Received: 30 September 2019  |  Revised: 7 February 2020  |  Accepted: 16 February 2020 DOI: 10.1111/geb.13089 R E S E A R C H P A P E R Biomes as evolutionary arenas: Convergence and conservatism in the trans-continental succulent biome Jens J. Ringelberg1  | Niklaus E. Zimmermann2,3  | Andrea Weeks4  | Matt Lavin5  | Colin E. Hughes1 1Department of Systematic and Evolutionary Botany, University of Zurich, Zurich, Abstract Switzerland Aim: Historically, biomes have been defined based on their structurally and function- 2Swiss Federal Research Institute WSL, ally similar vegetation, but there is debate about whether these similarities are super- Birmensdorf, Switzerland 3Environmental Systems Science, Swiss ficial, and about how biomes are defined and mapped. We propose that combined Federal Institute of Technology ETH, Zurich, assessment of evolutionary convergence of plant functional traits and phylogenetic Switzerland biome conservatism provides a useful approach for characterizing biomes. We focus 4Department of Biology, George Mason University, Fairfax, VA, USA on the little-known succulent biome, a trans-continentally distributed assemblage of 5Plant Sciences and Plant Pathology, succulent-rich, drought-deciduous, fire-free forest, thicket and scrub vegetation as a Montana State University, Bozeman, MT, USA useful exemplar biome to gain insights into these questions. Location: Global lowland (sub)tropics. Correspondence Jens J. Ringelberg, Department of Time period: Present. Systematic and Evolutionary Botany, Major taxa studied: Angiosperms. University of Zurich, Zollikerstrasse 107, Zurich 8008, Switzerland. Methods: We use a model ensemble approach to model the distribution of 884 spe- Email: jens.ringelberg@gmail.com cies of stem succulents, a plant functional group representing a striking example of Funding information evolutionary convergence. Using this model, phylogenies, and species occurrence Schweizerischer Nationalfonds zur data, we quantify phylogenetic succulent biome conservatism for 10 non-succulent Förderung der Wissenschaftlichen Forschung, Grant/Award Number: trans-continental plant clades including prominent elements of the succulent biome, 31003A_156140 and 31003A_182453/1; representing over 800 species. National Science Foundation, Grant/Award Number: DEB-0919179 and DEB-1403150 Results: The geographical and climatic distributions of stem succulents provide an objective and quantitative proxy for mapping the distribution of the succulent biome. Editor: Angela Moles High fractions of succulent biome occupancy across continents suggest all 10 non- succulent study clades are phylogenetically conserved within the succulent biome. Main conclusions: The trans-continental succulent and savanna biomes both show evolutionary convergence in key biome-related plant functional traits. However, in contrast to the savanna biome, which was apparently assembled via repeated local recruitment of lineages via biome shifts from adjacent biomes within continents, the succulent biome forms a coherent trans-continental evolutionary arena for drought- adapted tropical biome conserved lineages. Recognizing the important functional differences between the succulent-rich, grass-poor, fire-free succulent biome and the grass-dominated, succulent-poor, fire-prone savanna biome, and defining them This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2020 The Authors. Global Ecology and Biogeography published by John Wiley & Sons Ltd 1100  |  wileyonlinelibrary.com/journal/geb Global Ecol Biogeogr. 2020;29:1100–1113. RINGELBERG Et aL.      |  1101 as distinct seasonally dry tropical biomes, occupying essentially non-overlapping dis- tributions, provides critical insights into tropical biodiversity and the extent of biome stasis versus biome shifting. K E Y W O R D S biome definition, evolutionary arena, evolutionary convergence, lowland tropics, phylogenetic biome conservatism, stem succulence, succulent biome, trans-continental disjunction 1  | INTRODUC TION deserts, most research on tropical biodiversity divides the vegeta- tion of the lowland tropics into two broad categories: rain forests and Biomes have been defined as ‘globally distributed, structurally and savannas, or forest versus ‘open’ vegetation (Antonelli et al., 2018; functionally similar vegetation units’ (Moncrieff, Bond, & Higgins, Oliveras & Malhi, 2016; Staver, Archibald, & Levin, 2011). Savannas 2016), but the significance of these similarities remains much de- are found in areas with a pronounced dry season and abundant C4 bated. Some have argued that geographically disjunct areas of the grasses, and are prone to regular fires (Lehmann et al., 2019; Lehmann same biome are ecologically and evolutionarily independent, and & Parr, 2016). Rain forests, in contrast, receive year-round precipita- that any similarity is merely superficial (Corlett & Primack, 2006). tion, have few grasses, and do not experience fire (Eiserhardt et al., Others consider biomes to represent coherent evolutionary arenas 2017). However, a third major lowland tropical biome, the season- (sensu Nürk et al., 2019) with distinct histories and biotic and abiotic ally dry, drought-deciduous, but grass-poor and fire-free vegetation characteristics that shape and confine the evolution and distribu- referred to as the succulent biome sensu Schrire, Lavin, and Lewis, tion of the lineages inhabiting them (Eiserhardt, Couvreur, & Baker, 2005, has been recognized (Dexter et al., 2018; Gagnon, Ringelberg, 2017; Hughes, Pennington, & Antonelli, 2013; Moncrieff et al., 2016; Bruneau, Lewis, & Hughes, 2019; Lavin et al., 2004; Oliveira-Filho Pennington, Lavin, & Oliveira-Filho, 2009; Pennington, Lehmann, & et al., 2013; Schrire et al., 2005), but remains little known, lacks a Rowland, 2018). Detailed analyses of the lineages composing a biome quantitative distribution map, and is often neglected. can help to distinguish between these two contrasting hypotheses. Despite this neglect, the succulent biome includes import- If biomes coincide with distinct evolutionary arenas, due to similar ant hotspots of endemism (DRYFLOR, 2016; Marshall, Wieringa, selection pressures and a partially shared history, disjunct regions of & Hawthorne, 2016), most notably in Africa, where the hottest the same biome should share functional traits and functional groups, hotspots of range-restricted plant endemism are in the succu- that is, show a strong signal of evolutionary convergence across dis- lent biome (Marshall et al., 2016). The succulent biome is thus a tantly related lineages. If this is the case, these functional traits could high conservation priority, yet large portions of it are threatened also play an important role in defining and mapping biomes. In ad- (DRYFLOR, 2016; Kuemmerle et al., 2017; Pennington et al., 2018), dition, geographically disjunct regions of a biome should also share and neglected as conservation priorities compared to rain forests evolutionary lineages exhibiting phylogenetic biome conservatism, (Pennington et al., 2018). Ignorance of the succulent biome is also that is, lineages made up of species that retain ancestral ecological evident in many prominent recent studies investigating assembly traits and environmental distributions (Crisp et al., 2009; Donoghue of tropical diversity (e.g. Antonelli et al., 2018; Charles-Dominique & Edwards, 2014), and that are significantly confined within a biome et al., 2016), which often conflate the succulent biome with the cli- by strong adaptive barriers, but trans-continentally distributed matically similar but functionally distinct savanna biome. There is across its distribution. While striking examples of large-scale phy- thus a clear need for an objective quantitative map of the succulent logenetic biome conservatism have been demonstrated (e.g. Crisp biome, and a better understanding of the differences between the et al., 2009; Donoghue, 2008; Lavin et al., 2004; Segovia et al., succulent and savanna biomes (Pennington et al., 2018). 