Neotropical Entomology https://doi.org/10.1007/s13744-022-01010-4 ECOLOGY, BEHAVIOR AND BIONOMICS The Ground‑Dwelling Ant Fauna from a Cerrado Reserve in Southeastern Brazil: Vegetation Heterogeneity as a Promoter of Ant Diversity Heraldo Luis Vasconcelos1 · Rodrigo Machado Feitosa2 · Giselda Durigan3 Ruthe Emilia Oliveira Saraiva Leão1 · Karen Christina Ferreira Neves1 · Received: 12 August 2022 / Accepted: 16 November 2022 © Sociedade Entomológica do Brasil 2022 Abstract Ants represent one of the most diverse and ecologically important group of insects in tropical ecosystems, including in highly threatened ones such as the Brazilian Cerrado. Yet, a detailed understanding of the species diversity and composition of local Cerrado ant assemblages is lacking in many cases. Here we present the results of a comprehensive ant inventory performed within a region of the Cerrado (in São Paulo state) where most of the original vegetation has already been lost and where few conservation units exist. We performed consecutive surveys of the ant fauna that forage on the ground in replicated plots established in open savanna (campo sujo), dense savanna (cerrado sensu stricto), and forest (cerradão). Our surveys, with an estimated sample coverage of 99.4%, revealed a total of 219 species of ants from 60 genera, of which 36.1% were found in all the three vegetation types and 29.7% in just one. Rarefied species richness did not differ between vegetation types, but species composition differed markedly, especially between the two savannas in one hand and the forest in the other. Several species (60.1% of the 128 species analyzed) were significant “indicator” species due to their strong association with a given vegetation type. Overall, our findings reinforce the idea that habitat heterogeneity enhances ant diversity and that the mosaic of vegetation types that characterizes the Cerrado biome is one of the main factors explaining the elevated number of species that can be found at relatively small scales. Keywords Biological inventories · Hymenoptera · Diversity · Indicator species · Fire suppression · Savanna-forest transitions Introduction Species inventories play a key role in conservation biology, as they represent the foundation of studies involving the assessment and monitoring of biodiversity (Longino and Colwell 1997; Sodhi 2010). Baseline information about the species richness and composition of local assemblages Edited by Simão Dias Vasconcelos * Heraldo Luis Vasconcelos heraldo@ufu.br 1 Instituto de Biologia, Univ Federal de Uberlândia (UFU), Uberlândia, Minas Gerais, Brazil 2 Depto de Zoologia, Univ Federal do Paraná (UFPR), Curitiba, Paraná, Brazil 3 Floresta Estadual de Assis, Instituto de Pesquisas Ambientais, Assis, São Paulo, Brazil is essential to assess changes in biodiversity, such as those expected to occur in regions mostly highly threatened by human activities. Yet, in many cases, such information is lacking, and this is especially true for most insect groups. Home for the largest and most biodiverse of all tropical savannas, the Brazilian Cerrado has already lost over half its original vegetation, while only 8.2% is protected (Parente et al. 2021). These figures are even worse towards the southern limit of the Cerrado. In São Paulo state, for instance, less than 1% of the Cerrado vegetation (which once covered 14% of the area of the state) remains, and less than 20% of this vegetation is under some level of protection, either in public reserves or in private lands (Fiori and Fioravanti 2001). Furthermore, altered fire regimes are also threatening the conservation of the Cerrado biota, notably the open habitat specialists, which are susceptible to forest encroachment resulting from the suppression of savanna fires (Durigan and Ratter 2016; Abreu et al. 2017; Stevens et al. 2017). 13 Vol.:(0123456789) H. L. Vasconcelos et al. Ants (Hymenoptera: Formicidae) are a dominant group of insects in the Cerrado savannas, where they play a variety of ecological roles (Andersen and Vasconcelos 2022). The Cerrado ant fauna is highly diverse, containing at least 700 species (Andersen and Vasconcelos 2022). Nonetheless, this diversity is not uniformly distributed across the biome. In fact, contrasting to the usual latitudinal diversity pattern, within the Cerrado, more ant species are found at higher latitudes—where plant primary productivity is also higher—than at lower latitudes (Vasconcelos et al. 2018). Cerrado ant diversity is also sensitive to local variations in the structure of the vegetation. However, while some studies indicate that tree cover has a positive effect on ant diversity (Ribas et al. 2003; Pacheco and Vasconcelos 2012; Rabello et al. 2021), others have found the opposite pattern (Marques and Del-Claro 2006; Neves et al. 2013; Queiroz and Ribas, 2016). Importantly, ant species composition tends to change markedly as the structure of the vegetation changes (e.g., Neves et al. 