Bat communities of savanna biome in the Kruger National Park, South Africa

Bat communities of savanna biome in the Kruger National Park, South Africa | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Bat communities of savanna biome in the Kruger National Park, South Africa Markéta Staňková, Martin Hejda, Erin R. Brinkley, Jan Čuda, Sylvain Delabye, and 9 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5295291/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract The savanna habitats often harbour abundant and species-rich bat communities. Whether they represent mere ad hoc assemblages of incidentally co-occurring forms or distinct entities integrated by locally specific adaptations and balanced resource partitionings is largely unknown, as are the natural drivers shaping community variation at different spatial scales. An extensive dataset (130,888 acoustic bat records, 31 OTUs) was collected in 60 plots across Kruger National Park (KNP), South Africa; the plots were located (i) at perennial rivers, (ii) at seasonal rivers, and (iii) on dry crests away from any water source. Besides the effect of water availability, distance to campsites, and microgeographic variation on bat community richness and structure, we revealed (i) extensive homogeneity in community structure at local, subregional, and regional scales contrasting to a mosaic between-plot variation, (ii) absence of robust effects of environmental biotic and abiotic predictors on the distribution of individual OTUs and community variation, (iii) nearly identical pattern of habitat preferences in all community members approaching the centroid of KNP habitat variation, and (iv) an exceptionally high degree of community nestedness. These results suggest that the bat community of the KNP savanna biome represents a single entity consistently integrated with a network of coexistence relations that probably arose locally during long savanna history. savanna bats community structure spatial variation nestedness Kruger NP South Africa Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Introduction The savanna biome represents an essential component of global biodiversity that occupies a fifth of the earth’s land surface 1 , 2 and hosts a huge variety of biota that are rare in other habitats. This is particularly valid for southern Africa, where the savanna biome covers more than 54% of the total area 3 . The savanna biome is often presented as a dynamic mixture of semi-open habitats interfacing tropical forest and pure grassland biomes, with a climate characterized by low to intermediate rainfall and mild seasonality combined with tropic to subtropic thermal conditions 2 . Sankaran et al. 1 demonstrated that in Africa, the arid and semiarid savanna represents a “stable” formation in areas with mean annual precipitation of < 650 mm, while areas with higher precipitation represent an “unstable” system where the coexistence of herbs and trees depends on steady effects of disturbing factors (fire, herbivores). Yet, responses of particular biotic communities of savanna to extrinsic drivers promoting the habitat mosaic differ 4 , similar to responses of diverse community members to variation in vegetation cover and other variables of environmental context. The multidisciplinary project MOSAIK (Monitoring Savanna Biodiversity in Kruger National Park) was established to examine the patterns of spatial and temporal variation in syntopic savanna communities (plants, insects, birds, large mammals, and bats) by using a standardized monitoring design that consisted of 20 triplets of fixed plots (typically 6–8 km apart) each covering three main habitat types defined with regard to water availability (perennial rivers, seasonal rivers, and dry crests), disturbance levels, and landsystems of the region (Fig. 1 ). This design enabled to attribute the local, subregional and regional scales to patterns of within-plot variation 5 – 8 . The Kruger National Park (KNP) represents a suitable model area for such studies. It covers a large landscape unit of lowland semiarid savanna (19,175 km 2 , mean elevation 342 m a.s.l., mean annual precipitation 511 mm), protected for 120 years from anthropogenic influence. With all that, it provides an exceptional opportunity to analyze the organization of the savanna biome by natural drivers to the full extent. A plethora of previous research projects undertaken in KNP 4 , 9 – 12 provide a robust contextual background. Among others, this concerns detailed climatic and environmental data from high-resolution remote sensing 13 , 14 and diverse records of large-scale programs of long-term ground monitoring 15 – 17 . Thus, for KNP, comprehensive information is available on local variation in the biotic and abiotic drivers of landscape heterogeneity and its spatiotemporal dynamics 15 , 17 , including currently changing trends 5 , 18 such as, e.g., impact of elephants on large-scale pattern of habitat mosaic 19 – 21 . The present study deals with bats, a group for which KNP represents one of the pertinent hotspots of African diversity; inferring from more than 50 literary sources focused on bats in this area, the park is also one of the best-investigated regional units in Africa. Compared to former reports surveying mostly records obtained with the aid of mist netting 22 , 23 or details of reproduction biology of several local species 24 , the more recent studies on bat communities in Kruger NP and close surroundings are patterned by the influx of innovative field techniques, first of all, the application of ultrasonic detectors 25 and automated acoustic recordings. Detailed data on interspecific variation in echolocation calls of particular species and investigations on the complementarity of acoustic and non-acoustic survey methods 26 – 30 created a robust methodological platform enabling to integrate results of acoustic monitoring into macroecological analyses addressing the large-scale patterns of bat diversity in South Africa 31 – 36 . Yet, in contrast to robust information about the natural drivers associated with regional and supraregional variation in bat communities, the patterns of microgeographic variation at subregional and local scales and the feedback loop between local and regional dynamics still have not been adequately comprehended 36 . Bats differ from other animal groups studied within MOSAIK by extreme mobility (not constrained by territoriality as in birds), disposing them to respond