces, tree avenues, urban forests, walls, waterways, wastelands and woodland fragments (e.g. Klotz 1990; Pysˇek 1993; Deutschewitz et al. 2003; Benvenuti 2004). At a finer scale of analysis, less urbanized environments may tend to show more species than town centres as they contain larger vegetation patches, as for example shown by a study of birds resident in urban woods of Seoul, Korea (Park & Lee 2000). A similar scale dependence may occur in relation to the contribution of native vs. exotic species to the pattern reported here. Whilst Celesti-Grapow et al. (2006) find at a grid cell size of 1.6 km2 that for the flora of Rome factors increasing the presence of native species decrease the occurrence of exotic species and vice versa, Deutschewitz et al. (2003), using a grid cell size of c. 32 km2, for the flora of a district in Germany, and Ku¨hn et al. (2004), using a grid cell size of c. 130 km2, for the flora of Germany, report that the presence of native and exotic species are both positively correlated to the degree of urbanization. On a similar vein, the argument has been made that areas with a higher human presence have been sampled for species in a more thorough way (for plant species richness, e.g. in the urbanized area of Nairobi compared with other regions of north-western Kenya, Stadler et al. 2000), but it is unclear whether this methodological issue may be invoked to explain the scale dependence of the correlation between people and biodiversity. An open question is also whether the warmer climate reported in urbanized areas (other things being equal, higher temperatures are expected to increase the number of species that are able to subsist in a given area) is operating in a scaledependent way. We know that climate is predicted to change both at a global and regional level, but the implications for the correlation between human presence and species richness at different scales are largely unexplored. The positive relation of the correlation between human presence and species richness with study scale may provide an explanation of previous studies reporting the peaking of species richness at intermediate levels of urbanization. In these cases, the correlation coefficient between human presence and species richness is around zero. Examples of studies at intermediate scales reporting a peak of species

Letter

richness at intermediate levels of urbanization include birds in an urban–rural gradient in Santa Clara county, California, USA (Blair 1996), in coastal southern California, USA (Crooks et al. 2004), in Rome, Italy (Sorace 2001), in parks of towns in western France (Clergeau et al. 2001), as well as cavity-nesting birds in the rapidly urbanizing region around Seattle, Washington, USA (Blewett & Marzluff 2005). The level of urbanization also failed to be a simple predictor of species richness at intermediate scales of analysis for the flora in Berlin, Germany (Maurer et al. 2000) and in Great Britain (Roy et al. 1999). Studies at intermediate scales may report no or only a weak correlation between human presence and species richness because at this level of analysis the mechanisms involved at fine and coarse scales overlap and thus cancel out each other. Urbanization is a major force affecting community structure. Understanding these effects is essential in today’s rapidly modifying landscapes. It is important to realize that the scaling of human biodiversity co-occurrence can turn this correlation from positive at coarse scales to negative at finer scales. The top-right part of Fig. 1a,b (a positive correlation between human presence and species richness at coarse scales) should not be used to justify more urban sprawl over large regions or new developments close to regions of outstanding natural beauty and to nature reserves, as these correlations are largely an accident of history. People have tended to settle or have flourished in regions of high biodiversity, and may have in part increased the species richness of these regions with the deliberate or inadvertent introduction of species and modification of habitats, but because these regions were already the most productive, as a rule they already tended to harbour a richer array of species. Conversely, the take-home message from the bottom-left part of Fig. 1a,b (the negative impact of urbanization on species richness at a local scale) is that urban compaction and densification, i.e. the filling of the few undeveloped, green spaces left in towns, will inevitably make towns lose 5 species, lower the human-perceived quality of life, and increase the homogenization of urbanized landscapes and biota. Although nature conservation in towns should not only focus on patches of vegetation (Maurer et al. 2000), these should not be neglected, as they can harbour surprisingly high numbers of species, of both exotic and native origin (e.g. Ku¨hn et al. 2004). Also urban and suburban parks can exhibit high species richness, especially if they consist of seminatural habitats (Cornelis & Hermy 2004). From an applied point of view, the importance of patch size and context in urban conservation has been emphasized repeatedly. Although species richness may in some cases and for a certain lag-time not be affected by urbanization even at local scales, other indicators of biotic integrity can be used to illustrate the degradation which follows development.

