Region effects influence local tree species diversityRobert E. Ricklefsa,1 and Fangliang Heb,c,1

aDepartment of Biology, University of Missouri, St. Louis, MO 63121-4499; bSun Yat-sen University-Alberta Joint Lab for Biodiversity Conservation, State KeyLaboratory of Biocontrol, School of Life Sciences, Sun Yat-sen University, Guangzhou 510275, China; and cDepartment of Renewable Resources, Universityof Alberta, Edmonton, AB, T6G 2H1 Canada

Contributed by Robert E. Ricklefs, December 15, 2015 (sent for review May 20, 2015; reviewed by Holger Kreft, Jonathan A. Myers, Nigel C. A. Pitman)

Global patterns of biodiversity reflect both regional and local pro-cesses, but the relative importance of local ecological limits to speciescoexistence, as influenced by the physical environment, in contrast toregional processes including species production, dispersal, and extinc-tion, is poorly understood. Failure to distinguish regional influencesfrom local effects has been due, in part, to sampling limitations atsmall scales, environmental heterogeneity within local or regionalsamples, and incomplete geographic sampling of species. Here, we usea global dataset comprising 47 forest plots to demonstrate significantregion effects on diversity, beyond the influence of local climate,which together explain more than 92% of the global variation in localforest tree species richness. Significant region effects imply that large-scale processes shaping the regional diversity of forest trees exertinfluence down to the local scale, where they interact with localprocesses to determine the number of coexisting species.

climate effect | forest dynamics plots | global biodiversity |latitudinal gradient | tree species richness

Ecologists generally agree that large-scale patterns of diversityreflect a balance between regional processes of species pro-

duction, extinction, and dispersal, on one hand, and within-regionsorting of species based on adaptations to physical conditions ofthe environment, as well as competition among species for lim-iting resources, on the other hand (1–4). Nonetheless, the spatialscale down to which region effects extend has not been well re-solved (5–7), but has wide-ranging implications for understandingthe origins of patterns in local species richness. If the speciesrichness of a local assemblage were strictly limited by competitionand other local interactions among populations, new species couldnot be added without others being forced out, and we would ex-pect to find a common relationship between diversity and localenvironmental conditions across regions (6). However, if uniquehistorical and biogeographic features of each region influencedwithin-region diversification and extinction (8–10), these region-specific effects might contribute to the global pattern in localspecies richness. Efforts to disentangle these effects have met withlimited success and have led to a long-standing discussion of therelationship between local and regional diversity (6, 11–21).Analyses designed to distinguish local and regional influences

on diversity have found strong region effects in some cases (8,22) and weak or nonexistent region effects in others (15, 23–25).However, most studies that failed to find significant region ef-fects either have addressed large, biologically heterogeneoussamples, or they have used local samples (e.g., 0.1-ha “Gentry”plots) that are too small to characterize the diversity of localassemblages adequately (26, 27). Additionally, many large-scalesamples have been compiled from maps generated from presence-only museum records or from coarse-scale atlases that documentthe extent of species occupancy, not actual local occurrence. Suchdata often undercount local species richness. Moreover, manytests of the diversity–environment relationship have analyzed dataon local communities that extend over broad ranges of ecologicalconditions (e.g., tropical rainforests to arctic tundra and hotdeserts; ref. 25) with a range of biomes and vegetation typesunevenly represented among regions. These sampling issues haveconfounded the testing of region effects.

In this study, we analyze a dataset of tree species richness fromthe Center for Tropical Forest Science—Forest Global EarthObservatories (CTFS—ForestGEO; www.forestgeo.si.edu/; ref.28) to disentangle the influences of local climate and regionalfactors, i.e., differences between regions resulting from uniquehistories and geographic settings, on the global biodiversitypattern. The data represent 47 forest dynamics plots distributedworldwide (Fig. 1) with a median size of 25 ha, within which allindividual trees equal to or greater than 1 cm diameter at breastheight (DBH) were identified and counted (SI Appendix, TableS1). Plots of this size are large enough to include adequatesamples of species richness, but small enough to avoid sub-stantial heterogeneity in climate and vegetation structure withinthem. The CTFS data are complete censuses, and the speciesrichness in each plot is accurate. Many previous studies havebeen based on data from herbarium records of coarse-scalespecies range maps, and species richness is generally under-estimated, considerably so in some cases. Moreover, the forestplots represent a single vegetation type surveyed over a range ofenvironmental conditions. We assembled for each plot a set oflocal plot characteristics and climate data, and used generalizedlinear models to characterize the relationship between numberof tree species and local plot variables and to test the additionalstatistical effect of region on local species richness. Our analyseswere repeated for species richness with tree DBH ≥ 1 cm andDBH ≥ 10 cm.