2013; Rabello et al. 2021), and this has important implications for explaining ant diversity at the landscape scale (Pacheco and Vasconcelos 2012) and to better understand the indirect effects of fire on the Cerrado ant fauna (Maravalhas and Vasconcelos 2014; Andersen and Vasconcelos 2022). Here we present the results of a structured inventory (Longino and Colwell 1997) of the ant fauna that nest and/ or forage on ground, performed within one of the few Cerrado reserves found in São Paulo state (near the southern limit of the biome). Over a 4-year period, we sampled ants in replicated savanna and forest plots that were established at the Santa Bárbara Ecological Station (hereafter SBES). In addition to provide a comprehensive species list of the ants found at SBES, we attempted to answer the following questions: (a) How similar is the ant fauna from the different Cerrado vegetation types present at the SBES? (b) Which are the species most typical from each habitat? Overall, the results of our study point out for the importance of vegetation heterogeneity as a promoter of local ant diversity. Material and methods The Santa Bárbara Ecological Station (SBES) is located in Águas de Santa Bárbara, São Paulo, Brazil, within the coordinates 22°46′–22°51′S and 49°10–49°16′W, at an elevation of 600 to 680 m a.s.l. The local climate is classified as Koppen’s Cfa type, characterized by a wet summer and a dry winter (Alvares et al., 2013). The annual rainfall is around 1300 mm and mean monthly temperatures range from 18 to 22 °C (Meira-Neto et al. 2007). The soils are deep oxisols, characterized by high sand and low nutrient content, high saturation of aluminum, and low soil water holding capacity (Abreu et al. 2017). 13 SBES protects 2715 ha of Cerrado vegetation, including grasslands, savannas, and forests. However, since its creation in 1984 and associated with fire suppression policies, a substantial increase in tree biomass was detected over the entire area of the reserve, resulting in the loss of open habitats, thus threatening the conservation of the species associated with these habitats (Abreu et al. 2017). In 2015, a fire experiment was initiated (Durigan et al. 2020), and ant sampling took place in the plots designed for this experiment. A network of 30 plots (20 × 50 m each) was distributed across three savanna-forest transitions, distant 3 to 5 km from each other (Fig. S1). Twelve plots were established in forest areas (cerradão according to Brazilian terminology), 12 in dense savanna (cerrado sensu stricto), and six in open savanna (campo sujo). Ant sampling took place in December 2014, in January 2016, in December 2016, and in November 2020, but only the dense savanna plots were sampled in all four sampling years. The open savanna plots were sampled three times (no sampling in the second year) and the forest plots twice, once in the first and once in the last sampling year. Half of the open savanna plots and three-quarters of the dense savanna and forest plots were experimentally burned at least once over the course of the entire sampling period. A preliminary analysis, however, indicated that the experimental fires did not have a significant effect on the ground-dwelling ant fauna, at least not in the short term (Durigan et al. 2020). Five 2.5-m × 2.5-m grids were established in each sampling plot. The grids were set alongside the borders of each plot, keeping a minimum distance of 20 m between any two sampling grids. Four pitfall traps were set in each grid, totaling 20 traps per plot (over the course of the study we missed the traps installed in 17 grids, so our final sample size is 433 grids sampled rather than 450). Pitfall traps consisted of small plastic cups (250 ml, 8.5 cm high, and 7.8 cm in diameter) buried in the ground and partially filled with water and detergent. Pitfall traps remained in operation for 48 h and their contents were combined within grids. In the lab, ant workers were sorted into morphospecies and a representative specimen from each sample was dry mounted for subsequent identification using available taxonomic keys or by comparison with specimens previously identified by ant taxonomists deposited at the Zoological Collection of the Federal University of Uberlândia (UFU) and the Entomological Collection Padre Jesus Santiago Moure (DZUP) from the Federal University of Paraná (UFPR), where the specimens collected were also deposited. Sources used for species-level identification for each genus were the following: Acromyrmex (Gonçalves 1961); Crematogaster (Longino 2003); Ectatomma (Kugler and Brown 1982); Gnamptogenys, Holcoponera, and Poneracatnha (Camacho et al. 2020); Labidus and Neivamyrmex (Watkins 1976); Linepithema (Wild 2007); Megalomyrmex (Brandão 1990); Odontomachus (Brown 1976, 1978); Oxyepoecus (Albuquerque and The Ground‑Dwelling Ant Fauna from a Cerrado Reserve in Southeastern Brazil: Vegetation… Brandão 2009); Neoponera and Pachycondyla (Mackay and