The human–biodiversity correlation is scale dependent 21

Temporal studies may provide evidence about mechanisms underlying the scale dependence of the impact of human presence on species richness. For example, Tait et al. (2005) showed that, over the period 1836–2002, the plant species richness of the metropolitan area of Adelaide, South Australia, increased of nearly 50%, because species introductions outnumbered species extinctions. No net change was found for amphibians, reptiles and birds. It is important for temporal studies of the effect of urbanization on biodiversity to adopt a time frame which is long enough to enable significant variations to arise, as shown by Evans et al. (2006). These authors studied the relationship between avian and anuran species richness and human population in South Africa at two time points. Despite a marked increase in human population, there was no significant variation in the species–population relationship. Although the time-lag chosen for the analyses (1996–2001) may be too short for any signal to be perceived, this result would speak against a direct causal link between the two variables at the scale chosen for analyses. Over a longer period (1792 until the present day), Isocrono et al. (2006) showed that it is not the number of human beings present, but the degree of environmental pollution that determines lichen species richness in the metropolitan area of Turin, Italy. In this as in other urbanized areas of the developed world, recent increases in the species richness of lichens, concurrently with decreases in CO, NOx and SO2 emissions, have been reported, but generally these studies are carried out at a scale that is fine enough for a negative correlation between lichen species richness and urbanization levels to be observed even now that pollution levels have decreased (e.g. Giordani et al. 2002; Giordani 2006; Munzi et al. 2006). The direction and strength of the association between human presence and species richness (alpha diversity, i.e. the number of species present in sampled areas) varies with scale. As human presence has been reported to be a major driver of biotic regional homogenization (e.g. McKinney & Lockwood 1999), it is possible that species turnover amongst sampled areas (beta diversity) may not be positively related to human presence at coarse scales, as is the case for alpha diversity. However, this sort of study is less frequently found in the literature, which means that, in order to assess whether or not the biotic homogenization brought about by humans is scale dependent, there is still a need for primary studies of the relationship between human population presence and the beta diversity of various taxa at different scales. Similarly, there is a need for an assessment of whether or not habitat heterogeneity, nutrient inputs and pollution levels are scale dependent in human-modified habitats, in order to test formally whether differences in these factors across scales can be a mechanism responsible for the pattern reported here. Whether or not the correlation between human presence and species richness  2006 Blackwell Publishing Ltd/CNRS

22 M. Pautasso

responds to variations in scale in a similar way for different taxa also necessitates the collection of more data. Further investigations are also needed at different scales of analysis to determine the contribution of native and exotic species to the scale dependence of the spatial correlation between biodiversity and the presence of people. ACKNOWLEDGEMENTS