2019), the overall extent and prevalence of phylogenetic biome A defining characteristic of the succulent biome is the presence conservatism versus biome shifting remain debatable (Donoghue & of large stem succulents, that is, plants over 1 m height with stems Edwards, 2014; Edwards & Donoghue, 2013). Here we investigate containing specialized water-storing tissue. This functional plant these two important attributes of biomes, evolutionary convergence group distinguishes the succulent biome from savannas and rain and phylogenetic conservatism, to test to what extent biomes can be forests. Savannas dominated by C4 grasses experience regular fires considered as distinct evolutionary arenas. with most savanna species adapted to tolerate or resist fire (Maurin Quantifying phylogenetic biome conservatism and the conver- et al., 2014; Ratnam et al., 2011; Simon et al., 2009). Stem succulents gence of traits within a biome depends on how biomes are defined are notably vulnerable to fire (Cousins & Witkowski, 2012; Moe, and mapped, but there is no universally accepted framework for this Mobæk, & Narmo, 2009; Thomas, 1991) and almost absent from sa- (Higgins, Buitenwerf, & Moncrieff, 2016; Moncrieff et al., 2016), nor is vannas. Similarly, in rain forests stem succulent species are also typ- there a general agreement about the number and location of tropical ically absent. Stem succulence is a textbook example of evolutionary biomes. As pointed out by Dexter et al. (2018) and setting aside true convergence, occurring in over 20 plant families (Ávila-Lovera & 1102  |     RINGELBERG Et aL. Ezcurra, 2016; Eggli & Nyffeler, 2009; Griffiths & Males, 2017; Four stem succulent richness maps were generated: one by Ogburn & Edwards, 2010), and occupying a highly disjunct distri- counting the number of unique genera per 0.25° lat./long. grid bution across the tropics (Ávila-Lovera & Ezcurra, 2016; Ellenberg, cell, two by varying the grid cell sizes to 0.1° and 0.5°, and one by 1981; Ogburn & Edwards, 2010). counting unique species rather than genera. Since the global dis- Here we investigate the combination of evolutionary conver- tribution of stem succulent richness on these maps was strongly gence of plant functional traits and phylogenetic conservatism of influenced by the spatial sampling bias of the GBIF data (Meyer, lineages as a way to test the hypothesis that biomes can be defined Weigelt, & Kreft, 2016; which accounted for c. 75% of the data), and characterized as evolutionary arenas. We focus on the neglected we modelled the distribution of stem succulents, rather than using and little-known succulent biome as a particularly suitable system to it directly in further analyses. The four richness maps were used test this hypothesis, because it occupies a highly disjunct, trans-con- to generate four independent models, but since the results are tinental distribution. We generate an objective, quantitative global highly similar (Supporting Information Figures S1–S4), we discuss map of this biome to test the hypotheses that (a) the distribution the methods and results in detail only for the 0.25° genus map. of large stem succulent plant species, a prominent example of evo- lutionary convergence, is a good proxy for the distribution of the succulent biome, and (b) multiple non-succulent angiosperm clades 2.1.2 | Predictor variables that include common and dominant species in the succulent biome are phylogenetically conserved across this biome. Twenty-two climatic and topographic predictor variables were ob- tained from climatologies at high resolution for the earth’s land sur- face areas (CHELSA) 1.2 (Karger et al., 2017), WorldClim 2.0 (Fick 2  | METHODS & Hijmans, 2017) and EarthEnv.org (Amatulli et al., 2018; Wilson & Jetz, 2016) (Supporting Information Table S2), and aggregated to the Data handling and analyses were performed in R versions 3.3.2– same spatial resolution as the stem succulent richness maps using 3.6.0 (R Core Team, 2018). the ‘mean’ option of the aggregate function of the ‘raster’ package (Hijmans, 2018). In order to avoid multicollinearity problems when fitting models, we reduced the number of predictor variables by 2.1 | Stem succulent modelling calculating variance inflation factors (VIFs) and Pearson’s correla- tions between all predictors, removing those with VIFs ≥ 7 and 2.1.1 | Stem succulent occurrence data correlations ≥ 0.7, which resulted in 13 predictors being retained (Supporting Information Table S3). We assembled a checklist of accepted names and synonyms of 1,120 succulent species listed in several encyclopaedic compendia (Albers & Meve, 2004; Anderson, 2001; Eggli, 2001, 2004). We de- 2.1.3 | Distribution modelling fined stem succulents as plants with stems containing specialized water-storing tissue, including pachycaul species [e.g. Ceiba insignis We first produced a global model using all stem succulent occur- (Malvaceae)] and arborescent/caulescent giant-leaf succulents [e.g. rence points (Supporting Information Figure S5). However, this Aloe dichotoma (Asphodelaceae)]. As climbing, clambering, pendant, model yielded a poor match with the occurrence data in several epiphytic and shrubby (<1 m tall) succulent species do not form areas, such as the Sahara and Saudi Arabia. Therefore, we decided structural components of vegetation indicative of fire-free ecologies to model the distribution of stem succulents separately for the New and are more likely to occur in azonal areas and atypical climates, World and the Old World (Africa, Madagascar and Arabia). We ex- these species were excluded. tracted presence points from the stem succulent richness map at We downloaded occurrence records from the Global Biodiversity two levels of richness, i.e., with at least two or three unique gen- Facility (GBIF; www.gbif.org; see Supporting Information Table S1 era per grid cell. This resulted in 2,210 and 1,270 presence points, for DOIs), the Latin American Seasonally Dry Tropical Forest Floristic respectively, for the New World and 640 and 256 presence points Network (DryFlor; http://www.dryfl or.info) and the Southwestern for the Old World. For the whole of Australia and Asia, there were Environmental Information Network (http://swbio divers ity.org/ only five and zero presence points, which we deemed to be too few seinet). For c. 200 species on the checklist no occurrence data were to model properly. available. We performed extensive data cleaning, assigning syn- We used a model ensemble approach to map the spatial distri- onyms to their accepted names, and removing records not based on bution of stem succulents (Thuiller, Lafourcade, Engler, & Araújo, vouchered herbarium collections, those with imprecise coordinates 2009), selecting: (a) a generalized linear model (GLM), (b) a gener- (i.e. only degrees, no minutes or seconds), located in the sea, or oc- alized additive model (GAM), and (c) a random forest (RF). By com- curring outside the known distributions of species given by Albers bining these three methods, different degrees of model complexity and Meve (2004); Anderson (2001); Eggli (2001, 2004), plus culti- are incorporated (Merow et al., 2014). See Supporting Information vated records and country centroids. for details. RINGELBERG Et aL.      |  1103 We used two different regions (the New World and the Old tested for biome conservatism using the same approaches as Gagnon World), two different levels of determining presence data (two and et al. (2019): comparing the total number of biome shifts to a null three genera per grid cell), three different models (GLM, GAM and distribution, and measuring phylogenetic signal. Biome shifts were RF), and two different thresholds (optimizing Kappa and true skill assessed with stochastic character mapping using the make.simmap statistic), generating a total of 24 unique models. The final map of function of the ‘phytools’ package (Revell, 2012), run for 200 simu- the occurrence of stem succulents was created by combining all 12 lations under equal rates (ER) and symmetrical (SYM) models, and unique binary output maps for each region. compared to numbers of shifts obtained by re-running make.simmap The few grid cells that were not deemed suitable for stem suc- 99 times with randomized tip states. An observed number of biome culents by the models but did contain at least two observed genera shifts that falls within the lowest 5% of biome shifts obtained from were manually assigned as presence. For the assignment of species the randomized distribution is considered evidence for conserva- to the succulent biome (see below), the continuous stem succulents tism (Maddison & Slatkin, 1991). Phylogenetic signal was quantified map was converted to binary with a threshold of 33% (partially fol- using the fitDiscrete function of the ‘geiger’ package (Harmon, Weir, lowing Chala et al., 2016). Brock, Glor, & Challenger, 2008), which measures Pagel’s lambda (Pagel, 1999). Before running fitDiscrete the optimal character evo- lution model [ER, SYM or all rates different (ARD)] for each clade was 2.2 | Succulent biome conservatism determined using ‘phytool’s’ fitMk function. Some phylogenies con- tained polytomies, which were randomly resolved 10 different times 2.2.1 | Study clades using the multi2di function of the ‘ape’ package (Paradis & Schliep, 2019), creating a set of fully resolved trees. As fitDiscrete does not We identified 10 non-succulent amphi-Atlantic plant clades as abun- allow polymorphic character states, in clades with species assigned dant and prominent elements in the succulent biome, some from to more than one biome phylogenetic signal was determined sepa- previous studies (Gagnon et al., 2019; Lavin et al., 2004; Thiv et al., rately for each possible combination of biome states for polymorphic 2011). With the exception of 16 species of Bursera and Commiphora, species (with the exception of the Caesalpinia group, where the total none of the species of these clades was used for generating the stem number of combinations was very large, and fitDiscrete was run for succulents map. Per clade we assembled a taxonomic checklist (see 100 randomly selected combinations of polymorphic biome states Supporting Information) and occurrence data set using the same ap- instead). For clades with sets of trees, sets of possible biome states, proach as for the stem succulents (Supporting Information Table S1). or both, the median phylogenetic signal was calculated. High phylo- All species with over half of their occurrences co-occurring with genetic signal was interpreted as an indication of conservatism (but stem succulents (omitting multiple occurrences of the same species see Losos, 2008; Revell, Harmon, & Collar, 2008). at the same location) were assigned to the succulent biome, and re- maining species were assigned to savanna, rain forest, temperate or coastal biomes, based on their distribution and habitat information. 