Mackay 2010); Pheidole (Wilson 2003); Sericomyrmex (Jesovnik and Schultz 2017); Wasmannia (Longino and Fernández 2007). Specimens for which a species-level identification was not possible received a morphospecies code (the same used in UFU´s collection). We compared ant species richness between the three types of vegetation using both sample-based (Gotelli and Colwell 2001) and coverage-based (Chao and Jost 2012) rarefaction curves. Rarefaction curves were built in R version 4.0.5 (R Core Team 2021) using the packages “iNEXT” (Hsief et al. 2016) and “ggplot2” (Wickham 2016). We compared the level of similarity in ant species composition between the three vegetation types using the Sørensen index for presence or absence data. We used the “betapart” package to calculate the overall dissimilarity between each pair of vegetation types and to partition this dissimilarity into its turnover and nestedness components (Baselga 2010). We also used the “betapart” package to calculate the “abundance-based” overall dissimilarities between vegetation types (Bray–Curtis index). Abundance was calculated as the number of samples in which a species occurred divided by the total number of samples taken in a given vegetation type. By “sample,” we mean the combined contents of four pitfall traps set within each grid of 2.5 × 2.5 m in a given year. The ant species most characteristic of each type of vegetation were identified using the indicator species analysis (Dufrêne and Legendre 1997). For this, we first built a matrix containing information about the average abundance (mean per sampling year) of each ant species in each vegetation type. The analysis was restricted to species present in at least six of the 433 samples taken over the course of this study (in all vegetation types combined), given that the remaining species were too rare for any meaningful analysis of habitat association. The indicator species analysis combines information about the relative abundances (number of samples in which the species was recorded) and relative frequencies (number of plots in which the species occurred) of each species in each habitat to calculate an indicator value (IndVal) for each species in each habitat. Indicator values range from 0 (no indication) to 100% (perfect indication). For each species, the habitat with the largest IndVal was considered the most characteristic habitat. The significance of the largest IndVal of each species was tested using the Monte Carlo test with 4999 permutations. Species which received a significant IndVal were classified as either open savanna, dense savanna, or forest specialists. Results Our surveys, with an estimated sample coverage of 99.4% (Fig. 1b), revealed a total of 219 species or morphospecies of ants from 60 genera and eight subfamilies (Table S1). Species-level identification was possible for 64% of the species collected (Table S1). Despite the comprehensive sampling effort, 41 species (18.7%) were recorded in only one sample and 16 (7.3%) in only two. Among the species collected, three—Paratrachymyrmex bugnioni (Forel, 1912), Pheidole calimana Wilson, 2003, and Strumigenys lygatrix (Bolton, 2000)—were recorded for the first time in São Paulo state. In fact, the occurrence of P. bugnioni and P. calimana at SBES represents the southernmost record for these species. Overall, the most abundant species in our surveys were Pheidole fracticeps Wilson, 2003 (found in 76% of the samples), followed by Pheidole oxyops Forel, 1908 (71.6%), and Ectatomma permagnum Forel, 1908 (53.6%). Only five species were present in all the 30 sampling plots, and these were Brachymyrmex pictus Mayr, 1887, Cyphomyrmex rimosus (Spinola, 1851), Ectatomma edentatum Roger, 1863, Pheidole fracticeps, and Pheidole triconstricta Forel, 1886. The most diverse genera were Pheidole (35 species), followed by Camponotus (22 species), Solenopsis (16 species), Mycetomoellerius (10 species), and Brachymyrmex (9 species). More samples were taken in the dense savannas and, consequently, comparatively more species were recorded in this habitat (174 species) than in either the open savannas (137 species) or forests (141 species). However, as revealed by both the sample-based and coverage-based rarefaction/ extrapolation curves (Fig. 1), these three vegetation types seem to support a similar diversity of ant species. Rarefied species richness based on sample sizes was only slightly greater in the open and dense savannas than in forest, whereas the reverse was true when comparing the sample coverage rarefied species richness (Fig. 1). About one-third (36.1%) of the 219 species recorded were found in all the three vegetation types, whereas 29.7% were found in only one (Fig. 2). Ant assemblages in the open savanna shared much more species with assemblages from dense savanna than with those from the forest. Consequently, we found less dissimilarity in species composition between open and dense savanna than between each of