Many thanks to M. Baltisberger, A. Cannon, R.G. Davies, K.L. Evans, R. Fuller, K.J. Gaston, O. Holdenrieder, S. Jackson, G.P. Mondino, A.S.L. Rodrigues, D. Storch, L.B. Va´zquez and P.J. Weisberg for insights and discussions, and to K. England, B. Kench and three anonymous referees for commenting on a previous version of the draft. REFERENCES Arau´jo, M.B. (2003). The coincidence of people and biodiversity in Europe. Glob. Ecol. Biogeogr., 12, 5–12. Balmford, A., Moore, J.L., Brooks, T., Burgess, N., Hansen, L.A., Williams, P. et al. (2001). Conservation conflicts across Africa. Science, 291, 2616–2619. Beissinger, S.R. & Osborne, D.R. (1982). Effects of urbanization on avian community organization. Condor, 84, 75–83. Benvenuti, S. (2004). Weed dynamics in the Mediterranean urban ecosystem: ecology, biodiversity and management. Weed Res., 44, 341–354. Blair, R.B. (1996). Land use and avian species diversity along an urban gradient. Ecol. Appl., 6, 506–519. Blewett, C.M. & Marzluff, J.M. (2005). Effects of urban sprawl on snags and the abundance and productivity of cavity-nesting birds. Condor, 107, 678–693. Celesti-Grapow, L., Pysek, P., Jarosı´k, V. & Blasi, C. (2006). Determinants of native and alien species richness in the urban flora of Rome. Divers. Distrib., 12, 490–501. Chown, S.L., van Rensburg, B.J., Gaston, K.J., Rodrigues, A.S.L. & van Jaarsveld, A.S. (2003). Energy, species richness, and human population scale: conservation implications at a national scale. Ecol. Appl., 13, 1233–1241. Clergeau, P., Jokimaki, J. & Savard, J.P.L. (2001). Are urban bird communities influenced by the bird diversity of adjacent landscapes? J. Appl. Ecol., 38, 1122–1134. Clergeau, P., Croci, S., Jokima¨ki, J., Kaisanlahti-Jokima¨ki, M.L. & Dinetti, M. (2006a). Avifauna homogenisation by urbanisation: analysis at different European latitudes. Biol. Conserv., 127, 336– 344. Clergeau, P., Jokimaeki, J. & Snep, R. (2006b). Using hierarchical 7 levels for urban ecology. Trends Ecol. Evol. (in press). Cornelis, J. & Hermy, M. (2004). Biodiversity relationships in urban and suburban parks in Flanders. Landsc. Urban Plan., 69, 385–401. Crooks, K.R., Suarez, A.V. & Bolger, D.T. (2004). Avian assemblages along a gradient of urbanization in a highly fragmented landscape. Biol. Conserv., 115, 451–462. Deutschewitz, K., Lausch, A., Ku¨hn, I. & Klotz, S. (2003). Native and alien plant species richness in relation to spatial heterogeneity on a regional scale in Germany. Glob. Ecol. Biogeogr., 12, 299–311.  2006 Blackwell Publishing Ltd/CNRS

Letter

Ding, T.S., Yuan, H.W., Geng, S., Koh, C.N. & Lee, P.F. (2006). Macro-scale bird species richness patterns of the East Asian mainland and islands: energy, area and isolation. J. Biogeogr., 33, 683–693. Diniz-Filho, J.A.F., Bini, L.M., Pinto, M.P., Rangel, T.F.L.V.B., Carvalho, P. & Bastos, R.P. (2006). Anuran species richness, complementarity and conservation conflicts in Brazilian Cerrado. Acta Oecol., 29, 9–15. Erz, W. (1964). Populationso¨kologische Untersuchungen an der Avifauna zweier Nordwestdeutscher Grosssta¨dte. Z. Wiss. Zool., 170, 1–111. Evans, K.L. & Gaston, K.J. (2005). People, energy and avian species richness. Glob. Ecol. Biogeogr., 14, 187–196. Evans, K.L., van Rensburg, B.J., Gaston, K.J. & Chown, S.L. (2006). People, species richness and human population growth. Glob. Ecol. Biogeogr., 15, 625–636. Fudali, E. (2001). The ecological structure of the bryoflora of Wroclaw’s parks and cemeteries in relation to their localization and origin. Acta Soc. Bot. Poloniae, 70, 229–235. Gaston, K.J. (2005). Biodiversity and extinction: species and people. Prog. Phys. Geogr., 29, 239–247. Giordani, P. (2006). Is the diversity of epiphytic lichens a reliable indicator of air pollution? A case study from Italy. Environ. Pollut. 9 (in press). Giordani, P., Brunialti, G. & Alleteo, D. (2002). Effects of atmospheric pollution on lichen biodiversity (LB) in a Mediterranean region (Liguria, northwest Italy). Environ. Pollut., 118, 53–64. Hansen, A.J., Knight, R.L., Marzluff, J.M., Powell, S., Brown, K., Gude, P.H. et al. (2005). Effects of exurban development on biodiversity: patterns, mechanisms, and research needs. Ecol. Appl., 15, 1893–1905. Hardy, P.B. & Dennis, R.L.H. (1999). The impact of urban development on butterflies within a city region. Biodivers. Conserv., 8, 1261–1279. Hunter, M.L. & Yonzon, P. (1993). Altitudinal distribution of birds, mammals, people, forest, and parks in Nepal. Conserv. Biol., 7, 420–423. Isocrono, D., Matteucci, E., Ferrarese, A., Pensi, E. & Piervittori, R. (2006). Lichen colonization in the city of Turin (N Italy) based on current and historical data. Environ. Pollut., 145, 258–265. Jim, C.Y. (2002). Heterogeneity and differentiation of the tree flora in three major land uses in Guangzhou City, China. Ann. For. Sci., 59, 107–118. Klotz, S. (1990). Species/area and species/inhabitants relations in European cities. In: Urban Ecology – Plants and Plant Communities in Urban Environments (eds Sukopp, H., Hejny, S. & Kowarik, I.). SPB Academic Publishing, The Hague, pp. 99–103. Kubes, J. & Fuchs, R. (1998). Village as a bird refuge in cultural landscape (largely agricultural landscape, the Czech Republic). Ekol. Bratislava, 17, 208–220. Ku¨hn, I., Brandl, R. & Klotz, S. (2004). The flora of German cities is naturally species rich. Evol. Ecol. Res., 6, 749–764. Leveau, C.M. & Leveau, L.M. (2005). Avian community response to urbanization in the Pampean region, Argentina. Ornitol. Neotrop., 16, 503–510. Levin, S.A. (1992). The problem of pattern and scale in ecology. Ecology, 73, 1943–1967. Lopez-Moreno, I.R., Diaz-Betancourt, M.E. & Landa, T.S. (2003). Social insects in human environments – ants in the city of Coatepec (Veracruz, Mexico). Sociobiology, 42, 605–621.