3  | RESULTS Biome assignments based on the four different biome modelling ap- proaches were highly similar (Supporting Information Figure S6). In 3.1 | Stem succulent modelling some cases, species were assigned to more than one biome, and in rare cases, species with over half their occurrences in the succulent We obtained 47,575 quality-controlled occurrence points for 884 biome were assigned to a different biome (11 out of 839 total spe- species and infraspecific taxa of stem succulents, from 102 genera cies), or vice versa (18 species), based on expert knowledge. and 22 families (Supporting Information Table S4). Different model- ling methods and ways of counting presences yielded highly consist- ent results (Supporting Information Figures S1–S4 and Table S3). The 2.2.2 | Phylogeny reconstruction combined analysis of 24 unique models showing the distribution of stem succulents (Figure 1) is consistent with previous studies (Ávila- Phylogenies from each study clade were obtained either by down- Lovera & Ezcurra, 2016; Ellenberg, 1981; Ogburn & Edwards, 2010), loading previously generated phylogenies, or by inferring them using and, with some notable exceptions, closely matches the distribution published molecular data. See Supporting Information for details. of the succulent biome depicted by Schrire et al. (2005). Climatically, the distribution of grid cells containing at least two genera of stem succulents closely matches the proposed climate 2.2.3 | Phylogenetic biome conservatism of the succulent biome (Dexter et al., 2018; Oliveira-Filho et al., 2013; Schrire et al., 2005; Silva de Miranda et al., 2018; Figure 2). Several of our study clades are trans-continentally distributed yet Ninety percent of cells containing stem succulents receive less than (almost) completely restricted to the succulent biome, providing de- 1,300 mm annual precipitation, close to the 1,200 mm upper limit finitive evidence of phylogenetic succulent biome conservatism, and for succulent biome in eastern South America proposed by Oliveira- precluding the need for any formal tests. For the remaining clades we Filho et al. (2013) (Figure 2d). The climatic overlap between New and 1104  |     RINGELBERG Et aL. Old World succulents is high (Figure 2a), in contrast to the mark- continents in general (Moncrieff, Hickler, & Higgins, 2015). The only edly incomplete climatic overlap between savannas on different exception to the pattern of overlap is that some New World succu- continents (Lehmann et al., 2014) and across biomes on different lents occur at higher elevations (Figure 2b), while some Old World RINGELBERG Et aL.      |  1105 F I G U R E 1   The trans-continental distribution of the succulent biome. Map showing the modelled probability distribution of stem succulents, averaged over 24 unique models (see Methods) with a cut-off at a minimum probability of occurrence of 33%. Colours on the map correspond to fraction of models predicting the occurrence of stem succulents in each grid cell. The inset shows Hawai’i. The distribution of stem succulents provides a proxy for the distribution of the succulent biome. The full global map showing the absence of stem succulents from most of Asia and Australia is in Supporting Information Figure S1. Images provide an overview of the vegetation, landscapes, and stem succulent species and growth forms that make up the succulent biome, illustrating both the convergent evolution of stem succulence, the key plant functional type defining this biome, and variation in stature and vegetation cover ranging from scrub and thicket vegetation with an open cover of small trees and shrubs, to closed deciduous seasonally dry tropical forest (SDTF). All these vegetation types are notably poor in grasses, rich in stem succulents and fire-free, a set of characteristics that distinguishes the succulent from the savanna biome: (1) Baja California, Mexico, cactus-rich dry thorn scrub; (2) Pacific coastal Oaxaca, Mexico, SDTF; (3) Loja province, southern Ecuador, pachycaul Malvaceae in SDTF; (4) the inter-Andean Marañon Valley, Peru, SDTF; (5) north-east Brazil, Caatinga vegetation (cactus-rich dry thorn scrub); (6) northern Argentina, dry thorn scrub with Cactaceae; (7) Tenerife, Canary Islands, Spain, tabaibal-cardonal (coastal succulent scrub) with Euphorbia canariensis (Euphorbiaceae); (8) Socotra, Yemen, succulent shrubland with Adenium obesum subsp. socotranum (Apocynaceae) and Dracaena cinnabari (Ruscaceae); (9) northern Kenya, Somali-Masai Acacia–Commiphora woodland with Euphorbia atroflora and Euphorbia magnicapsula subsp. lacertosa; (10) south-west Madagascar, spiny forest; (11) southern Namibia, desert scrub with Aloe dichotoma and Aloe littoralis (Asphodelaceae); (12) Eastern Cape, South Africa, succulent thicket vegetation with Euphorbia tetragona. Photos courtesy of: 1, 2 & 4 Colin Hughes, 3 Gwilym Lewis, 5 Domingos Cardoso, 6 Edeline Gagnon, 7 Mark Carine, 8 Luciano Napolitano (http://www.travel -tour-guide.com), 9 Frans Noltee (http://enjoy succu lents.com), 10 Andrea Weeks, 11 Roman Kellenberger, 12 Richard Cowling [Colour figure can be viewed at wileyonlinelibrary.com] succulents occur in areas that are slightly warmer, with higher solar Figures S22 and S23). This high trans-continental succulent biome radiation and evapotranspiration rates. In addition, occurrences of occupancy applies to six of the study clades, each with over 85% of succulent biome extend further into the Northern Hemisphere in its species in the succulent biome. While the remaining four clades the New World (Supporting Information Figure S11), and overall the have relatively fewer species in this biome, they still show moderate biome shows an intriguing bimodal latitudinal distribution concen- to high levels of succulent biome conservatism, as indicated by a high trated mainly between 10° and 30° north and south of the equator, phylogenetic signal and a low number of biome shifts compared to a presenting an exception to the general latitudinal diversity gradient. randomized biome distribution (Gagnon et al., 2019; Table 1). Many other plant and some animal clades also appear to show striking suc- culent biome conservatism (Supporting Information Table S5). 3.2 | Succulent biome conservatism Using occurrence data (83,373 records for 839 species; Figure 3) and 4  | DISCUSSION our newly generated map we quantified the extent to which species of 10 non-succulent amphi-Atlantic plant clades can be assigned to 4.1 | Biome mapping the succulent biome to assess phylogenetic succulent biome con- servatism (Table 1). A high fraction of succulent biome occupancy The occurrence of functionally and structurally similar but geo- combined with a trans-continental distribution provides strong evi- graphically disjunct and floristically distinct vegetation formations dence for biome conservatism. For example, Bursera is a ubiquitous, was first noted by von Humboldt over 200 years ago, and biomes abundant and often dominant tree in Mexican seasonally dry tropi- have played a prominent role in ecology, evolution and biogeog- cal forests (SDTFs; De-Nova et al., 2012), while its sister group, the raphy ever since (Moncrieff et al., 2016; Mucina, 2019). However, genus Commiphora, is equally abundant in dry parts of Africa and there is controversy about how biomes are defined and mapped Madagascar, lending its name to the Acacia–Commiphora woodlands of (Higgins et al., 2016; Moncrieff et al., 2016; Mucina, 2019). Most the Somali-Masai region (Supporting Information Figure S14) (Gostel, biome maps are based on combinations of climatic data, existing Phillipson, & Weeks, 2016; White, 1983). Ninety-three percent of the vegetation maps