Fig 1.
Parque Municipal Mário Viana, Nova Xavantina, Mato Grosso, Brazil. Within a transect (green points) and along an environmental gradient from cerrado sensu stricto (open canopy, drier, warmer) to cerradão (closed canopy, moister, cooler), 25 evenly spaced arrays of Y-shaped pitfall traps with drift fences were placed to sample velvet ants for 12 months; each array had a datalogger to record microclimate parameters of relative humidity and temperature. The overview map depicts the extent of Cerrado (light green) within Brazil (grey) and the municipality of Nova Xavantina location (white star).
Fig 2.
Principal Components Analysis (PCA) describing the variation of microclimatic parameters collected for 12 months in 25 arrays of Y-shaped pitfall traps with drift fences, along an environmental gradient from cerrado sensu stricto to cerradão at Parque Municipal Mário Viana, Nova Xavantina, Mato Grosso, Brazil. Filled circles represent pitfall trap arrays. Trap array numbers increase from cerrado sensu stricto to cerradão. The first two principal components (PC1 and PC2) explained 86% of the variation. Arrow length and direction are proportional to correlation between microclimatic parameters and ordination axes. Largest eigenvectors: PC1: Tsda, Tsd, Hmin, Tmax, Tmaxa, and Tmean; PC2: Hmax, Hmaxa, and Hsda. Legend: Hmaxa, absolute maximum relative humidity (%); Hmina, absolute minimum relative humidity (%); Hmax, maximum relative humidity (%); Hmin, minimum relative humidity (%); Hmean, mean relative humidity (%); Hsda, absolute standard deviation of relative humidity (%); Hsd, standard deviation of relative humidity (%); Tmaxa, absolute maximum temperature (°C); Tmina, absolute minimum temperature (°C); Tmax, maximum temperature (°C); Tmin, minimum temperature (°C); Tmean, mean temperature (°C); Tsda, absolute standard deviation of temperature (°C); Tsd, standard deviation of temperature (°C).
Fig 3.
Relative abundance distribution.
Bar plot of velvet ant species captured for 12 months in 25 arrays of Y-shaped pitfall traps with drift fences, along an environmental gradient from cerrado sensu stricto to cerradão at Parque Municipal Mário Viana, Nova Xavantina, Mato Grosso, Brazil. Bar colours code for the subtribe which the species pertain to; red: Sphaeropthalmina, blue: Pseudomethocina, orange: Ephutina, and light blue: Smicromyrmina. Legend: Cac, Callomutilla crucigera; Cev, Cephalomutilla vivata; Daa, Darditilla araxa; Da1, Darditilla sp. 01; Da2, Darditilla sp. 02; Da3, Darditilla sp. 03; Da4, Darditilla sp. 04; Da5, Darditilla sp. 05; Da6, Darditilla sp. 06; Ep2, Ephuta sp. 02; Ep5, Ephuta sp. 05; Ep6, Ephuta sp. 06; Ep7, Ephuta sp. 07; Ep8, Ephuta sp. 08; Ep9, Ephuta sp. 09; Ep11, Ephuta sp. 11; Ep12, Ephuta sp. 12; Ep14, Ephuta sp. 14; Ep15, Ephuta sp. 15; Ep16, Ephuta sp. 16; Ep18, Ephuta sp. 18; Hpm, Hoplocrates monacha; Hob, Hoplomutilla biplagiata; Hop, Hoplomutilla pollens; Hot, Hoplomutilla triumphans; Hrf, Horcomutilla fronticornis; Lmv, Lophomutilla vina; Ls1, Lophostigma sp. 01; Ls2, Lophostigma sp. 02; Mih, Mickelia harpyia; Pem, Pertyella mayri; Prs, Protophotopsis sulcifrons; Pr1, Protophotopsis sp. 01; Psg, Pseudomethoca gounellei; Ps5, Pseudomethoca sp. 05; Ps6, Pseudomethoca sp. 06; Ps7, Pseudomethoca sp. 07; Pt1, Ptilomutilla sp. 01; Suc, Suareztilla centrolineata; Taf, Tallium festivum; Ta4, Tallium sp. 04; Ta5, Tallium sp. 05; Tim1, Timulla sp. 01; Tim2, Timulla sp. 02; Tra, Traumatomutilla andrei; Trbc, Traumatomutilla bellicosa; Trbf, Traumatomutilla bellifera; Trga, Traumatomutilla gausapata; Trge, Traumatomutilla geographica; Trid, Traumatomutilla indica; Trint, Traumatomutilla integella; Trir, Traumatomutilla ira; Trl, Traumatomutilla laida; Trma, Traumatomutilla maipa; Trmo, Traumatomutilla moesta; Tro, Traumatomutilla ocellaris; Trp, Traumatomutilla parallela; Trs, Traumatomutilla sancta; Trt, Traumatomutilla tristis; Trv, Traumatomutilla vivax; Viv, Vianatilla victura; Xy1, Xystromutilla sp. 01.
Fig 4.
Sample-based (A) and individual-based (B) rarefaction curves of velvet ants captured for 12 months in 25 arrays of Y-shaped pitfall traps with drift fences, along an environmental gradient from cerrado sensu stricto to cerradão at Parque Municipal Mário Viana, Nova Xavantina, Mato Grosso, Brazil.
Fig 5.