the savannas and forest. Furthermore, while turnover accounted for only 54.3% of the overall dissimilarity between open and dense savanna, it accounted for 97.1% of the dissimilarity between open savanna and forest and 75.2% of the dissimilarity between dense savanna and forest. We found similar levels of overall dissimilarity in species composition between open and dense savanna using either abundance (Bray–Curtis dissimilarity index = 0.280) or presence/absence data (Sørensen dissimilarity index = 0.228). In contrast, dissimilarity was much greater using abundance than presence/absence data when comparing each of the savannas with forest (Fig. 2). The indicator value analysis revealed that 60.1% of the 128 species analyzed (those present in at least 6 samples) 13 H. L. Vasconcelos et al. Fig. 1 Rarefaction (solid lines) and extrapolation (dashed lines) curves, showing the cumulative number of ant species recorded in each habitat in relation to A the sampling effort (cumulative number of samples) or B the sampling completeness (sample coverage estimator). In A, numbers within parentheses represent, respectively, the total number of samples taken and the observed species richness in each habitat. In bold is the rarified species richness (and 95% confidence intervals) in the dense savanna and forest habitats for a sam- ple size = 89 samples, which represents the total number of samples taken in open savanna (the habitat with the lowest sampling effort). Numbers in B represent, respectively, the estimated sample coverage and the observed species richness. The rarified species richness in the open and dense savanna habitats, considering the sampling coverage achieved in the forest habitat (the habitat with the lowest sampling coverage), is shown in bold and dense savannas) and four of the more “closed” habitats (primarily found both in dense savanna and forest) (Table 1). Discussion Overall species richness Fig. 2 Venn diagram showing the number of ant species recorded exclusively within a given habitat, and the number of species shared with one or two other habitats. Numbers outside the circles represent two measures of the overall dissimilarity in species composition between ant assemblages in different habitats: the Sørensen index for species presence or absence data, and (within parentheses) the Bray– Curtis index, which took into account the relative abundances of each species in each habitat presented a significant indicator value (Table 1). Of these, 33 species were primarily associated with the open savanna, 15 with the dense savanna, and 18 with the forest habitat. Another seven species were more characteristic of the savanna habitat in general (primarily found both in open 13 With an estimated sample coverage of over 99%, our ant inventory at SBES represents one of the most complete inventories of the ground-dwelling ant fauna of a Cerrado locality. In total, we recorded 219 species, of which 174 were found in cerrado stricto sensu (dense savanna), which is the most typical and dominant Cerrado vegetation. Comparing the ant diversity found at SBES with other Cerrado areas is difficult given differences in sampling methodology, effort, and/or design between our study and others (e.g., Silva et al. 2004 found much less species at two Cerrado sites in São Paulo than we found at SBES, and this can be attributed to the fact that their sampling was restricted to the ant fauna that visit sardine baits). To our knowledge, only one other study performed ant surveys in multiple types of Cerrado vegetation and in multiple years as was done in here. Interestingly, this study (Camacho and Vasconcelos 2015), which took place in a 400-ha reserve located ca. 500 km north of SBES, revealed a surprisingly similar number of grounddwelling species (226 species) as in the present study. The The Ground‑Dwelling Ant Fauna from a Cerrado Reserve in Southeastern Brazil: Vegetation… Table 1 List of the ant species with a significant indicator value, together with their indicator values (in %, where zero = no indication and 100 is perfect indication) in each of the three sampled habitats Species Open savanna indicators Acromyrmex balzani Forel, 1893 Atta laevigata Smith, 1858 Camponotus innocens Forel, 1909 Camponotus leydigi Forel, 1866 Camponotus melanoticus Emery, 1894 Camponotus substitutus Forel, 1899 Cyphomyrmex transversus Emery, 1894 Dorymyrmex sp. 6 Ectatomma brunneum Smith, 1858 Ectatomma opaciventre Roger, 1861 Ectatomma planidens Borgmeier, 1939 Ectatomma tuberculatum Olivier, 1792 Forelius brasiliensis (Forel, 1908) Forelius sp. 9 Gnamptogenys sulcata (Smith, 1858) Gracilidris pombero Wild & Cuezzo, 2006 Kalathomyrmex emeryi (Forel, 1907) Mycetagroicus cerradensis Brandão & MayhéNunes, 2001 Mycetomoellerius kempfi (Fowler, 1982) Myrmicocrypta sp.1 Neoponera verenae Forel, 1922 Pheidole jujuyensis Forel, 1913 Pheidole nr. mapinguar Pheidole schwarzmeieri Borgmeier, 1939 Pheidole sp. 106 Pheidole