ResultsDifferences between continental-scale regions (“region effects”)alone explained 73% of the variance in species richness amongplots for both DBH ≥ 1 and DBH ≥ 10 cm (Fig. 2 and Table 1).

Significance

The number of species in local ecological assemblages isconstrained by competition and other local interactions be-tween populations. However, ecologists have not agreed onthe degree to which the unique history and geography ofcontinental regions also influence local species richness.Here, based on forest plots distributed throughout theworld, we show that the species richness of forest treesvaries among continental regions independently of local cli-mate. In particular, forest plots in Europe and North Americacontain fewer species than expected from the statistical re-lationship of diversity to present-day local climate. Thus, thenumber of species co-occurring locally reflects, to a signifi-cant extent, region characteristics, including geographic andgeologic history, that influence evolutionary diversificationand regional extinction.

Author contributions: R.E.R. and F.H. designed research, performed research, analyzeddata, and wrote the paper.

Reviewers: H.K., University of Göttingen; J.A.M., Washington University in St. Louis; andN.C.A.P., The Field Museum.

The authors declare no conflict of interest.1To whom correspondence may be addressed. Email: ricklefs@umsl.edu or fhe@ualberta.ca.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1523683113/-/DCSupplemental.

674–679 | PNAS | January 19, 2016 | vol. 113 | no. 3 www.pnas.org/cgi/doi/10.1073/pnas.1523683113

Dow

nloa

ded

by g

uest

on

June

21,

202

0

Because region effects are confounded with local climate andhabitat attributes, we further included climate and additionallocal plot variables listed in SI Appendix, Table S1 and selectedamong these full models by a stepwise procedure. The “best”model selected by the Akaike Information Criterion (AIC) foreach DBH class is presented in Fig. 3 and Table 2. Although theinfluence of individual climate variables can be complex, theoverall results are consistent and clear: The models for DBH ≥ 1and DBH ≥ 10 cm retain a reduced set of local climate variables,in addition to sample size per plot (number of stems) and region,to explain >92% of the global variation in local tree diversity(Fig. 3 and Table 2). Thus, three major sources of variation arestatistically associated with local tree species richness: (i) bio-geographic region, (ii) sample size, and (iii) local climate.Significant region effects show that species richness varies

among continents after accounting for the influence of localclimate and plot characteristics. The unique contribution of theregion effects to variation in species richness can be quantified bycomparing the difference in explained variance between the fullmodel with region effects, number of stems, and climate (R2 =0.931 for DBH ≥ 1 cm and 0.921 for DBH ≥ 10 cm) and thereduced model that excludes the region effects (R2 = 0.863 and0.821, respectively). Although the unique contribution of regionamounts to only 7% and 10% of the total variance, the uniquecontributions of local plot variables are also small—20% and19%, respectively. The strong association of variation in climatewith region (SI Appendix, Fig. S2) makes it difficult to partitionthe effects of each unambiguously, except to say that both makesignificant contributions to variation in diversity. Several of theforest plots are located on islands (Taiwan, Hainan Island, thePhilippines, Borneo, New Guinea, Sri Lanka, and Puerto Rico),and we tested for an isolation effect differentiating continentaland island localities. We did not detect an additional effect ofisolation when region effects were considered alone (Fig. 2 andTable 1). However, in the full model including climate variables,isolation (i.e., island locations) was significantly associated withlower species richness (P ≤ 1.33e-03) for both DBH ≥ 1 andDBH ≥ 10 cm size classes (Fig. 3 and Table 2).

The sampling effect, as measured by the number of stems in aplot, is another strong predictor of local diversity. Because plotsize (area) is strongly correlated with the number of stems (R2 =0.57 for DBH ≥ 1 and 0.66 for DBH ≥ 10 cm), plot size itself wasnot retained as a significant effect by the stepwise model selection.The climate variables represent two types of factors (Fig. 3 and

Table 2): (i) energy factors, including mean temperature ofwarmest quarter (BIO10) (standardized effect size, β̂ = 0.99, P =2.6e-05) for DBH ≥ 1 cm, and Mean temperature of wettestquarter (BIO8) (β̂= 1.47, P = 2.5e-04) for DBH ≥ 10 cm; and (ii)climate seasonality factors (variability in temperature and wateravailability), including isothermality (BIO3: reduced temperatureseasonality) (β̂ = 1.25, P = 1.4e-05) and precipitation seasonality(BIO15) (β̂= −0.65, P = 1.3e-03) for DBH ≥ 1 cm, and these sametwo factors with β̂ = 1.37 (P = 7.0e-05) and β̂ = −0.84 (P = 3.9e-04), respectively, for DBH ≥ 10 cm. Although the effect of specific

Fig. 1. Global distribution of the 47 CTFS plots. The number associated with each plot is its size in hectares. The base vegetation map is the 2012 MODISglobal land cover map (www.landcover.org/data/lc/) with IGBP Land Cover Type Classification.