and expert opinion (Olson et al., 2001), or remotely c. 235 species in the Bursera-Commiphora clade are confined to the sensed vegetation types (Friedl et al., 2010). Here we use a differ- succulent biome across continents and there are two trans-Atlantic ent approach, by mapping a key plant functional group. Existing disjunctions spanning areas of succulent biome within this clade biome maps that explicitly take into account functionality are based (Table 1, Figure 3 and Supporting Information Figures S14 and S15). on mechanistic models (Prentice et al., 1992), regressions of trait Similarly, Parkinsonia and Delonix are common and conspicuous trees observations against climate (van Bodegom, Douma, & Verheijen, in dry parts of the Neotropics (Pérez-García, Meave, & Cevallos- 2014), or remotely sensed broad functional types (Higgins et al., Ferriz, 2012) and the dry spiny forests of Madagascar (Babineau & 2016). Our method is markedly different: by identifying (practically) Bruneau, 2017), respectively (Supporting Information Figure S22), all species that make up the functional group known as stem succu- and all 27 species of the Parkinsonia-Delonix clade are restricted to lents, mapping these species-by-species and quantifying global rich- the succulent biome, again with two separate succulent biome trans- ness patterns, we use a bottom-up approach to biome mapping. This Atlantic disjunctions (Table 1, Figure 3 and Supporting Information approach has two advantages. First, although here performed on a 1106  |     RINGELBERG Et aL. (a) Old World New World elevation precipitation temperature PC1 (37.6%) ( b ) Elevation (c) Length of dry season (d) Mean annual precipitation ( e ) Mean annual temperature .26 .4 .12 .14 .13 .2 .06 .07 0 0 0 0 0 1,000 2,000 3,000 4,000 0 2 4 6 8 10 12 0 1,000 2,000 3,000 0 5 10 15 20 25 30 Metres Months Millimetres Degrees Celsius F I G U R E 2   Climate of the succulent biome. Subfigures depict climatic values of all 2,850 0.25° lat./long. grid cells with at least two unique stem succulent genera (see Methods). (a) Principal components analysis (PCA) showing the high climatic overlap between the succulent biome in the New and Old Worlds. Arrows indicate direction, but not magnitude, of the same four climatic variables depicted in b–e. See Supporting Information Figure S9 for a PCA depicting all climatic variables used in this study. (b) Elevational distribution of the succulent biome. (c) Length of the dry season (consecutive months with rainfall < 100 mm/month). Over half of all cells have a dry season of at least 10 months. (d) Distribution of mean annual rainfall, which closely matches that of South American seasonally dry tropical forests (SDTFs) (Dexter et al., 2018: their fig. 2), except that stem succulents also occur in drier areas. (e) Mean annual temperature distribution, showing that the succulent biome mainly occurs in frost free areas (also see Supporting Information Figure S10) [Colour figure can be viewed at wileyonlinelibrary.com] global scale, if species occurrence data are densely sampled, there insightful for defining and mapping biomes (Moncrieff et al., 2016). is no reason this approach cannot work on more local scales to pro- Possibilities might include high-resolution global maps of crassu- duce a higher resolution map that more accurately reveals the local lacean acid metabolism photosynthesis, annual versus perennial interdigitation of biomes in parts of Mexico and south-east Africa plant life history, plant growth forms including lianas and geoxyles, where the succulent biome is currently over-projected (see below). C4 grasses, leaves with drip tips, sclerophyly, deciduousness and Second, this method could be used for any functional trait. This is spinescence; there are many options. A complication with this ap- not to imply that every functional trait is a proxy for a unique biome, proach is spatial biases in available species occurrence data (Meyer but simply that global maps of key functional traits would be very et al., 2016), but as we show, this problem can be circumvented by Proportion of cells PC2 (22.7%) length of dry season RINGELBERG Et aL.      |  1107 Bourreria clade Bursera - Commiphora Caesalpinia group Gambelia clade Leucaena - Dichrostachys clade Parkinsonia - Delonix clade Pictetia clade Prosopis clade Sideroxylon Thamnosma F I G U R E 3   Phylogenies, geographical distributions and succulent biome occupancy for 10 non-succulent angiosperm clades. On the maps and phylogenies, red corresponds to species/occurrences in the succulent biome, and blue in other biomes. Black squares on the tips of the phylogenies indicate New World species, absence of a square indicates Old World. Branches of the phylogenies are coloured based on the biome optimizations (see Methods). The insets on the maps of the Caesalpinia group and the Leucaena–Dichrostachys clade show Hawai’i and several islands in the South Pacific Ocean. Detailed phylogenies with terminal names and larger sized distribution maps are in Supporting Information Figures S12–S31 [Colour figure can be viewed at wileyonlinelibrary.com] species distribution modelling, and occurrence data are rapidly ex- model. First, there is a trade-off between specificity and general- panding. A logical next step would be to map not just occurrences of ity, most notably in the spatial scale of the modelling. While our functional traits, but also their abundance, based on global plot data selected scale (0.25° long./lat. grid cells) is not that coarse, there (Oliveira-Filho et al., 2013). is still scope for heterogeneity within cells deemed uniform by While we are confident that the output of our models the model. Second, stem succulents can occur on azonal land- (Figure 1) accurately depicts the global distribution of stem scape features such as inselbergs, termitaria (Moe et al., 2009), succulents, an important caveat is that stem succulents do not and rocky outcrops that are less fire-prone than surrounding sa- necessarily occur at every location predicted by our model, nor vannas (Cousins & Witkowski, 2012; Pérez-García et al., 2012; do we claim they are absent from all areas not predicted by our Thomas, 1991). Third, while much digitized species occurrence 1108  |     RINGELBERG Et aL. data are available, some regions, such as India and north-east Africa, are significantly under-represented (Meyer et al., 2016). Finally, a handful of stem succulents (c. 15 species) are known from more mesic areas outside of the succulent biome, perhaps reflecting the difficulties associated with how to define suc- culents in general, and stem succulents in particular (Eggli & Nyffeler, 2009; Ogburn & Edwards, 2010). These factors probably explain why, although highly consistent on a global scale, locally there are minor differences among the four stem succulent maps (Supporting Information Figures S1–S4), and also likely play a role in the apparent over-projection in Mexico (see below). Despite these minor shortcomings, the map of the global distribution of stem succulents is as accurate and detailed as is possible based on available data, and provides a useful quantitative global map of the succulent biome for macroevolutionary studies. 4.2 | The trans-continental succulent biome In the New World, the distribution of stem succulents includes all areas of SDTF (DRYFLOR, 2016; Pennington et al., 2009; Pennington, Prado, & Pendry, 2000) in the Brazilian Caatinga, the Bolivian Chiquitania, Piedmont in Argentina, dry inter-Andean val- leys, semi-arid coastal zones of Colombia and Venezuela, and dry forests in the Caribbean, Central America and Mexico (Figure 1). However, stem succulents have a broader range than SDTFs, ex- tending into drier areas in northern Mexico and the arid coasts of Peru and Chile. When the distribution of stem succulents is modelled separately for occurrences with mean annual precipita- tion above and below 1,000 mm, the > 1,000 mm ‘wetter’ model (Supporting Information Figure S7) more closely resembles the typical distribution of SDTFs (DRYFLOR, 2016; Pennington et al., 2000). In contrast, most New World stem succulent occurrences fall into the ‘drier’ < 1,000 mm category, and this model maps the more arid areas of the New World (Supporting Information Figure S7), including the Caatinga, even though this is usually considered typi- cal SDTF (de Queiroz, Cardoso, Fernandes, & Moro, 2017). While SDTFs span a range of precipitation (Dexter et al., 2018; Pennington et al., 2009), our results suggest that typical SDTFs occur mainly across the wetter portion of the succulent biome, which overall also encompasses more arid scrub and thickets, but always with a more or less open cover of deciduous shrubs, small trees, and stem suc- culents. In Mexico the taxonomic affinities of the arid northern de- serts to wetter southern SDTFs (Pérez-García et al., 2012) fit