sp. 70 Pheidole vafra Santschi, 1923 Pogonomyrmex naegelli Emery, 1878 Rogeria sp. 1 Solenopsis nr. frank Solenopsis nr. goeldii Solenopsis nr. latastei Solenopsis substituta Santschi, 1925 Indicators of savanna habitat in general Apterostigma sp. 1 Dorymyrmex sp. 10 Ectatomma permagnum Forel, 1908 Linepithema cerradense Wild, 2007 Mycetomoellerius urichii (Forel, 1893) Mycocepurus goeldii (Forel, 1893) Pheidole nr. germani Dense savanna indicators Dinoponera grandis (Guérin-Méneville, 1838) Hypoponera sp. 3 Linepithema angulatum Emery, 1894 Linepithema micans Forel, 1908 Mycetomoellerius sp. 15 Mycetomoellerius sp. 36 Oxyepoecus rastratus (Mayr, 1887) Open savanna Dense savanna Forest 59 83 49 30 49 65 61 41 47 67 48 45 61 65 50 30 33 52 1 0 18 1 8 27 34 3 0 28 12 26 2 9 11 1 0 1 0 0 0 0 4 0 0 0 0 0 0 1 0 0 12 0 0 1 47 48 46 30 55 51 41 31 59 58 37 57 77 58 53 0 17 0 1 30 11 23 1 33 5 6 29 13 32 7 0 2 1 0 11 0 0 0 0 0 0 1 0 0 0 49 54 39 46 48 42 44 40 46 48 44 39 46 53 1 0 1 8 0 9 1 19 0 0 36 8 12 4 54 42 58 52 62 50 49 0 0 0 7 15 1 0 13 H. L. Vasconcelos et al. Table 1 (continued) Species Open savanna Dense savanna Forest Pachycondyla harpax (Fabricius, 1804) Pheidole radoszkowskii Mayr, 1884 Pseudomyrmex pallidus (Smith, 1855) Pseudomyrmex tenuis (Fabricius 1804) Rogeria sp. 5 Solenopsis loretana Santschi, 1936 Solenopsis nr. basalis Wasmannia auropunctata (Roger, 1863) Indicators of dense savanna and forest Atta sexdens Linnaeus, 1758 Nylanderia docilis (Forel, 1908) Pheidole oxyops Forel, 1908 Pheidole fracticeps Wilson, 2003 Forest indicators Acromyrmex subterraneus (Forel, 1893) Camponotus iheringi Forel, 1908 Camponotus lespesii Forel, 1886 Carebara brevipilosa (Fernández, 2004) Cyphomyrmex laevigatus Weber, 1938 Hypoponera nr. parva Hypoponera sp. 7 Hypoponera sp. 13 Linepithema aztecoides Wild, 2007 Nylanderia nr. caeciliae Odontomachus chelifer (Latreille, 1802) Pachycondyla striata Smith, 1858 Pheidole calimana Wilson, 2003 Pheidole lovejoyi Wilson, 2003 Pheidole nr. obscurior Pheidole sp. 136 Solenopsis nr. brevicolis Solenopsis nr. westwoodi 20 38 9 12 0 25 3 39 54 55 62 47 50 54 49 56 20 2 3 3 0 12 19 2 0 10 18 26 50 48 36 31 34 37 43 43 0 0 5 0 0 0 0 0 0 0 3 14 0 1 13 0 15 3 0 1 7 3 0 1 0 1 6 1 34 11 1 0 32 0 21 0 56 46 79 59 58 38 73 79 54 54 56 71 95 71 56 42 62 35 fact that over 200 ant species can be found within relatively small Cerrado areas (as shown here and in previous studies; Camacho and Vasconcelos 2015; Oliveira and Feitosa 2021) is surprising, notably when considering that current estimates indicate the occurrence of only ca. 700 species in the entire biome (Andersen and Vasconcelos 2022). That local ant diversity is high relative to the regional diversity can be explained, at least in part, because many of the species found within local assemblages are widespread species. This is the case for most of the species that are numerically dominant within local assemblages, such as P. oxyops and E. permagnum. However, this is not to say that Cerrado ants are of least conservation concern as the rampant rate of habitat destruction in the biome is likely to be reducing the number, size, and level of isolation of the remaining populations. One example is that of Dinoponera grandis (GuérinMéneville, 1838) (Fig. 3), a relatively widespread species 13 in the southern portion of South America (Dias and Lattke 2021) but whose population recorded at SBES represents one of the last known populations of this remarkable species in São Paulo (R. Feitosa & J. Lattke, unpublished). Furthermore, many Cerrado ant species appear to be naturally rare and, as such, are also of conservation concern. Examples of rare species (currently known from only a very few locations in South America) found at SBES include Gnamptogenys nana Kempf, 1960, Neoponera agilis Forel, 1901, and Strumigenys lygatrix (Fig. 3). Interestingly, none of these three species were recorded in the plots that became forest as result of savanna woody encroachment. Similarly, with a single exception, no D. grandis was found in the forest plots. Finally, it is important to mention that most of the speciesrich genera of Cerrado ants still require taxonomic revision and/or lack molecular information, and therefore, many of the individual species that we now regard as widespread may The Ground‑Dwelling Ant Fauna from a Cerrado Reserve in Southeastern Brazil: Vegetation… Fig. 3 Species of potential conservation concern recorded at Santa Bárbara Ecological Station. A Dinoponera grandis (DZUP 549,812, image by Amanda Dias), B Gnamptogenys nana, C Neoponera agilis, and D Strumigenys lygatrix in fact represent a complex of geographically distinct species (Feitosa et al. 2022). Species diversity and composition in different vegetation types Our results indicate a high level of similarity in the overall species richness of ground-dwelling ants between open savanna (campo sujo), dense savanna (cerrado stricto sensu), and forest (cerradão), despite the marked structural and floristic differences between these three vegetation types. Interestingly, our coverage-based rarefaction curves showed that, at an estimated sample coverage of about 80% or less, ant species richness was lower in forest than in either open or dense savannas, while at higher levels of sample coverage within each habitat this difference disappeared. This points out for the importance of considering differences in sampling completeness when comparing ant diversity in different habitats (cf. Chao and Jost 2012). In contrast to species richness patterns, we found marked differences in ant species composition between the different Cerrado habitats. There was a large number of species that showed significant habitat associations, even though the three types of vegetation we surveyed were in close proximity to each other (i.e., dispersal limitation was not an issue). These findings reinforce the idea that habitat heterogeneity 13 H. L. Vasconcelos et al. enhances ant diversity at the landscape scale (Pacheco and Vasconcelos 2012; Neves et al. 2013). As observed also for plants (Durigan et al. 2003), the mosaic of vegetation types that characterizes the Cerrado seems, therefore, one of the main factors in explaining the high number of ant species that can be found at relatively small spatial scales. Maintaining the mosaic of vegetation types should, therefore, be the target for management interventions aiming at Cerrado biodiversity. Differences in species composition were particularly pronounced between the forest in one hand and the two savanna habitats in the other, although somewhat greater between open savanna and forest than between dense savanna and forest. In addition to find several species that seem exclusive of the forest or of the savanna habitat, we often found sharp differences in the abundance of those species that were recorded in both types of habitats. In other words, species that were relatively common in forest were relatively rare in the savannas, and vice-versa, suggesting that the former may represent marginal habitats for ant populations that are most typical in savanna (i.e., they are sink populations), while savannas represent marginal habitats for forest ants. In this sense, woody encroachment in fire-suppressed savanna areas is likely to represent a threat for the conservation of the savanna ant fauna not only because the forests established in former savanna areas are inadequate habitats for the savanna specialists (Andersen et al. 2006; Abreu et al. 2017), but also because these forests, over the long run, may not be able to sustain the populations of the species that are not strictly restricted to savanna. Our study has listed a relatively large number of ant species that are primarily associated with the forest or with the savanna habitats (or, else, are habitat generalists). Compilation of such information in other localities would be instrumental to determine the extent to which savanna and forest ant assemblages represent two alternative compositional states and thus to assess the generality of our findings. Further information about the habitat affinities of different ant species would also be of value for studies that use ants as bioindicators of ecological change (reviewed in Andersen 1997; Ribas et al. 2012). This is evidenced, for instance, by a study which showed that the prevalence of savannaassociated ant species is strongly correlated with the extent of forest fires in the Amazon (Paolucci et al. 2017). Finally, information on habitat affinities may also be of importance for assessing the impacts of climatic change on ants, given that savanna and forest ants can respond differentially to the same climatic drivers (Vasconcelos et al. 2018). Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s13744-022-01010-4. Acknowledgements We thank Jésica Vieira, Lino Zuanon, Elmo Koch, Jonas Maravalhas, and Renata Pacheco for their invaluable help with 13 the field and/or laboratory work, and W. Hoffmann who helped design and set up the fire experiment at SBES. Author Contribution H. L. V. and G. D. designed the study; K. C. F. N., R. E. O. S. L., and H. L. V. were involved in data collection and preparation; R. M. F. provided taxonomic expertise; H. L. V. led the writing; and all authors contributed to manuscript preparation. Funding Financial support was provided by the Brazilian Council of Research and Scientific Development (grants 304628/2020–4 to HLV, 301495/2019–0 to RMF, and 309709/2020–2 to GD) and the U.S. National Science Foundation for funding the fire experiment (NSF grant DEB1354943). Declarations Conflict of Interest HLV is member of the Editorial Board of Neotropical Entomology, and the manuscript was independently handled by another editor. References Abreu RCR, Hoffmann WA, Vasconcelos HL, Pilon NA, Rossatto DR, Durigan G (2017) The biodiversity cost of carbon sequestration in tropical savanna. Sci Adv 3:594–598. https://doi.org/10.1126/ sciadv.1701284 Alvares CA, Stape JL, Sentelhas PC, Gonçalves JLM, Sparovek G (2013) Köppen’s climate classification map for Brazil. Meteorol Z 22:711–728. https://doi.org/10.1127/0941-2948/2013/0507 Albuquerque NL, Brandão CRF (2009) A revision of the Neotropical Solenopsidini ant genus Oxyepoecus Santschi, 1926 (Hymenoptera: Formicidae: Myrmicinae): 2. Final Key for species and revision of the Rastratus species-group. Pap Avulsos Zool 49:289–309. https://doi.org/10.1590/S0031-10492009002300001 Andersen AN (1997) Using ants as bioindicators: multiscale issues in ant community ecology. Conserv Ecol 1:8. https://doi.org/10. 