Fig. 2. Magnitude of the regional effects on species richness. Neotropicalsites define the baseline for the region effects. Coefficients of the negativebinomial generalized linear models (Table 1) and their 95% confidence in-tervals for DBH ≥ 1 cm (Left) and 10 cm (Right), respectively. The regionaleffects models explain 73.3% of variation in richness with DBH ≥ 1 cm (i.e.,R2 = 0.733; Left) and 72.6% of variation in richness with DBH ≥ 10 cm (Right).

Ricklefs and He PNAS | January 19, 2016 | vol. 113 | no. 3 | 675

ECOLO

GY

Dow

nloa

ded

by g

uest

on

June

21,

202

0

climatic variables on species richness can differ, both highertemperature (BIO8 and BIO10) and isothermality (BIO3) areassociated with higher local plot diversity. Precipitation seasonality(BIO15), for which high values are associated with prolonged dryseasons, is associated with reduced tree species richness.Actual evapotranspiration (AET, a measure of water flux) and

potential evapotranspiration (PET, a measure of thermal input)were not retained as significant independent variables in themultiple regressions in our study although they are positivelycorrelated with species richness in some analyses (23, 25).However, because variation in evapotranspiration across theCTFS plots is strongly correlated with mean temperature of thewarmest quarter (BIO10) (R2 = 0.59 for AET; R2 = 0.55 for PET),mean temperature of wettest quarter (BIO8) (R2 = 0.59, 0.41),and isothermality (BIO3) (R2 = 0.50, 0.50), our results supportthese previous studies. Similarly, although the number of treespecies decreased with increasing absolute latitude north or southof the equator (SI Appendix, Fig. S3; R2 = 0.77 for DBH ≥ 1 andR2 = 0.74 for DBH ≥ 10 cm), latitude per se was not retainedunder stepwise model selection because of its close correlationwith climate variables: isothermality (R2 = 0.75), mean temper-ature of wettest quarter (R2 = 0.47), and mean temperature ofwarmest quarter (R2 = 0.58), respectively. Finally, we note thattopographic variables (i.e., elevation and elevation range within aplot) were not significant effects for either the 1-cm DBH or the10-cm DBH size class. Thus, within-plot habitat variation prob-ably does not strongly influence the species richness of localassemblages, at least among these forest plots in the context ofglobal variation.

DiscussionGlobal patterns of biodiversity continue to attract attention (2,25, 29–31), but the relative influence of historical and regionalfactors versus local habitat and climate on global variation intaxonomic diversity remains controversial. The problem derivesto some degree from incomplete and heterogeneous diversitydata, uncertainties about the regional distributions of species,and inconsistencies in sampling scale. We have tried to circum-vent these issues by relating local diversity in plots of similar size,

whose tree diversity has been assessed by identical censusmethods, to a widely used set of climate variables in addition toregion and plot-specific traits. The local tree taxonomic diversityof the 47 plots included in this study showed a clear globalpattern, exhibiting the typical latitudinal gradient of decliningvalues away from the equator (SI Appendix, Fig. S3), whichparallels gradients in several climate variables. In addition, ouranalyses show that differences in tree species richness betweenregions, presumably related to their unique history and physi-ography (32–34), contribute substantially to global variation inlocal tree species richness in forested habitats.The most consistent effects of Region were the low climate-

and stem density-adjusted species richness for both DBH ≥1 and≥10 cm trees in the three European plots and the NorthAmerican forests compared with forests in temperate Asia (Fig.3). The relatively low diversity in North America and, especially,in Europe compared with the forests of eastern Asia is wellknown. Explanations include more moderate effects of late Cenozoicclimate cooling and glaciation in eastern Asia (see ref. 10 for effectsin Europe), lower time-integrated regional area in Europe (32),extinction related to mountain building and increasing dryness inWestern North America (35), and the geographic complexity oftemperate eastern Asia, providing more opportunity for speciesformation (8). It is striking that the effect on species richness ofregional processes appears to extend down to the local scale,emphasizing the spatial and temporal continuity of pattern andprocess (36). The effect of isolation on the local diversity offorests on several islands around the globe is also evident inour analyses.African forests are considered to be less diverse than tropical