with this wider definition of the succulent biome, as does the modified descriptor SDTFW (‘SDTF and Woodland’) suggested by de Queiroz et al. (2017). Nevertheless, our model appears to over-project suc- culents to areas where they are less abundant in mid-elevation pine–oak forests in Mexico. This is a function of the wide ecologi- cal, geographical and altitudinal amplitude of stem succulents, and especially Cactaceae, in the New World, as well as the small-scale interdigitation of the succulent biome with these climatically similar pine–oak forests. TA B L E 1   Non-succulent angiosperm study clades and results of the phylogenetic biome conservatism tests Species in clade (fraction in Species in phylogeny (fraction in Biome shifts significantly Trans-Atlantic disjunctions Clade succulent biome) succulent biome) lower than random? Phylogenetic signal (fraction in succulent biome) Bourreria clade (Ehretiaceae) 56 (.88) 26 (.88) No 1 1 (1.00) Bursera–Commiphora 235 (.93) 181 (.96) No .99 2 (1.00) (Burseraceae) Caesalpinia group (Leguminosae) 199 (.68) 175 (.73) Yes .83 7 (.71) Gambelia clade (Plantaginaceae) 11 (.91) 8 (.88) – – 1 (1.00) Leucaena–Dichrostachys clade 100 (.79) 78 (.78) Yes .76 2 (1.00) (Leguminosae) Parkinsonia–Delonix clade 27 (1.00) 21 (1.00) – – 2 (1.00) (Leguminosae) Pictetia clade (Leguminosae) 52 (.87) 45 (.87) Yes .48 1 (1.00) Prosopis clade (Leguminosae) 61 (.70) 35 (.69) No 1 1 (.00) Sideroxylon (Sapotaceae) 87 (.59) 41 (.59) Yes .99 1 (1.00) Thamnosma (Rutaceae) 11 (1.00) 14 (1.00) – – 1 (1.00) Note: The second column only lists the number of species per clade for which occurrence data are available; some clades contain more species in total than are listed. The fractions of succulent biome occupancy reflect species exclusively occurring in the succulent biome, not species occurring in the succulent biome in addition to a different biome. RINGELBERG Et aL.      |  1109 Another discrepancy between our stem succulent model and 1981; Holtum et al., 2016). It is possible that longer-term climatic os- traditional SDTF maps is the Chaco. The Chaco has characteris- cillations exclude stem succulents from Australia, as the CHELSA cli- tics of SDTFs (seasonally dry, fire-free, grass-poor), is clearly not matic data used in this study only reflect the last c. 35 years (Karger a savanna, as indicated in the latest global map of grassy biomes et al., 2017). Alternatively, the deep history of fire in Australia (Crisp, (Lehmann et al., 2019), appears to comprise a mosaic of elements Burrows, Cook, Thornhill, & Bowman, 2011) may have rendered the from several biomes (Segovia et al., 2019), and has been considered continent unsuitable for stem succulents throughout the Cenozoic, a distinct biome based on differences in soils, occurrence of frost, in line with the idea that most of northern and central Australia are and floristic composition (DRYFLOR, 2017; Pennington et al., 2000; assigned as grassy, fire-prone savanna (Lehmann et al., 2019). The Silva de Miranda et al., 2018). The affinities of the Chaco remain de- absence of stem succulents from most of Asia is in line with the batable (Kuemmerle et al., 2017; Segovia et al., 2019). Twenty-three view that ‘dry forests’ in Asia are better regarded as savannas be- species of stem succulents occur in the Chaco, some of them en- cause they also have a flammable C4 grass layer (Dexter et al., 2015; demic there, and the Chaco is clearly included in our succulent biome Lehmann et al., 2019; Ratnam et al., 2011). model (Figure 1). While this does not mean the Chaco is a typical SDTF, based on occurrence of stem succulents, the Chaco and other Neotropical SDTFs cannot be distinguished. 4.3 | Succulent biome phylogenetic conservatism In the Old World, the major centres of stem succulent diversity are in the northern succulent Karoo, the Namib desert, and the Acacia– We demonstrate high levels of phylogenetic succulent biome conserv- Commiphora woodlands of the Somali-Masai region in Africa, western atism in a cohort of non-succulent plant clades that comprises promi- Madagascar (spiny forests in the south and deciduous forests in the nent and in some cases dominant elements of this biome (Figure 3 and north), and coastal Arabia (Figure 1). The distribution in continental Table 1). This suggests that the succulent biome forms a tightly deline- Africa closely matches the Arid Corridor (De Winter, 1971; Poynton, ated evolutionary arena for drought-adapted plants. These patterns of 1995) or Arid Flora (Linder, 2014), well known for its many disjunctly succulent biome conservatism are especially striking given that many distributed taxa (De Winter, 1971; Jürgens, 1997; Linder, 2014; of the species are separated by large geographical (often trans-conti- Poynton, 1995; Thiv et al., 2011). The most important differences be- nental) disjunctions from their nearest relatives yet still occur within tween our model and the map of Schrire et al. (2005) are in south-east the succulent biome (Figure 3), suggesting that the high levels of biome Africa. Coastal subtropical thicket vegetation of the Eastern Cape of conservatism are not simply the result of limited dispersal abilities. This South Africa has a notable presence of stem succulents (Figure 1), and confirms earlier results (Gagnon et al., 2019; Lavin et al., 2004; Thiv is included in the succulent biome (Cowling, Procheş, & Vlok, 2005; et al., 2011) and suggests that for lineages spanning the trans-conti- Linder, 2014). Our stem succulent model also extends into inland nentally distributed succulent biome it has been ‘easier to move than to areas of south-east Africa generally not considered part of the coastal evolve’ as highlighted by Donoghue (2008, 2019). This shows that the thicket vegetation [although they feature Arid Corridor disjunctions idea that local recruitment and independent origins of biomes within (Jürgens, 1997; Poynton, 1995; Thiv et al., 2011)]. This region contains continents is a universal rule governing biomes in general (Pennington many stem succulent Apocynaceae, Asphodelaceae, Euphorbiaceae et al., 2018), clearly is not applicable to the succulent biome with its and Malvaceae, but is also characterized by the presence of C4 grasses high levels of trans-continental phylogenetic conservatism. and regular fires and generally classified as savanna (White, 1983). We Succulent biome phylogenetic conservatism, far from being re- suggest that stem succulents in this area occupy locally fire-free azonal stricted to legumes where it was first documented (Donoghue, 2019; sites (rocky gullies and outcrops and termitaria; Cousins & Witkowski, Gagnon et al., 2019; Lavin et al., 2004), is prevalent across a range of 2012; Moe et al., 2009), with a mosaic of pockets of succulent biome plant families, and certain animal lineages (Supporting Information across a region of predominantly savanna. The other discrepancy with Table S5). These findings are comparable to some of the most strik- Schrire et al. (2005)’s map is that our model does not predict stem ing examples of global phylogenetic biome conservatism (Crisp et al., succulents in coastal Iran, Pakistan and northwest India due to lack of 2009; Donoghue & Smith, 2004; Wiens & Donoghue, 2004), sug- occurrence data (Supporting Information Figures S1–S4), even though gesting that phylogenetic conservatism may be most apparent at stem succulents are known from those regions (Eggli, 2004). this broad biome level (Segovia et al., 2019). Furthermore, our results The near absence of stem succulents and any occurrences of benefit from the additional rigour that comes from a quantitative the succulent biome from Australia is striking and well known, but biome model and use of detailed species occurrence data to objec- poorly understood (Ellenberg, 1981; Holtum et al., 2016). Several in- tively and quantitatively assign species to biomes. troduced stem succulent cactus species have invaded large parts of Australia (Ellenberg, 1981; Holtum et al., 2016), indicating that stem succulents can thrive there, and our Old and New World models 4.4 | Biomes as evolutionary arenas both predict large parts of Australia as climatically suitable for stem succulents (Supporting Information Figure S8). Our models thus We show that the succulent biome shows striking convergence of disagree with Ellenberg’s view that the climatic envelope of stem functional traits [in stem succulence, early burst pre-rain leaf flush- succulents in Africa and America is absent in Australia (Ellenberg, ing (Donoghue, 2019; Gagnon et al., 2019; Oliveira-Filho et al., 2013) 1110  |     RINGELBERG Et aL. and spinescence (see below)], as well as high levels of phylogenetic evolutionary convergence and phylogenetic biome conservatism succulent biome conservatism in 10 non-succulent clades (Table 1). (Corlett & Primack, 2006). In other words, the confluence of evolutionary convergence and phylogenetic biome conservatism strongly suggests the succulent biome forms a well-defined evolutionary arena (Gagnon et al., 2019; 5  | CONCLUSIONS Lavin et al., 2004; Oliveira-Filho et al., 2013; Schrire et al., 2005; Thiv et al., 2011). However, current global biome maps do not rec- We highlight the utility of the convergence of key plant functional ognize the succulent biome (e.g. Friedl et al., 2010; Higgins et al., traits for defining and mapping the little-known succulent biome. 