5751/ES-00014-010108 Andersen AN, Hertog T, Woinarski JC (2006) Long-term fire exclusion and ant community structure in an Australian tropical savanna: congruence with vegetation succession. J Biogeogr 33:823–832. https://doi.org/10.1111/j.1365-2699.2006.01463.x Andersen AN, Vasconcelos HL (2022) Historical biogeography shapes functional ecology: inter-continental contrasts in responses of savanna ant communities to stress and disturbance. J Biogeogr 49:590–599. https://doi.org/10.1111/jbi.14343 Baselga A (2010) Partitioning the turnover and nestedness components of beta diversity. Global Ecol Biogeogr 19:134–143. https://doi. org/10.1111/j.1466-8238.2009.00490.x Brandão CRF (1990) Systematic revision of the Neotropical ant genus Megalomyrmex Forel (Hymenoptera: Formicidae: Myrmicinae), with the description of thirteen new species. Arq Zool 31:411– 481. https://doi.org/10.11606/issn.2176-7793.v31i5p1-91 Brown WL (1976) Contributions toward a reclassification of the Formicidae. Part VI. Ponerinae, tribe Ponerini, subtribe Odontomachiti. Section A. Introduction, subtribal characters. Genus Odontomachus. Stud Entomol 19:67–171 Brown WL (1978) Contributions toward a reclassification of the Formicidae. Part VI. Ponerinae, tribe Ponerini, subtribe Odontomachiti. Section B. Genus Anochetus and bibliography. Stud Entomol 20:549–638 Camacho GP, Vasconcelos HL (2015) Ants of the Panga Ecological Station, a Cerrado reserve in central Brazil. Sociobiology 62:281– 295. https://doi.org/10.13102/sociobiology.v62i2.281-295 The Ground‑Dwelling Ant Fauna from a Cerrado Reserve in Southeastern Brazil: Vegetation… Camacho GP, Franco W, Feitosa RM (2020) Additions to the taxonomy of Gnamptogenys Roger (Hymenoptera: Formicidae: Ectatomminae) with an updated key to the New World species. Zoo 3:450–476 https://doi.org/10.11646/zootaxa.4747.3.2 Chao A, Jost L (2012) Covered-based rarefaction and extrapolation: standardizing samples by completeness rather than size. Ecology 93:2533–2547. https://doi.org/10.1890/11-1952.1 Dias AM, Lattke JE (2021) Large ants are not easy - the taxonomy of Dinoponera Roger (Hymenoptera: Formicidae: Ponerinae). Eur J Taxon 784:1–66. https://doi.org/10.5852/ejt.2021.784.1603 Dufrêne M, Legendre P (1997) Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecol Monogr 67:345–366. https://doi.org/10.2307/2963459 Durigan G, Ratter JA (2016) The need for a consistent fire policy for Cerrado conservation. J Appl Ecol 53:11–15. https://doi.org/10. 1111/1365-2664.12559 Durigan G, Siqueira MF, Franco GADC, Bridgewater S, Ratter JA (2003) The vegetation of priority areas for cerrado conservation in São Paulo State, Brazil. Edinb Journ Bot 60:217–241. https:// doi.org/10.1017/S0960428603000155 Durigan G, Pilon NAL, Abreu RCR, Hoffmann WA, Martins M, Fiorillo BF, Antunes AZ, Carmignotto AP, Maravalhas JB, Vieira J, Vasconcelos HL (2020) No net loss of species diversity after prescribed fires in the Brazilian savanna. Front Glob Chang 3:13. https://doi.org/10.3389/ffgc.2020.00013 Feitosa RM, Camacho GP, Silva TSR, Ulysséa MA, Ladino N, Oliveira AM, Albuquerque EZ et al (2022) Ants of Brazil: an overview based on 50 years of diversity studies. Syst Biodiv 20:1. https://doi.org/10.1080/14772000.2022.2089268 Fiori AM, Fioravanti C (2001) Os caminhos para salvar o Cerrado paulista. Pesquisa FAPESP. https://revistapesquisa.fapesp.br/oscaminhos-para-salvar-o-cerrado-paulista. Accessed July 2022 Gonçalves CR (1961) O gênero Acromyrmex no Brasil (Hym. Formicidae). Stud Entomol 4:113–180 Gotelli NJ, Colwell RK (2001) Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecol Lett 4:379–391. https://doi.org/10.1046/j.1461-0248. 2001.00230.x Hsief TC, Ma KH, Chao A (2016) iNEXT: an R package for rarefaction and extrapolation of species diversity (Hill numbers). Method Ecol Evol 7:1451–1456. https://doi.org/10.1111/2041210X.12613 Jesovnik A, Schultz TR (2017) Revision of the fungus-farming ant genus Sericomyrmex Mayr (Hymenoptera, Formicidae, Myrmicinae). ZooKeys 670:1–109. https://doi. org/ 10. 