Asian and tropical American forests (37–40); however, the Af-rican forests represented in the CTFS forest plots have compa-rable species richness to those elsewhere. Possibly the few Africanplots were selected for their high diversity compared with drierforests elsewhere on the continent. The climate variables at theAfrican CTFS sites group closely with those at the tropical Asianand Neotropical sites (SI Appendix, Fig. S2). Given the matchingclimates at the tropical sites, and acknowledging the small numberof selected African plots, the CTFS plots provide no evidence ofregion effects influencing the diversity of trees across the tropics.Variation in AET, which is associated with habitat productivity,

and PET, which measures the heat load (water evaporation po-tential) of the environment, are often associated with variation inregional species richness (e.g., refs. 23 and 25). However, in thepresence of other related climate variables, AET and PET werenot retained by our model selection. Our results suggest that treespecies richness is influenced by energy and water flux through theenvironment (and their synergetic combination), as reflected bythe variables our models selected (e.g., BIO8, mean temperatureof wettest quarter; and BIO10, mean temperature of warmestquarter). However, some climate variables, such as BIO3 (iso-thermality = daily temperature range divided by the annual tem-perature range) and BIO15 (precipitation seasonality), which arenot correlated with AET and PET across the forest locations, areretained in our models for forest tree species richness. This resultsuggests that energy and water alone are probably not sufficientpredictors; absence of seasonal variation in energy (BIO3) andwater (BIO15) also plays an important role in determining theglobal richness pattern, consistent with the analysis of Kreft andJetz (25).Finally, although it is clear that regional processes, along with

variation in climate and sample size, jointly explain variation inlocal diversity of forest trees, it is important to note the difficultyof separating these effects statistically. Climate variables showstrong latitudinal gradients, which are associated with the per-vasive pattern of decreasing diversity with increasing latitude (SIAppendix, Fig. S3), and they also differ consistently between re-gions at the same latitude, particularly within temperate latitudes

Table 1. The negative binomial generalized linear regressionfor modeling region effects on species richness with DBH ≥ 1 cmand DBH ≥ 10 cm

Explanatory variables β̂ SE z Pr(>jzj)DBH ≥ 1 cm

Intercept 5.801 0.217 26.6 <2e-16Africa 0.256 0.397 0.7 0.52Europe −3.071 0.422 −7.3 3.48e-13Nearctic −2.087 0.302 −6.9 4.85e-12Temperate Asia −0.946 0.270 −3.5 4.66e-04Tropical Asia 0.217 0.269 0.8 0.42

DBH ≥ 10 cmIntercept 5.465 0.235 23.3 <2e-16Africa 0.213 0.497 0.4 0.67Europe −3.268 0.549 −6.0 2.59e-09Nearctic −2.407 0.341 −7.1 1.64e-12Temperate Asia −0.951 0.292 −3.3 0.0011Tropical Asia 0.298 0.291 1.0 0.31

The log link function is logðμÞ= βx, where β is the vector of regressioncoefficients (including the intercept) and x is the character region variable(Africa, Europe, Nearctic, Neotropical, Temperate Asia, Tropical Asia). Be-cause the variables have been standardized to a range of 0–1, the β coeffi-cients are standardized effect sizes. The region effects were coded inreference to the Neotropical forests (the intercept). Compared with thebaseline forests, European, North American, and Temperate Asian forestshave reduced tree diversity.

676 | www.pnas.org/cgi/doi/10.1073/pnas.1523683113 Ricklefs and He

Dow

nloa

ded

by g

uest

on

June

21,

202

0

(SI Appendix, Fig. S2). Because the forest plots in each regionare distributed across a unique range of climate variables, thequalitative variable region is redundant, to some degree, on thequantitative climate variables, and region effects per se tend tobe obscured in multiple regression models. Previous studiesfailing to find region effects at the local scale may have con-founded region and climate in their analyses.Differences in diversity between regions, independent of pre-

sent climate, are broadly appreciated and are generally attrib-uted to region size and unique history and geography (8, 32, 33,40), including the recent influence of Pleistocene glaciations (10,41–43) at temperate latitudes. Regional diversity is built up throughan excess of species formation over extinction. The rates of boththese processes presumably vary among regions owing to dif-ferences in their size and geographic complexity, leading to dif-ferent regional levels of diversity (32). How species are sortedwithin regions, which establishes patterns of local (alpha) di-versity and turnover of species between locations (beta diversity),should depend on how populations interact, primarily throughcompetition for resources, which can also promote evolutionarydiversification over environmental gradients. The strong statis-tical influence of region on local species richness suggests thatmembership in local communities is not limited locally, butrather responds to regional pressures of species production andthe spread of populations within regions to occupy extendedgeographic areas and ranges of ecological conditions.