2016; Olson et al., 2001; but see Pennington et al., 2018), and it has We show that this biome forms a tightly delineated geographi- received limited attention in recent years. This neglect of the succu- cally disjunct, trans-continental evolutionary arena for drought- lent biome has far reaching implications for understanding tropical adapted plant lineages, providing some of the strongest evidence diversity. to date that ecology has played a key role in dictating the geo- For example, recognizing the succulent biome is important graphical turnover of clades across the lowland tropics (Segovia for understanding the origins of the savanna biome, especially in et al., 2019). We suggest that the dichotomy between succulent- Africa. Charles-Dominique et al. (2016) argued that contempora- rich, grass-poor, fire-free and succulent-lacking, grass-dominated, neous evolution of spinescence and diversification of bovids in the fire-prone vegetation reflects a fundamental functional distinc- mid-Miocene underpinned the rise of African savannas. However, tion underpinning recognition of two distinct non-forest tropical regions with high diversity of spiny plant species (their fig. 2b) and lowland biomes, both of which are ‘open’ vegetation formations bovids (their fig. 2c and e) substantially overlap with the succulent with seasonally dry climates, and both of which are ancient and biome (our Figure 1), rather than comprising savanna as suggested merit attention (Bond, 2005; Cowling et al., 2005; White, 1983). by Charles-Dominique et al. (2016), a view reinforced by the rela- This is supported by the largely non-overlapping distributions of tive paucity of perennial grasses in these areas (Lehmann et al., the succulent and savanna biomes, as depicted in Figure 1 and the 2019). Spinescent plants are prevalent across many plant lineages most recent global map of grassy biomes (Lehmann et al., 2019). throughout the succulent biome (e.g. the ‘Spiny Forest’ of south- Such a division is also very much in line with broad division of west Madagascar; see also Cowling et al., 2005; de Queiroz et al., the lowland tropics into three main biomes—rain forest, savanna 2017; Pérez-García et al., 2012), providing another example of evo- and succulent (Dexter et al., 2018; Lavin et al., 2004; Pennington lutionary convergence of an important plant functional trait asso- et al., 2018; Schrire et al., 2005)—rather than the two broad cat- ciated with this biome. It seems likely that initial diversification of egories prevalent in many macroevolutionary and macroecologi- spiny plant lineages somewhat earlier than the arrival of bovids in cal studies. Finally, we show that the confluence of evolutionary Africa, and well before the rise to dominance of C4 grasses (Charles- convergence and phylogenetic biome conservatism can provide a Dominique et al., 2016: fig 4), could reflect, not the emergence of strong indication that geographically disjunct areas of the same savannas, but rather the earlier evolution of spinescence in plant biome form a single coherent evolutionary arena, suggesting that clades occupying the succulent biome, which is thought to pre-date in-depth cross-continental comparisons of convergence and con- the savanna biome (Gagnon et al., 2019). servatism in other biomes would be worthwhile. It remains to be seen to what extent other global biomes con- stitute evolutionary arenas characterized by similar combinations ACKNOWLEDG MENTS of convergence and conservatism to those demonstrated here for We thank Peter Linder for insightful comments on an earlier the succulent biome. While the flora of the global savanna biome version of this manuscript, Achilleas Psomas and Rafael Wüest shows several examples of evolutionary convergence in fire-re- for sharing custom R scripts, William Hawthorne, Mats Thulin lated traits such as geoxyles, there is little evidence of global phy- and Alan Forrest for additional species occurrence records of logenetic savanna biome conservatism; instead, different parts African stem succulents and particular endemic species, the pho- of this biome were apparently assembled via repeated local or tographers of the pictures in Figure 1, Emma Ringelberg for as- in situ recruitment and adaptation of lineages via biome shifts sisting with assembling the figures, the Swiss National Science from geographically adjacent biomes within the same continent Foundation (Grants 31003A_156140 and 31003A_182453/1 to (Maurin et al., 2014; Moncrieff et al., 2014; Simon et al., 2009). CEH) for funding, and the US National Science Foundation (Grants The flora of the Mediterranean biome also displays a suite of con- DEB-0919179 and DEB-1403150 to AW), which funded the acqui- vergent plant functional traits, and there are at least forty genera sition of sequence data for Bursera and Commiphora. occurring in two or more of its regions suggesting some degree of trans-continental Mediterranean biome conservatism (although DATA AVAIL ABILIT Y S TATEMENT no genus is known from all five Mediterranean regions; Ackerly A raster file of the succulent biome map, occurrence data of stem & Onstein, 2018). Whether these and other biomes constitute succulents and the study clades, and an R script for cleaning up oc- evolutionary arenas remains to be determined via more thorough, currence data can be found on Dryad (doi:10.5061/dryad.08kpr cross-continental comparisons and assessments of the degrees of r4zs). RINGELBERG Et aL.      |  1111 ORCID at the Cretaceous-Palaeogene boundary. Nature Communications, 2, Jens J. Ringelberg https://orcid.org/0000-0003-0567-5210 193. https://doi.org/10.1038/ncomm s1191 de Queiroz, L. P., Cardoso, D., Fernandes, M. F., & Moro, M. F. (2017). Niklaus E. Zimmermann https://orcid. Diversity and evolution of flowering plants of the Caatinga Domain. org/0000-0003-3099-9604 In J. M. C. Da Silva, I. R. Lear, & M. Tabarelli (Eds.), Caatinga: The Andrea Weeks https://orcid.org/0000-0002-0000-4541 largest tropical dry forest region in South America (pp. 23–63). Cham, Matt Lavin https://orcid.org/0000-0003-4205-1802 Switzerland: Springer International. De Winter, B. (1971). Floristic relationships between the northern Colin E. Hughes https://orcid.org/0000-0002-9701-0699 and southern arid areas in Africa. Mitteilungen Der Botanischen Staatssammlung München, 10, 424–437. R E FE R E N C E S De-Nova, J. A., Medina, R., Montero, J. C., Weeks, A., Rosell, J. A., Ackerly, D. D., & Onstein, R. E. (2018). Case study 2: Assembly of med- Olson, M. E., … Magallón, S. (2012). Insights into the historical con- iterranean-type floras: Convergence, exaptation and evolutionary struction of species-rich Mesoamerican seasonally dry tropical for- predisposition. In K. J. Esler, A. L. Jacobsen, & R. B. Pratt (Eds.), The ests: The diversification of Bursera (Burseraceae, Sapindales). New biology of mediterranean-type ecosystems (pp. 27–34). Oxford, UK: Phytologist, 193(1), 276–287. https://doi.org/10.1111/j.1469-8137. Oxford University Press. 2011.03909.x Albers, F., & Meve, U. (Eds.). (2004). Illustrated handbook of succulent Dexter, K. G., Pennington, R. T., Oliveira-Filho, A. T., Bueno, M. L., plants: Asclepiadaceae. Berlin, Germany: Springer-Verlag. Silva de Miranda, P. L., & Neves, D. M. (2018). Inserting tropical Amatulli, G., Domisch, S., Tuanmu, M. N., Parmentier, B., Ranipeta, A., dry forests into the discussion on biome transitions in the tropics. Malczyk, J., & Jetz, W. (2018). A suite of global, cross-scale topo- Frontiers in Ecology and Evolution, 6, 104. https://doi.org/10.3389/ graphic variables for environmental and biodiversity modeling. fevo.2018.00104 Scientific Data, 5, 1–15. https://doi.org/10.1038/sdata.2018.40 Dexter, K. G., Smart, B., Baldauf, C., Baker, T. R., Balinga, M. P. B., Anderson, E. F. (2001). The cactus family. Portland, OR: Timber Press. Brienen, R., … Pennington, R. T. (2015). Floristics and biogeogra- Antonelli, A., Zizka, A., Carvalho, F. A., Scharn, R., Bacon, C. D., Silvestro, phy of vegetation in seasonally dry tropical regions. International D., & Condamine, F. L. (2018). Amazonia is the primary source of Forestry Review, 17, 10–32. https://doi.org/10.1505/146554 8158 Neotropical biodiversity. Proceedings of the National Academy of 15834859 Sciences of the United States of America, 115(23), 6034–6039. https:// Donoghue, M. J. (2008). A