3897/ zooke ys.670.11839 Kugler C, Brown W (1982) Revisionary and other studies on the ant genus Ectatomma, including the description of two new species. Search Agric (ithaca N y) 24:1–8 Longino JT (2003) The Crematogaster (Hymenoptera, Formicidae, Myrmicinae) of Costa Rica. Zootaxa 151:1–150 https://doi.org/ 10.11646/zootaxa.151.1.1 Longino JT, Colwell RK (1997) Biodiversity assessment using structured inventory: capturing the ant fauna of a tropical rain forest. Ecol Appl 7:1263–1277. https://doi.org/10.2307/2641213 Longino JT, Fernández F (2007) Taxonomic review of the genus Wasmannia. In: Snelling RR, Fisher BL, Ward PS (eds) Advances in ant systematics: homage to E.O. Wilson – 50 years of contributions. Memoirs of the American Entomological Institute, Philadelphia, USA, 271–289 Mackay WP, Mackay EE (2010) The systematics and biology of the new world ants of the genus Pachycondyla (Hymenoptera: Formicidae). Edwin Mellon Press, Lewiston, USA Maravalhas J, Vasconcelos HL (2014) Revisiting the pyrodiversity– biodiversity hypothesis: long-term fire regimes and the structure of ant communities in a Neotropical savanna hotspot. J Appl Ecol 51:1661–1668. https://doi.org/10.1111/1365-2664.12338 Marques GDV, Del-Claro K (2006) The ant fauna in a Cerrado area: the influence of vegetation structure and seasonality (Hymenoptera: Formicidae). Sociobiol 47:235–252 Meira-Neto JA, Martins FR, Valente GE (2007) Composição florística e espectro biológico na Estação Ecológica de Santa Bárbara, Estado de São Paulo, Brasil. Rev Arv 31:907–922. https://doi.org/ 10.1590/S0100-67622007000500015 Neves FS, Queiroz-Dantas KS, da Rocha WD, Delabie JHC (2013) Ants of three adjacent habitats of a transition region between the Cerrado and Caatinga biomes: the effects of heterogeneity and variation in canopy cover. Neotrop Entomol 42:258–268. https:// doi.org/10.1007/s13744-013-0123-7 Oliveira AM, Feitosa RM (2021) Save the survivors: the remarkable ant diversity of the last protected fragment of savanna in Southern Brazil. Ins Soc 68:49–58. https:// doi. org/ 10. 1007/ s00040-020-00804-2 Parente L, Nogueira S, Baumann L, Almeida C, Maurano L, Affonso AG, Ferreira L (2021) Quality assessment of the PRODES Cerrado deforestation data. Remot Sens Appl Soc Environ 21:100444. https://doi.org/10.1016/j.rsase.2020.100444 Pacheco R, Vasconcelos HL (2012) Habitat diversity enhances ant diversity in a naturally heterogeneous Brazilian landscape. Biodivers Conserv 21:797–809. https:// doi. org/ 10. 1007/ s10531-011-0221-y Paolucci LN, Schoereder JH, Brando PM, Andersen AN (2017) Fireinduced forest transition to derived savannas: cascading effects on ant communities. Biol Conserv 214:295–302. https://doi.org/10. 1016/j.biocon.2017.08.020 Queiroz ACM, Ribas CR (2016) Canopy cover negatively affects arboreal ant species richness in a tropical open habitat. Braz J Biol 76:864–870. https://doi.org/10.1590/1519-6984.02015 R Core Team (2021) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/ Rabello AM, Parr CL, Queiroz AC, Braga DL, Santiago GS, Ribas CR (2021) Taxonomic and functional approaches reveal different responses of ant assemblages to land-use changes. Basic Appl Ecol 54:39–49. https://doi.org/10.1016/j.baae.2021.04.001 Ribas CR, Schoereder JH, Pic M, Soares SM (2003) Tree heterogeneity, resource availability, and larger scale processes regulating arboreal ant species richness. Austral Ecol 28:305–314. https:// doi.org/10.1046/j.1442-9993.2003.01290.x Ribas CR, Campos RB, Schmidt FA, Solar RR (2012) Ants as indicators in Brazil: a review with suggestions to improve the use of ants in environmental monitoring programs. Psyche: J Entomol 2012:23. https://doi.org/10.1155/2012/636749 Silva RR, Brandão CR, Silvestre R (2004) Similarity between Cerrado localities in Central and Southeastern Brazil based on the dry season bait visitors ant fauna. Stud Neotrop Fauna Environ 39: 191–199. https://doi.org/10.1080/01650520412331271783 Sodhi NS (2010) Invaluable biodiversity inventories. In: Sodhi NS, Ehrlich PR (eds) Conservation biology for all. Oxford University Press, Oxford, pp 40–41 Stevens N, Lehmann CER, Murphy BP, Durigan G (2017) Savanna woody encroachment is widespread across three continents. Glob Chang Biol 23:235–244. https://doi.org/10.1111/gcb.13409 Vasconcelos HL, Maravalhas JB, Feitosa RM, Pacheco R, Neves KC, Andersen AN (2018) Neotropical savanna ants show a reversed latitudinal gradient of species richness, with climatic drivers reflecting the forest origin of the fauna. J Biogeogr 45:248–258. https://doi.org/10.1111/jbi.13113 Watkins JF (1976) The identification and distribution of New World army ants (Dorylinae: Formicidae). Baylor University Press, Waco, Texas, USA 13 H. L. Vasconcelos et al. Wickham H (2016) ggplot2: elegant graphics for data analysis. Springer-Verlag, New York. https:// doi. org/ 10. 1007/ 978-3-319-24277-4 Wild AL (2007) Taxonomic revision of the ant genus Linepithema (Hymenoptera: Formicidae). Univ Calif Plub Entomol 126:1–151 Wilson EO (2003) Pheidole in the New World. A dominant, hyperdiverse ant genus. Harvard University Press, Cambridge, MA Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 13 Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.