Letter

Luck, G.W., Ricketts, T.H., Daily, G.C. & Imhoff, M. (2004). Alleviating spatial conflict between people and biodiversity. Proc. Natl Acad. Sci. USA, 101, 182–186. Maffi, L. (2005). Linguistic, cultural, and biological diversity. Annu. Rev. Anthropol., 34, 599–617. Manne, L.L. (2003). Nothing has yet lasted forever: current and threatened levels of biological and cultural diversity. Evol. Ecol. Res., 5, 517–527. Maurer, U., Peschel, T. & Schmitz, S. (2000). The flora of selected urban land-use types in Berlin and Potsdam with regard to nature conservation in cities. Landsc. Urban Plan., 46, 209–215. McKinney, M.L. (2002). Do human activities raise species richness? Contrasting patterns in United States plants and fishes. Glob. Ecol. Biogeogr., 11, 343–348. McKinney, M.L. (2006). Correlated non-native species richness of birds, mammals, herptiles and plants: scale effects of area, human population and native plants. Biol. Invasions, 8, 415–425. McKinney, M.L. & Lockwood, J.L. (1999). Biotic homogenization: a few winners replacing many losers in the next mass extinction. Trends Ecol. Evol., 14, 450–453. Moore, A.A. & Palmer, M.A. (2005). Invertebrate biodiversity in agricultural and urban headwater streams: implications for conservation and management. Ecol. Appl., 15, 1169–1177. Munzi, S., Ravera, S. & Caneva, G. (2006). Epiphytic lichens as indicators of environmental quality in Rome. Environ. Pollut. 11 (in press). Park, C.R. & Lee, W.S. (2000). Relationship between species composition and area in breeding birds of urban woods in Seoul, Korea. Landsc. Urban Plan., 51, 29–36. Pautasso, M. & Gaston, K.J. (2005). Resources and global avian assemblage structure in forests. Ecol. Lett., 8, 282–289. Pickett, S.T.A., Cadenasso, M.L., Grove, J.M., Nilon, C.H., Pouyat, R.V., Zipperer, W.C. et al. (2001). Urban ecological systems: linking terrestrial ecological, physical, and socioeconomic components of metropolitan areas. Annu. Rev. Ecol. Syst., 32, 127–157. Pysˇek, P. (1993). Factors affecting the diversity of flora and vegetation in central European settlements. Vegetatio, 106, 89–100. Real, R., Barbosa, A.M., Porras, D., Kin, M.C., Marquez, A.L., Guerrero, J.C. et al. (2003). Relative importance of environment, human activity and spatial situation in determining the distribution of terrestrial mammal diversity in Argentina. J. Biogeogr., 30, 939–947. Ricklefs, R.E. (2004). A comprehensive framework for global patterns in biodiversity. Ecol. Lett., 7, 1–15. Ricotta, C., Celesti-Grapow, L., Avena, G. & Blasi, C. (2001). Topological analysis of the spatial distribution of plant species richness across the city of Rome (Italy) with the echelon approach. Landsc. Urban Plan., 57, 69–76. Roy, D.B., Hill, M.O. & Rothery, P. (1999). Effects of urban land cover on the local species pool in Britain. Ecography, 22, 507–515. Soh, M.C.K., Sodhi, N.S. & Lim, S.L.H. (2006). High sensitivity of montane bird communities to habitat disturbance in Peninsular Malaysia. Biol. Conserv., 129, 149–166. Sorace, A. (2001). Value to wildlife of urban-agricultural parks: a case study from Rome urban area. Environ. Manage., 28, 547–560. Stadler, J., Trefflich, A., Klotz, S. & Brandl, R. (2000). Exotic plant species invade diversity hot spots: the alien flora of northwestern Kenya. Ecography, 23, 169–176.