phylogenetic perspective on the distribution doi.org/10.1073/pnas.17138 19115 of plant diversity. Proceedings of the National Academy of Sciences USA, Ávila-Lovera, E., & Ezcurra, E. (2016). Stem-succulent trees from the 105(1), 11549–11555. https://doi.org/10.1073/pnas.08019 62105 Old and New World tropics. In G. Goldstein & L. S. Santiago (Eds.), Donoghue, M. J. (2019). Adaptation meets dispersal: Legumes in the Tropical tree physiology: Adaptations and responses in a changing envi- land of succulents. New Phytologist, 222(4), 1667–1669. https://doi. ronment (pp. 45–65). Cham, Switzerland: Springer International. org/10.1111/nph.15834 Babineau, M., & Bruneau, A. (2017). Phylogenetic and biogeograph- Donoghue, M. J., & Edwards, E. J. (2014). Biome shifts and niche evolution ical history of the Afro-Madagascan genera Delonix, Colvillea and in plants. Annual Review of Ecology, Evolution, and Systematics, 45(1), Lemuropisum (Fabaceae: Caesalpinioideae). Botanical Journal of the 547–572. https://doi.org/10.1146/annur ev-ecols ys-120213 -091905 Linnean Society, 184, 59–78. https://doi.org/10.1093/botli nnean / Donoghue, M. J., & Smith, S. A. (2004). Patterns in the assembly of box009 temperate forests around the Northern Hemisphere. Philosophical Bond, W. J. (2005). Large parts of the world are brown or black: A different Transactions of the Royal Society B: Biological Sciences, 359, 1633– view on the ‘Green World’ hypothesis. Journal of Vegetation Science, 1644. https://doi.org/10.1098/rstb.2004.1538 16(3), 261–266. https://doi.org/10.1658/1100-9233(2005)016[0261: DRYFLOR: Banda-R, K., Delgado-Salinas, A., Dexter, K. G., Linares- LPOTWA ]2.0.CO;2 Palomino, R., Oliveira-Filho, A., … Pennington, R. T. (2016). Plant Chala, D., Brochmann, C., Psomas, A., Ehrich, D., Gizaw, A., Masao, C. A., diversity patterns in neotropical dry forests and their conservation … Zimmermann, N. E. (2016). Good-bye to tropical alpine plant giants implications. Science, 353(6306), 1383–1387. under warmer climates? Loss of range and genetic diversity in Lobelia DRYFLOR: Pennington, R. T., Banda-R, K., Delgado-Salinas, A., Dexter, K. rhynchopetalum. Ecology and Evolution, 6(24), 8931–8941. G., Galetti, L., … Särkinen, T. E. (2017). Response to “Forest conserva- Charles-Dominique, T., Davies, T. J., Hempson, G. P., Bezeng, B. S., Daru, tion: Remember Gran Chaco.” Science, 355(6324), 465–466. B. H., Kabongo, R. M., … Bond, W. J. (2016). Spiny plants, mammal Edwards, E. J., & Donoghue, M. J. (2013). Is it easy to move and easy browsers, and the origin of African savannas. Proceedings of the to evolve? Evolutionary accessibility and adaptation. Journal of National Academy of Sciences of the United States of America, 113(38), Experimental Botany, 64(13), 4047–4052. E5572–E5579. https://doi.org/10.1073/pnas.16074 93113 Eggli, U. (Ed.). (2001). Illustrated handbook of succulent plants: Corlett, R. T., & Primack, R. B. (2006). Tropical rainforests and the need Monocotyledons. Berlin, Germany: Springer-Verlag. for cross-continental comparisons. Trends in Ecology and Evolution, Eggli, U. (Ed.). (2004). Illustrated handbook of succulent plants: Dicotyledons. 21(2), 104–110. https://doi.org/10.1016/j.tree.2005.12.002 Berlin, Germany: Springer-Verlag. Cousins, S. R., & Witkowski, E. T. F. (2012). African aloe ecology: A review. Eggli, U., & Nyffeler, R. (2009). Living under temporarily arid conditions— Journal of Arid Environments, 85, 1–17. https://doi.org/10.1016/j.jarid Succulence as an adaptive strategy. Bradleya, 27, 13–36. https://doi. env.2012.03.022 org/10.25223 /brad.n27.2009.a10 Cowling, R. M., Procheş, Ş., Vlok, J., & van Staden, J. (2005). On the Eiserhardt, W. L., Couvreur, T. L. P., & Baker, W. J. (2017). Plant phy- origin of southern African subtropical thicket vegetation. South logeny as a window on the evolution of hyperdiversity in the trop- African Journal of Botany, 71(1), 1–23. https://doi.org/10.1016/S0254 ical rainforest biome. New Phytologist, 214, 1408–1422. https://doi. -6299(15)30144- 7 org/10.1111/nph.14516 Crisp, M. D., Arroyo, M. T. K., Cook, L. G., Gandolfo, M. A., Jordan, G. J., Ellenberg, H. (1981). Ursachen des Vorkommens und Fehlens von McGlone, M. S., … Linder, H. P. (2009). Phylogenetic biome conserva- Sukkulenten in den Trockengebieten der Erde. Flora, 171, 114–169. tism on a global scale. Nature, 458(7239), 754–756. Fick, S. E., & Hijmans, R. J. (2017). WorldClim 2: New 1-km spatial reso- Crisp, M. D., Burrows, G. E., Cook, L. G., Thornhill, A. H., & Bowman, D. lution climate surfaces for global land areas. International Journal of M. J. S. (2011). Flammable biomes dominated by eucalypts originated Climatology, 37, 4302–4315. https://doi.org/10.1002/joc.5086 1112  |     RINGELBERG Et aL. Friedl, M. A., Sulla-Menashe, D., Tan, B., Schneider, A., Ramankutty, N., Maddison, W. P., & Slatkin, M. (1991). Null models for the number of Sibley, A., & Huang, X. (2010). MODIS Collection 5 global land cover: evolutionary steps in a character on a phylogenetic tree. Evolution, Algorithm refinements and characterization of new datasets. Remote 45(5), 1184–1197. https://doi.org/10.1111/j.1558-5646.1991. Sensing of Environment, 114(1), 168–182. https://doi.org/10.1016/j. tb043 85.x rse.2009.08.016 Marshall, C. A. M., Wieringa, J. J., & Hawthorne, W. D. (2016). Bioquality Gagnon, E., Ringelberg, J. J., Bruneau, A., Lewis, G. P., & Hughes, C. E. hotspots in the tropical African flora. Current Biology, 26(23), 3214– (2019). Global succulent biome phylogenetic conservatism across the 3219. https://doi.org/10.1016/j.cub.2016.09.045 pantropical Caesalpinia group (Leguminosae). New Phytologist, 222(4), Maurin, O., Davies, T. J., Burrows, J. E., Daru, B. H., Yessoufou, K., 1994–2008. https://doi.org/10.1111/nph.15633 Muasya, A. M., … Bond, W. J. (2014). Savanna fire and the origins of Gostel, M. R., Phillipson, P. B., & Weeks, A. (2016). Phylogenetic recon- the “underground forest” of Africa. New Phytologist, 204, 201–214. struction of the Myrrh genus, Commiphora (Burseraceae), reveals Merow, C., Smith, M. J., Edwards, T. C., Guisan, A., McMahon, S. M., multiple radiations in Madagascar and clarifies infrageneric relation- Normand, S., … Elith, J. (2014). What do we gain from simplicity ver- ships. Systematic Botany, 41(1), 67–81. sus complexity in species distribution models? Ecography, 37(12), Griffiths, H., & Males, J. (2017). Succulent plants. Current Biology, 27(17), 1267–1281. https://doi.org/10.1111/ecog.00845 R890–R896. https://doi.org/10.1016/j.cub.2017.03.021 Meyer, C., Weigelt, P., & Kreft, H. (2016). Multidimensional biases, gaps Harmon, L. J., Weir, J. T., Brock, C. D., Glor, R. E., & Challenger, W. (2008). and uncertainties in global plant occurrence information. Ecology GEIGER: Investigating evolutionary radiations. Bioinformatics, 24(1), Letters, 19, 992–1006. https://doi.org/10.1111/ele.12624 129–131. https://doi.org/10.1093/bioin format ics/btm538 Moe, S. R., Mobæk, R., & Narmo, A. K. (2009). Mound building termites Higgins, S. I., Buitenwerf, R., & Moncrieff, G. R. (2016). Defining func- contribute to savanna vegetation heterogeneity. Plant Ecology, tional biomes and monitoring their change globally. Global Change 202(1), 31–40. https://doi.org/10.1007/s1125 8-009-9575-6 Biology, 22(11), 3583–3593. https://doi.org/10.1111/gcb.13367 Moncrieff, G. R., Bond, W. J., & Higgins, S. I. (2016). Revising the biome Hijmans, R. J. (2018). raster: Geographic data analysis and modeling. concept for understanding and predicting global change impacts. R package version 2.8-4. https://CRAN.R-proje ct.org/packa Journal of Biogeography, 43(5), 863–873. https://doi.org/10.1111/ ge=raster jbi.12701 Holtum, J. A. M., Hancock, L. P., Edwards, E. J., Crisp, M. D., Crayn, D. Moncrieff, G. R., Hickler, T., & Higgins, S. I. (2015). Intercontinental di- M., Sage, R., & Winter, K. (2016). Australia lacks stem succulents vergence in the climate envelope of major plant biomes. Global but is it depauperate in plants with crassulacean acid metabolism Ecology and Biogeography, 24(3), 324–334. https://doi.org/10.1111/ (CAM)? Current Opinion in Plant Biology, 31, 109–117. https://doi. geb.12257 org/10.1016/j.pbi.2016.03.018 Moncrieff, G. R., Lehmann, C. E. R., Schnitzler, J., Gambiza, J., Hiernaux, Hughes, C. E., Pennington, R. T., & Antonelli, A. (2013). Neotropical plant P., Ryan, C. M., … Higgins, S. I. (2014). Contrasting architecture of evolution: Assembling the big picture. Botanical Journal of the Linnean key African and Australian savanna tree taxa drives intercontinental Society, 171(1), 1–18. https://doi.org/10.1111/boj.12006 structural divergence. Global Ecology and Biogeography, 23(11), 1235– Jürgens, N. (1997). Floristic biodiversity and history of African arid re- 1244. https://doi.org/10.1111/geb.12205 gions. Biodiversity and Conservation, 6, 495–514. Mucina, L. (2019). Biome: Evolution of a crucial ecological and bio- Karger, D. N., Conrad, O., Böhner, J., Kawohl, T., Kreft, H., Soria-Auza, geographical concept. New Phytologist, 222, 97–114. https://doi. R. W., … Kessler, M. (2017). Climatologies at high resolution for org/10.1111/nph.15609 the earth’s land surface areas. Scientific Data, 4, 1–20. https://doi. Nürk, N. M., Linder, H. P., Onstein, R. E., Larcombe, M. J., Hughes, C. org/10.1038/sdata.2017.122 E., Fernández, L. P., … Pirie, M. D. (2019). Diversification in evolu- Kuemmerle, T., Altrichter, M., Baldi, G., Cabido, M., Camino, M., Cuellar, tionary arenas—Assessment and synthesis. BioRxiv. https://doi. E., … Zak, M. (2017). Forest conservation: Remember Gran Chaco. org/10.1101/636803 Science, 355(6324), 465. https://doi.org/10.1038/sdata.2017.122 Ogburn, R. M., & Edwards, E. J. (2010). The ecological water-use Lavin, M., Schrire, B. P., Lewis, G., Pennington, R. T., Delgado–Salinas, A., strategies of succulent plants. In J.