Local diversity is the outcome of complex interactions of localand regional processes, as emphasized by the present analysis.The ambiguity of the local–regional relationship in the ecologicalliterature arises, in large part, from the difficulty of separatingprocesses acting on local and regional scales. This problem isexacerbated by the uneven sampling effort associated with muchempirical data, which typically have been gathered for otherpurposes, and by the unique climate variables associated witheach biogeographic region. Nonetheless, understanding the na-ture of the local–regional diversity relationship is essential forecologists seeking to integrate theory addressing the maintenanceof diversity at the local community scale, including species–habitatrelationships and species interactions, and at the regional level,including continental-scale attributes that influence species for-mation and extinction.

Materials and MethodsSince the early 1980s, when Steven Hubbell established the first 50-ha plot onBarro Colorado Island, Panama (44, 45), the CTFS of the Smithsonian In-stitution has developed a global network of forest plots across Africa,Europe, Asia, and the Americas. The network comprises of more than 60plots within which it monitors approximately 6 million free-standing trees/shrubs with DBH ≥ 1 cm, belonging to nearly 10,000 species. We compiledspecies richness data for 47 of these plots (Fig. 1 and SI Appendix, Table S1).Most of the plots are 5–50 ha in area and contain an average of 1.1 × 105

individuals (≥1 cm DBH) and 1.4 × 104 individuals (≥10 cm DBH). We ana-lyzed the data for the two size classes of individuals separately; the ≥1 cm

Fig. 3. Magnitude of the effects of region, number of plot stems, and significant local climate and plot variables predicting species richness. (Upper) Coefficientsof the negative binomial generalized linear models (Table 2) and their 95% confidence intervals for DBH ≥ 1 cm (Left) and 10 cm (Right), respectively. Neotropicalsites define the baseline for the region effects. (Lower) Relationships between the observed species richness and the richness predicted from the correspondinggeneralized linear models for the two DBH classes. If the models predicted species richness perfectly, the points would fall on the diagonal lines.

Ricklefs and He PNAS | January 19, 2016 | vol. 113 | no. 3 | 677

ECOLO

GY

Dow

nloa

ded

by g

uest

on

June

21,

202

0

DBH data include the ≥10 cm DBH data and, thus, are not completelyindependent.

Most of the plot data were accompanied by local measurements of thephysical environment: elevation, annual precipitation, mean annual tem-perature, and number of dry months (SI Appendix, Table S1) (28). For each ofthe 47 plots, we also downloaded climate variables from www.worldclim.org/bioclim, a dataset widely used for ecological niche modeling (46, 47).The 19 bioclimatic variables, BIO1–BIO19, describe average temperature andprecipitation, and seasonal variation in temperature and precipitation (48).A canonical correlation analysis between 14 nonredundant bioclimatic var-iables (SI Appendix, Table S1) and the locally measured climate variables

yielded three significant canonical correlations with adjusted R2 valuesranging from 0.989 (with local mean annual temperature) to 0.682 (localnumber of dry months) (SI Appendix, Table S2). Thus, the bioclimatic vari-ables, which were available for all of the plots, provide an adequate de-scription of the local climate, particularly temperature. Accordingly, theexplanatory variables included in this analysis were as follows:

(i) Local plot characteristics: latitude; plot area; number of stems; eleva-tion; elevation range within each plot, which is related to local heteroge-neity (49), although additional heterogeneity related to variation in soils(50) is not accounted for in this analysis.

(ii) Local climate variables: the 19 bioclimatic variables (48) plus AET andPET (compiled from the Global Evapotranspiration and Water Balance DataSets of Ahn and Tateishi; ref. 51). Because some of the bioclimatic variables,AET, and PET are highly intercorrelated, we excluded variables havingpairwise Pearson correlation coefficients > 0.95 to avoid collinearity. Twelvebioclimatic variables plus AET and PET were finally included in our analyses(SI Appendix, Table S1).