The human–biodiversity correlation is scale dependent 23

Sutherland, W.J. (2003). Parallel extinction risk and global distribution of languages and species. Nature, 423, 276–279. Tait, C.J., Daniels, C.B. & Hill, R.S. (2005). Changes in species assemblages within the Adelaide Metropolitan Area, Australia, 1836–2002. Ecol. Appl., 15, 346–359. Tikhonova, G.N., Tikhonov, I.A. & Bogomolov, P.L. (2006). Impact of a small city on the structure of small mammal fauna in forests of the northeastern Moscow region. Russ. J. Ecol., 37, 278–283. Tomiałojc´, L. (1970). Quantitative studies on the synanthropic avifauna of Legnica and its environs. Acta Ornithol., 12, 294–392. Va´zquez, L.B. & Gaston, K.J. (2006). People and mammals in Mexico: conservation conflicts at a national scale. Biodivers. Conserv., 15, 2397–2414. Wania, A., Kuhn, I. & Klotz, S. (2006). Plant richness patterns in agricultural and urban landscapes in Central Germany – spatial gradients of species richness. Landsc. Urban Plan., 75, 97–110. Wu, J. (2004). Effects of changing scale on landscape pattern analysis: scaling relations. Landsc. Ecol., 19, 125–138. Zanette, L.R.S., Martins, R.P. & Ribeiro, S.P. (2005). Effects of urbanization on Neotropical wasp and bee assemblages in a Brazilian metropolis. Landsc. Urban Plan., 71, 105–121. APPENDIX 1: REFERENCES FROM WHICH DATA USED IN ANALYSES WERE OBTAINED Arau´jo, M.B. (2003). The coincidence of people and biodiversity in Europe. Glob. Ecol. Biogeogr., 12, 5–12. Blair, R.B. (1996). Land use and avian species diversity along an urban gradient. Ecol. Appl., 6, 506–519. Celesti-Grapow, L., Pysek, P., Jarosı´k, V. & Blasi, C. (2006). Determinants of native and alien species richness in the urban flora of Rome. Divers. Distrib., 12, 490–501. Chown, S.L., van Rensburg, B.J., Gaston, K.J., Rodrigues, A.S.L. & van Jaarsveld, A.S. (2003). Energy, species richness, and human population scale: conservation implications at a national scale. Ecol. Appl., 13, 1233–1241. Clergeau, P., Savard, J.P.L., Mennechez, G. & Falardeau, G. (1998). Bird abundance and diversity along an urban–rural gradient: a comparative study between two cities on different continents. Condor, 100, 413–425. Crooks, K.R., Suarez, A.V. & Bolger, D.T. (2004). Avian assemblages along a gradient of urbanization in a highly fragmented landscape. Biol. Conserv., 115, 451–462. Deutschewitz, K., Lausch, A., Ku¨hn, I. & Klotz, S. (2003). Native and alien plant species richness in relation to spatial heterogeneity on a regional scale in Germany. Glob. Ecol. Biogeogr., 12, 299–311. Donnelly, R. & Marzluff, J.M. (2004). Importance of reserve size and landscape context to urban bird conservation. Conserv. Biol., 18, 733–745. Evans, K.L. & Gaston, K.J. (2005). People, energy and avian species richness. Glob. Ecol. Biogeogr., 14, 187–196. Evans, K.L., van Rensburg, B.J., Gaston, K.J. & Chown, S.L. (2006) People, species richness and human population growth. Glob. Ecol. Biogeogr., 15, 625–636. Fernandez-Juricic, E. (2000). Bird community composition patterns in urban parks of Madrid: the role of age, size and isolation. Ecol. Res., 15, 373–383.