-C. Kader & M. Delseny (Eds.), Thulin, M., … Wojciechowski, M. F. (2004). Metacommunity process Advances in botanical research (Vol. 55, pp. 179–225). Amsterdam, the rather than continental tectonic history better explains geograph- Netherlands: Elsevier. ically structured phylogenies in legumes. Philosophical Transactions Oliveira-Filho, A. T., Cardoso, D., Schrire, B. D., Lewis, G. P., Pennington, of the Royal Society B: Biological Sciences, 359(1450), 1509–1522. R. T., Brummer, T. J., … Lavin, M. (2013). Stability structures tropical https://doi.org/10.1098/rstb.2004.1536 woody plant diversity more than seasonality: Insights into the ecol- Lehmann, C. E. R., Anderson, T. M., Sankaran, M., Higgins, S. I., ogy of high legume-succulent-plant biodiversity. South African Journal Archibald, S., Hoffmann, W. A., … Bond, W. J. (2014). Savanna veg- of Botany, 89, 42–57. https://doi.org/10.1016/j.sajb.2013.06.010 etation-fire-climate relationships differ among continents. Science, Oliveras, I., & Malhi, Y. (2016). Many shades of green: The dynamic tropi- 343(6170), 548–552. cal forest–savannah transition zones. Philosophical Transactions of the Lehmann, C. E. R., Griffith, D. M., Simpson, K. J., Anderson, T. M., Royal Society B: Biological Sciences, 371, 20150308. Archibald, S., Beerling, D. J., … Osborne, C. P. (2019). Functional Olson, D. M., Dinerstein, E., Wikramanayake, E. D., Burgess, N. D., Powell, diversification enabled grassy biomes to fill global climate space. G. V. N., Underwood, E. C., … Kassem, K. R. (2001). Terrestrial ecore- BioRxiv. https://doi.org/10.1101/583625 gions of the world: A new map of life on Earth. BioScience, 51(11), Lehmann, C. E. R., & Parr, C. L. (2016). Tropical grassy biomes: Linking 933–938. https://doi.org/10.1641/0006-3568(2001)051[0933:- ecology, human use and conservation. Philosophical Transactions of TEOTWA ]2.0.CO;2 the Royal Society B: Biological Sciences, 371, 20160329. Pagel, M. (1999). Inferring the historical patterns of biological evolution. Linder, H. P. (2014). The evolution of African plant diversity. Nature, 401(6756), 877–884. Frontiers in Ecology and Evolution, 2, 38. https://doi.org/10.3389/ Paradis, E., & Schliep, K. (2019). ape 5.0: An environment for modern fevo.2014.00038 phylogenetics and evolutionary analyses in R. Bioinformatics, 35(3), Losos, J. B. (2008). Phylogenetic niche conservatism, phylogenetic sig- 526–528 nal and the relationship between phylogenetic relatedness and eco- Pennington, R. T., Lavin, M., & Oliveira-Filho, A. (2009). Woody plant logical similarity among species. Ecology Letters, 11(10), 995–1003. diversity, evolution, and ecology in the tropics: Perspectives from https://doi.org/10.1111/j.1461-0248.2008.01229.x seasonally dry tropical forests. Annual Review of Ecology Evolution RINGELBERG Et aL.      |  1113 and Systematics, 40, 437–457. https://doi.org/10.1146/annur ev.ecols Staver, A. C., Archibald, S., & Levin, S. (2011). The global extent and de- ys.110308.120327 terminants of savanna and forest as alternative biome states. Science, Pennington, R. T., Lehmann, C. E. R., & Rowland, L. M. (2018). Tropical 334(6053), 230–232. savannas and dry forests. Current Biology, 28(9), R541–R545. https:// Thiv, M., van der Niet, T., Rutschmann, F., Thulin, M., Brune, T., & doi.org/10.1016/j.cub.2018.03.014 Linder, H. P. (2011). Old-New World and trans-African disjunctions Pennington, R. T., Prado, D. E., & Pendry, C. A. (2000). of Thamnosma (Rutaceae): Intercontinental long-distance dispersal Neotropical seasonally dry forests and Quaternary vegeta- and local differentiation in the succulent biome. American Journal of tion changes. Journal of Biogeography, 27, 261–273. https://doi. Botany, 98(1), 76–87. https://doi.org/10.3732/ajb.1000339 org/10.1046/j.1365-2699.2000.00397.x Thomas, P. A. (1991). Response of succulents to fire: A review. International Pérez-García, E. A., Meave, J. A., & Cevallos-Ferriz, S. R. S. (2012). Journal of Wildland Fire, 1(1), 11–22. https://doi.org/10.1071/WF991 Flora and vegetation of the seasonally dry tropics in Mexico: Origin 0011 and biogeographical implications. Acta Botanica Mexicana, 100(1), Thuiller, W., Lafourcade, B., Engler, R., & Araújo, M. B. (2009). BIOMOD - A 149–194. platform for ensemble forecasting of species distributions. Ecography, Poynton, J. C. (1995). The “arid corridor” distribution in Africa: A search 32, 369–373. https://doi.org/10.1111/j.1600-0587.2008.05742.x for instances among amphibians. Madoqua, 19, 45–48. van Bodegom, P. M., Douma, J. C., & Verheijen, L. M. (2014). A fully Prentice, I. C., Cramer, W., Harrison, S. P., Leemans, R., Monserud, R. traits-based approach to modeling global vegetation distribution. A., & Solomon, A. M. (1992). A global biome model based on plant Proceedings of the National Academy of Sciences USA, 111(38), 13733– physiology and dominance, soil properties and climate. Journal of 13738. https://doi.org/10.1073/pnas.13045 51110 Biogeography, 19(2), 117–134. White, F. (1983). The vegetation of Africa. Paris, France: UNESCO. Ratnam, J., Bond, W. J., Fensham, R. J., Hoffmann, W. A., Archibald, Wiens, J. J., & Donoghue, M. J. (2004). Historical biogeography, ecology S., Lehmann, C. E. R., … Sankaran, M. (2011). When is a “forest” a and species richness. Trends in Ecology and Evolution, 19(12), 639– savanna, and why does it matter? Global Ecology and Biogeography, 644. https://doi.org/10.1016/j.tree.2004.09.011 20(5), 653–660. https://doi.org/10.1111/j.1466-8238.2010.00634.x Wilson, A. M., & Jetz, W. (2016). Remotely sensed high-resolution global R Core Team. (2018). R: A language and environment for statistical comput- cloud dynamics for predicting ecosystem and biodiversity distribu- ing. Vienna, Austria: R Foundation for Statistical Computing. https:// tions. PLoS Biology, 14(3), e1002415. https://doi.org/10.1371/journ www.R-projec t.org/ al.pbio.1002415 Revell, L. J. (2012). phytools: An R package for phylogenetic compara- tive biology (and other things). Methods in Ecology and Evolution, 3(2), 217–223. https://doi.org/10.1111/j.2041-210X.2011.00169.x Revell, L. J., Harmon, L. J., & Collar, D. C. (2008). Phylogenetic signal, BIOSKE TCH evolutionary process, and rate. Systematic Biology, 57(4), 591–601. https://doi.org/10.1080/10635 150802 302427 Jens J. Ringelberg is a PhD student in the group of Colin Hughes Schrire, B. D., Lavin, M., & Lewis, G. P. (2005). Global distribution patterns at the University of Zurich, Switzerland. His work focuses on the of the Leguminosae: Insights from recent phylogenies. Biologiske Skrifter, 55, 375–422. biogeography and phylogenomics of caesalpinioid legumes. Segovia, R. A., Pennington, R. T., Baker, T. R., de Souza, F. C., Neves, D. M., Davis, C. C., … Dexter, K. G. (2019). Freezing and water availabil- SUPPORTING INFORMATION ity structure the evolutionary diversity of trees across the Americas. Additional Supporting Information may be found online in the BioRxiv. https://doi.org/10.1101/728717 Silva de Miranda, P. L., Oliveira-Filho, A., Pennington, R. T., Neves, D. M., Supporting Information section. Baker, T. R., & Dexter, K. G. (2018). Using tree species inventories to map biomes and assess their climatic overlaps in lowland tropi- cal South America. Global Ecology and Biogeography, 27(8), 899–912. How to cite this article: Ringelberg JJ, Zimmermann NE, https://doi.org/10.1111/geb.12749 Weeks A, Lavin M, Hughes CE. Biomes as evolutionary Simon, M. F., Grether, R., de Queiroz, L. P., Skema, C., Pennington, R. arenas: Convergence and conservatism in the trans- T., & Hughes, C. E. (2009). Recent assembly of the Cerrado, a neo- tropical plant diversity hotspot, by in situ evolution of adaptations continental succulent biome. Global Ecol Biogeogr. to fire. Proceedings of the National Academy of Sciences USA, 106(48), 2020;29:1100–1113. https://doi.org/10.1111/geb.13089 20359–20364. https://doi.org/10.1073/pnas.09034 10106