Correspondence analysis (CA) describing the variation on the distribution of velvet ants (only those species with abundance higher than average) captured for 12 months in 25 arrays of Y-shaped pitfall traps with drift fences, along an environmental gradient from cerrado sensu stricto to cerradão at Parque Municipal Mário Viana, Nova Xavantina, Mato Grosso, Brazil. Filled circles represent pitfall trap arrays. Trap array numbers increase from cerrado sensu stricto to cerradão. The first two correspondence axes (CA1 and CA2) explained 42.7% of the variation. Legend: Da1, Darditilla sp. 01; Da6, Darditilla sp. 06; Ep5, Ephuta sp. 05; Ep6, Ephuta sp. 06; Hop, Hoplomutilla pollens; Hot, Hoplomutilla triumphans; Hrf, Horcomutilla fronticornis; Mih, Mickelia harpyia; Pem, Pertyella mayri; Psg, Pseudomethoca gounellei; Taf, Tallium festivum; Ta5, Tallium sp. 05; Tim1, Timulla sp. 01; Trbf, Traumatomutilla bellifera; Trge, Traumatomutilla geographica; Trint, Traumatomutilla integella; Trmo, Traumatomutilla moesta; Trp, Traumatomutilla parallela; Trs, Traumatomutilla sancta.
Fig 6.
Monthly distribution of velvet ant species captures and richness by traps, collected for 12 months in 25 arrays of Y-shaped pitfall traps with drift fences, along an environmental gradient from cerrado sensu stricto to cerradão at Parque Municipal Mário Viana, Nova Xavantina, Mato Grosso, Brazil. Trap array numbers increase from cerrado sensu stricto to cerradão.
Fig 7.
Constrained species ordination.
Best (reduced) Canonical Correspondence Analysis (CCA) model, depicting relationships between (A) velvet ant species (with abundance higher than average) and microclimatic parameters, and (B) pitfall trap arrays and microclimatic parameters, collected for 12 months in 25 arrays of Y-shaped pitfall traps with drift fences, along an environmental gradient from cerrado sensu stricto to cerradão at Parque Municipal Mário Viana, Nova Xavantina, Mato Grosso, Brazil. Species scores are represented by abbreviations, microclimatic parameters by arrows and pitfall trap array position by circles and numbers. Coloured filled circles code for the subtribe which the species pertain to; red: Sphaeropthalmina, blue: Pseudomethocina, orange: Ephutina, and light blue: Smicromyrmina. Grey filled circles represent pitfall trap arrays. Trap array numbers increase from cerrado sensu stricto to cerradão. Jointly, graphs reflect species distributions along the microclimate parameter space. Arrow lengths represent strength of association between microclimate parameters and ordination axes. Legend: Hmean, mean relative humidity (%); Tmina, absolute minimum temperature (°C); Da1, Darditilla sp. 01; Da6, Darditilla sp. 06; Ep5, Ephuta sp. 05; Ep6, Ephuta sp. 06; Hop, Hoplomutilla pollens; Hot, Hoplomutilla triumphans; Hrf, Horcomutilla fronticornis; Mih, Mickelia harpyia; Pem, Pertyella mayri; Psg, Pseudomethoca gounellei; Taf, Tallium festivum; Ta5, Tallium sp. 05; Tim1, Timulla sp. 01; Trbf, Traumatomutilla bellifera; Trge, Traumatomutilla geographica; Trint, Traumatomutilla integella; Trmo, Traumatomutilla moesta; Trp, Traumatomutilla parallela; Trs, Traumatomutilla sancta.
Fig 8.
Partitioning of total deviance in velvet ant species turnover along an environmental gradient from cerrado sensu stricto to cerradão at Parque Municipal Mário Viana, Nova Xavantina, Mato Grosso, Brazil, according to Generalized Dissimilarity Modelling (GDM) using distance matrices of microclimate parameters and spatial distance as predictors. The stacked histogram represents fractions of total deviance in species turnover explained exclusively by each predictor distance matrix, their shared contribution, as well as deviance left unexplained.
Fig 9.
Generalized dissimilarity modelling of velvet ant species turnover across 25 arrays of Y-shaped pitfall traps with drift fences, distributed along an environmental gradient from cerrado sensu stricto to cerradão and sampled for 12 months in Nova Xavantina, Mato Grosso, Brazil. (A) Relationship between observed velvet ant turnover for each trap array pair and predicted ecological distance between those trap arrays. (B) Relationship between observed and predicted velvet ant turnover. Generalized dissimilarity model-fitted I-splines (partial regression fits) of spatial distance (C) and selected environmental variables (D-J) as predictors. The steeper the slope of the I-spline, the greater the predicted species turnover on that portion of the gradient. The maximum height of each curve indicates the total amount of species turnover associated with that predictor and the relative importance of that predictor in explaining species turnover holding all other predictors constant.
Fig 10.
Variance Components Analysis (VCA) of microclimate preferences among taxonomic levels of velvet ants captured for 12 months in 25 arrays of Y-shaped pitfall traps with drift fences, along an environmental gradient from cerrado sensu stricto to cerradão at Parque Municipal Mário Viana, Nova Xavantina, Mato Grosso, Brazil. VCA used Restricted Maximum Likelihood (REML) to fit a linear mixed model (LMM) where the intercept is the only fixed effect. Values represent means of species scores along the first axis of a Canonical Correspondence Analysis relating number of captures to microclimate parameters and are distributed among four taxonomic levels (subfamily, tribe, subtribe and genus).
Table 1.
Variance Components Analysis.