(iii) Region: a character variable indicating tropical Africa, temperateEurope, tropical Asia, temperate Asia, tropical America (Neotropics), andNearctic (North America). We conducted additional analyses with the Ne-arctic separated into eastern and western temperate North America, butobtained almost identical results. The models without separation wereslightly favored by AIC and are therefore reported in this study.

(iv) Isolation: plots on islands (Taiwan, Hainan Island, the Philippines,Borneo, Puerto Rico, Sri Lanka, and New Guinea) were coded as island(isolation) in contrast to the continental plots (nonisolation).

Plot area and number of stems were log-transformed to account for powerrelationships with species richness. All other numeric variables (x) werestandardized to 0–1 range by (x − xmin)/(xmax − xmin).

Because species richness values for the 47 plots exhibit a negative binomialdistribution (SI Appendix, Fig. S1), we used negative binomial generalizedlinear models with a log link function to model richness as a function of theregion effects, isolation, stem density, areal effect, and local climatic andenvironmental variables. We initially modeled species richness as a functionof region (Fig. 2 and Table 1). Because the six regions can be distinguished bydifferent climate conditions (SI Appendix, Fig. S2), region effects can beconfounded with local climate and habitat conditions. Accordingly, we ex-tended the region effects model to include all of the local climate and plotvariables listed in SI Appendix, Table S1.

A stepwise selection procedure was used for model selection based on AIC.The importance of a variable can be evaluated by comparing a full model thatincludes all related variables and a reduced model excluding the variable, orvariables, of interest. The deviance difference between the two modelsfollows a χ2 distribution.

Statistical analyses in this study were conducted by using R (www.r-project.org/). The function “glm.nb” in the package MASS was used to model thenegative binomial generalized linear models.

ACKNOWLEDGMENTS. We thank many people for generously providingus with their plot data for this work, including Norm Bourg, SarayudhBunyavejchewin, Deliang Chen, Rick Condit, Stuart Davies, Gregory Gilbert,Dong He, Robert Howe, Patrick Jansen, Minxi Jiang, Guangze Jin, Kamil Kral,Yiching Lin, Yanyan Liu, Zhengrong Luo, Jim Lutz, William J. McShea,Vojtech Novotny, Alexandre A. de Oliveira, Jan den Ouden, I-Fang Sun, SeanThomas, Xihua Wang, Yi-Hui Wang, George Weiblen, Amy Wolf, Han Xu,Yan Zhu, and many others who have contributed to establishing the theCenter for Tropical Forest Science (CTFS) global plots. Xinghua Sui preparedFig. 1. This work was supported by Sun Yat-sen University, the NaturalSciences and Engineering Research Council (Canada), and the Curators of theUniversity of Missouri.

1. Lortie CJ, et al. (2004) Rethinking plant community theory. Oikos 107(2):433–438.

2. Ricklefs RE (2004) A comprehensive framework for global patterns in biodiversity.Ecol Lett 7(1):1–15.

3. Weiher E, Keddy PA (1995) Assembly rules, null models, and trait dispersion: Newquestions from old patterns. Oikos 74(1):159–164.

4. Cavender-Bares J, Kozak KH, Fine PVA, Kembel SW (2009) The merging of communityecology and phylogenetic biology. Ecol Lett 12(7):693–715.

5. Harrison S, Cornell H (2008) Toward a better understanding of the regional causes oflocal community richness. Ecol Lett 11(9):969–979.

6. Cornell HV, Harrison SP (2012) Regional effects as important determinants of localdiversity in both marine and terrestrial systems. Oikos 122(2):288–297.

7. White EP, Hurlbert AH (2010) The combined influence of the local environment andregional enrichment on bird species richness. Am Nat 175(2):E35–E43.

8. Qian H, Ricklefs RE (2000) Large-scale processes and the Asian bias in species diversityof temperate plants. Nature 407(6801):180–182.

9. Ricklefs RE, Schwarzbach AE, Renner SS (2006) Rate of lineage origin explains thediversity anomaly in the world’s mangrove vegetation. Am Nat 168(6):805–810.

10. Svenning JC (2003) Deterministic Plio-Pleistocene extinctions in the European cool-temperate tree flora. Ecol Lett 6(7):646–653.

11. Srivastava D (1999) Using local-regional richness plots to test for species saturation:Pitfalls and potentials. J Anim Ecol 68(1):1–16.

12. Loreau M (2000) Are communities saturated? On the relationship between α, β, and γdiversity. Ecol Lett 3:73–76.

13. Cornell HV (1999) Unsaturation and regional influences on species richness in eco-logical communities: A review of the evidence. Ecoscience 6(3):303–315.