 2006 Blackwell Publishing Ltd/CNRS

24 M. Pautasso

Fraterrigo, J.M. & Wiens, J.A. (2005). Bird communities of the Colorado Rocky Mountains along a gradient of exurban development. Landsc. Urban Plan., 71, 263–275. Gaston, K.J. & Evans, K.L. (2004). Birds and people in Europe. Proc. R. Soc. Lond. B, Biol. Sci., 271, 1649–1655. Germaine, S.S., Rosenstock, S.S., Schweinsburg, R.E. & Richardson, W.S. (1998). Relationships among breeding birds, habitat, and residential development in Greater Tucson, Arizona. Ecol. Appl., 8, 680–691. Hope, D., Gries, C., Zhu, W.X., Fagan, W.F., Redman, C.L., Grimm, N.B. et al. (2003). Socioeconomics drive urban plant diversity. Proc. Natl Acad. Sci. USA, 100, 8788–8792. Ku¨hn, I. & Klotz, S. (2006). Urbanization and homogenization – comparing the floras of urban and rural areas in Germany. Biol. Conserv., 127, 292–300. Ku¨hn, I., Brandl, R. & Klotz, S. (2004). The flora of German cities is naturally species rich. Evol. Ecol. Res., 6, 749–764. Lee, P.F., Ding, T.S., Hsu, F.H. & Geng, S. (2004). Breeding bird species richness in Taiwan: distribution on gradients of elevation, primary productivity and urbanization. J. Biogeogr., 31, 307– 314. Miller, J.R., Wiens, J.A., Hobbs, N.T. & Theobald, D.M. (2003). Effects of human settlement on bird communities in lowland riparian areas of Colorado (USA). Ecol. Appl., 13, 1041–1059. Posa, M.R.C. & Sodhi, N.S. (2006). Effects of anthropogenic land use on forest birds and butterflies in Subic Bay, Philippines. Biol. Conserv., 129, 256–270.

 2006 Blackwell Publishing Ltd/CNRS

Letter

Ricotta, C., Grapow, L.C., Avena, G. & Blasi, C. (2001). Topological analysis of the spatial distribution of plant species richness across the city of Rome (Italy) with the echelon approach. Landsc. Urban Plan., 57, 69–76. Rottenborn, S.C. (1999). Predicting the impacts of urbanization on riparian bird communities. Biol. Conserv., 88, 289–299. Roy, D.B., Hill, M.O. & Rothery, P. (1999). Effects of urban land cover on the local species pool in Britain. Ecography, 22, 507–515. Sandstro¨m, U.G., Angelstam, P. & Mikusinski, G. (2006). Ecological diversity of birds in relation to the structure of urban green space. Landsc. Urban Plan., 77, 39–53. Sorace, A. (2001). Value to wildlife of urban-agricultural parks: a case study from Rome urban area. Environ. Manage., 28, 547–560. Tilghman, N.G. (1987). Characteristics of urban woodlands affecting breeding bird diversity and abundance. Landsc. Urban Plan., 14, 481–495. Va´zquez, L.B. & Gaston, K.J. (2006). People and mammals in Mexico: conservation conflicts at a national scale. Biodivers. Conserv., 15, 2397–2414.

Editor, Thomas Crist Manuscript received 28 August 2006 First decision made 29 September 2006 Manuscript accepted 10 October 2006