14. Cornell HV, Lawton JH (1992) Species interactions, local and regional processes, and limitsto the richness of ecological communities: A theoretical perspective. J Anim Ecol 61:1–12.

15. Francis AP, Currie DJ (2003) A globally consistent richness-climate relationship forangiosperms. Am Nat 161(4):523–536.

16. Huston MA (1999) Local processes and regional patterns: Appropriate scales for un-derstanding variation in the diversity of plants and animals. Oikos 86(3):393–401.

Table 2. The output of the negative binomial generalized linearmodel for species richness with DBH ≥ 1 cm and ≥10 cm

Explanatory variables β̂ SE z Pr(>jzj)DBH ≥ 1 cm)Intercept 0.335 0.729 0.460 0.65Region effect

Africa −0.113 0.223 −0.505 0.61Europe −1.232 0.372 −3.310 9.31e-04Nearctic −0.653 0.253 −2.580 9.89e-03Temperate Asia 0.307 0.222 1.385 0.17Tropical Asia 0.377 0.152 2.483 0.013Island −0.407 0.117 −3.472 5.17e-04

log(ntree1) 0.347 0.055 6.355 2.08e-10Isothermality (BIO3) 1.253 0.289 4.337 1.44e-05Mean temp of warmest

quarter (BIO10)0.993 0.236 4.207 2.59e-05

Precipitation seasonality(BIO15)

−0.647 0.201 −3.215 1.30e-03

DBH ≥ 10 cmIntercept −0.599 0.873 −0.685 0.49Region effect

Africa −0.085 0.276 −0.307 0.76Europe −1.786 0.456 −3.913 9.11e-05Nearctic −0.823 0.337 −2.443 0.015Temperate Asia 0.333 0.256 1.303 0.19Tropical Asia 0.551 0.169 3.263 1.10e-03Island −0.431 0.134 −3.209 1.33e-03

log(ntree10) 0.424 0.069 6.163 7.14e-10Isothermality (BIO3) 1.370 0.345 3.976 7.01e-05Mean temp of wettest

quarter (BIO8)1.472 0.402 3.658 2.54e-04

Precipitation seasonality(BIO15)

−0.842 0.237 −3.546 3.91e-04

The log link function is logðμÞ= βx, where β is the vector of regressioncoefficients (including the intercept) and x is the vector of explanatory var-iables. The region effects were coded in reference to the Neotropical forests(the intercept). The effect of island (i.e., isolation effect) is in reference tothe mainland plots; −0.407 for the island plots (for DBH ≥ 1 cm) indicatesthat species richness on islands is significantly (P = 0.0005) lower than that inthe mainland plots (given that the effects of other variables were accountedfor). The model was selected by stepwise procedure using the criterion of AIC.

678 | www.pnas.org/cgi/doi/10.1073/pnas.1523683113 Ricklefs and He

Dow

nloa

ded

by g

uest

on

June

21,

202

0

17. Karlson RH, Cornell HV (2002) Species richness of coral assemblages: Detecting re-gional influences at local spatial scales. Ecology 83(2):452–463.

18. Belmaker J, Jetz W (2015) Relative roles of ecological and energetic constraints, di-versification rates and region history on global species richness gradients. Ecol Lett18(6):563–571.

19. Eiserhardt WL, Borchsenius F, Plum CM, Ordonez A, Svenning JC (2015) Climate-drivenextinctions shape the phylogenetic structure of temperate tree floras. Ecol Lett 18(3):263–272.

20. Kerkhoff AJ, Moriarty PE, Weiser MD (2014) The latitudinal species richness gradientin New World woody angiosperms is consistent with the tropical conservatism hy-pothesis. Proc Natl Acad Sci USA 111(22):8125–8130.

21. Lessard JP, et al. (2012) Strong influence of regional species pools on continent-widestructuring of local communities. Proc Biol Sci 279(1727):266–274.

22. Latham RE, Ricklefs RE (1993) Global patterns of tree species richness in moist forests:energy-diversity theory does not account for variation in species richness. Oikos 67(2):325–333.

23. Currie DJ, Paquin V (1987) Large-scale biogeographical patterns of species richness oftrees. Nature 329:326–327.

24. Francis AP, Currie DJ (1998) Global patterns of tree species richness in moist forest:Another look. Oikos 81:598–602.

25. Kreft H, Jetz W (2007) Global patterns and determinants of vascular plant diversity.Proc Natl Acad Sci USA 104(14):5925–5930.

26. Simová I, et al. (2011) Global species-energy relationship in forest plots: Role ofabundance, temperature and species climatic tolerances. Glob Ecol Biogeogr 20(6):842–856.

27. Baraloto C, et al. (2013) Rapid simultaneous estimation of aboveground biomass andtree diversity across Neotropical forests: A comparison of field inventory methods.Biotropica 45(3):288–298.

28. Anderson-Teixeira KJ, et al. (2015) CTFS-ForestGEO: A worldwide network monitoringforests in an era of global change. Glob Change Biol 21(2):528–549.

29. Allen AP, Gillooly JF (2006) Assessing latitudinal gradients in speciation rates andbiodiversity at the global scale. Ecol Lett 9(8):947–954.

30. Currie DJ, et al. (2004) Predictions and tests of climate-based hypotheses of broad-scale variation in taxonomic richness. Ecol Lett 7(12):1121–1134.

31. Gaston KJ (2000) Global patterns in biodiversity. Nature 405(6783):220–227.32. Fine PVA, Ree RH (2006) Evidence for a time-integrated species-area effect on the

latitudinal gradient in tree diversity. Am Nat 168(6):796–804.33. Jetz W, Fine PV (2012) Global gradients in vertebrate diversity predicted by historical

area-productivity dynamics and contemporary environment. PLoS Biol 10(3):e1001292.

34. Kreft H, Jetz W (2010) A framework for delineating biogeographical regions based onspecies distributions. J Biogeogr 37(11):2029–2053.

35. Wolfe JA (1994) Tertiary climatic changes at middle latitudes of western NorthAmerica. Palaeogeogr Palaeoclimatol Palaeoecol 108(3–4):195–205.

36. Ricklefs RE (2008) Disintegration of the ecological community. Am Nat 172(6):741–750.

37. Gentry AH (1988) Patterns of plant community diversity and floristic composition onenvironmental and geographical gradients. Ann Mo Bot Gard 75:1–34.

38. Whitmore TC (1984) Tropical Rain Forests of the Far East (Clarendon, Oxford).39. Parmentier I, et al. (2007) The odd man out? Might climate explain the lower tree

α-diversity of African rain forests relative to Amazonian rain forests? J Ecol 95(5):1058–1071.

40. Slik JWF, et al. (2015) An estimate of the number of tropical tree species. Proc NatlAcad Sci USA 112(24):7472–7477.

41. Latham RE, Ricklefs RE (1993) Continental comparisons of temperate-zone tree spe-cies diversity. Species Diversity: Historical and Geographical Perspectives, eds Ricklefs RE,Schluter D (Univ of Chicago Press, Chicago), pp 294–314.

42. Sauer JD (1988) Plant Migration. The Dynamics of Geographic Patterning in SeedPlant Species (Univ of California Press, Berkeley, CA).

43. Svenning J-C, Skov F (2007) Could the tree diversity pattern in Europe be generatedby postglacial dispersal limitation? Ecol Lett 10(6):453–460.

44. Hubbell SP, Foster RB (1992) Short-term dynamics of a neotropical forest: Why eco-logical research matters to tropical conservation and management. Oikos 63:48–61.

45. Condit R (1995) Research in large, long-term tropical forest plots. Trends Ecol Evol10(1):18–22.

46. Hijmans RJ, Graham CH (2006) The ability of climate envelope models to predict theeffect of climate change on species distributions. Glob Change Biol 12(12):1–10.

47. Elith J, Leathwick JR (2009) Species distribution models: Ecological explanation andprediction across space and time. Annu Rev Ecol Evol Syst 40:677–697.

48. Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A (2005) Very high resolution in-terpolated climate surfaces for global land areas. Int J Climatol 25(15):1965–1978.

49. Brown C, et al. (2013) Multispecies coexistence of trees in tropical forests: Spatialsignals of topographic niche differentiation increase with environmental heteroge-neity. Proc Biol Sci 282(1800):20130502.

50. Baldeck CA, et al. (2012) Soil resources and topography shape local tree communitystructure in tropical forests. Proc Biol Sci 280(1753):20122532.

51. Ahn C-H, Tateishi R (1994) Development of a global 30-minute grid potentialevapotranspiration data set. Photogrammetry and Remote Sensing 33(2):12–21.

Ricklefs and He PNAS | January 19, 2016 | vol. 113 | no. 3 | 679

ECOLO

GY

Dow

nloa

ded

by g

uest

on

June

21,

202

0