rapidly to spatial variation in dispersal and capacity of their feeding resources, notwithstanding some unique features of their biology: small body size, temporal aggregation in day-time resting colonies, dependence upon the availability of taxon-specific roosts, etc. Studies in the Neotropical region revealed that the abundances and species richness of bat communities in the savanna biome correspond to those of communities in the rainforest habitats 37 – 44 and this obviously holds also for the Ethiopian region. Numerous studies addressing this topic in southern and central Africa were comprehensively reviewed by Schoeman and Monadjem 36 . Here, the peak richness of savanna bats occurs in northeastern southern Africa, where at local and regional scales, savanna bat assemblages include more than 30 species 31 , 45 – 48 . Schoeman and Monadjem 36 stress that in terms of alpha and gamma diversity, bats exceed any other group of mammals inhabiting the savanna subregion and raise the question of whether these species-rich assemblages are structured in any way. Thus, the central questions for any study of savanna bat communities are whether they represent (i) mere ad hoc constituted assemblages of incidentally co-occurring species attracted by the local patches of the habitats to which they are primarily adapted 49 or (ii) integrated entities established by a long-term coexistence and balanced partitioning of the resources inherently provided by the savanna biome and its dynamics 50 , 51 ? Our paper, based on an extensive dataset of automated whole-night acoustic recordings of foraging bat communities collected under standardized setting of the MOSAIK project, is intended to contribute to these topics by testing the effects of diverse natural drivers, analyzing variation in community patterns at point, local, and regional scales and revealing the commonalities across these scales. Results Structure of the dataset. Within the set of all 120 primary records (plot/night), the total abundances (number of individual bat passes recorded during one night) ranged from 22 to 7,003 (mean ± S.D. = 1090 ± 1199), the number of recorded OTUs (acoustic parataxa) varied from 4 to 19 (mean 12.3 ± 3.22). For the sets of 60 plot records (February and November samples merged), the respective values were abundances of 122 to 8,565 (mean 1887 ± 1642, for spatial pattern see Fig. 2 A), and OTUs from 10 to 22 (mean 15.0 ± 2.32). The rank distributions (RAD) of abundances and species richness for seasonal plot records and all plots merged fit well to a log-normal scaling with both variables mutually evenly scaled (R 2 = 0.358 and 0.228, respectively) except for a set of 11 samples with excessively high abundances (Supplementary File S2.1). In total, 93.7% of all samples were composed exclusively of OTUs belonging to Vespertilionidae, Molossidae, and Emballonuridae; representation of other groups was more or less incidental, and they were omitted from the present analysis (compare Supplementary File S 1 and Table 1 ). The rank distribution of total abundances of individual OTUs (Supplementary File S2.1) showed a nearly ideal log-normal distribution (R 2 = 0.87). The core of community samples in all plots in both seasons was composed of four eudominant (d > 10%) and three dominant (10% 5%) OTUs (CHAPUM – Chaerephon pumilus , TADAEG – Tadarida aegyptiaca , EPTHOT – Eptesicus hottentotus (but see below), NEONAN – Afronycteris nana , and CHAANS – Chaerephon ansorgei , NEOCAP – Laephoptis capensis , SCODIN – Scotophilus dinganii , respectively) with slight differences in their abundance ranks among particular geographic units (and particularly between February and November samples). Together with other five, these OTUs appeared in more than 90% of the samples; Table 1 , Fig. 2 B). Effect of environmental variables on bat species richness, diversity, abundance, and composition. Habitat type (perennial rivers, seasonal rivers, crests) had a highly significant effect (p < 0.001) on bat species richness. Presence/absence of mopane ( Colophospermum mopane , a dominant tree species at some plots; p = 0.001), landsystem (p = 0.002), north versus south (p = 0.004), longitude (p = 0.04), and distance to camps p = 0.042) also significantly affected bat species richness. The relation between distance to camps and bat species richness contained a significant quadratic term. The effect of mopane was also significant as a continuous variable based on a visual estimate of its cover (p = 0.039; Fig. 3 ). Further, there were significant interactions between the north/south location of plots and habitat as well as between landsystem and habitat (p = 0.006 and p = 0.033, respectively). In terms of the direction of these effects, plots near perennial and seasonal rivers harboured significantly more species compared to plots on crests (p = 0.002 and p = 0.001, respectively). Further, plots in the Satara landsystem harboured more bat species compared to plots in the Phalaborwa and Letaba landsystems (p = 0.007 and p = 0.014, respectively). Plots in the south of KNP harboured more species compared to plots in the north. Similarly, plots without dominant mopane harboured more species than plots with mopane present (Tables 2 and 3 ). Table 2 Direct ordination analysis (RDA) on the effects of environmental variables on the dominance structure of bat community in particular sites (as a response variable, n = 120). A list of the variables with significant effects revealed by forward selection is presented. categorical explained vairability (%) p-value all predictors 12.5 0.024 vegetation parameters all predictors 47.6 0.006 richness of shrubs 14.2 0.008 PCO1 - vegetation 12.7 0.014 PCO3 - vegetation 11.4 0.032 grass cover 7.7 0.044 richness of herbs 7.4 0.046 environmental context all predictors 45.6 0.028 variability in precipitation (S.D.) 24.8 0.002 distance to parkboundary 15.4 0.002 lattitude 8.9 0.034 Habitat type had a highly significant effect on the Shannon diversity of bats (p < 0.001), and there were also significant effects of mopane presence/absence (p = 0.002), mopane cover (p = 0.003), landsystem (p = 0.007), north/south location (p = 0.012), and distance to streams (p = 0.039). The relation between Shannon diversity and distance to streams contained a significant quadratic term. There was also a significant interaction between landsystem and habitat, and between north/south and habitat (p = 0.009 and p = 0.01, respectively). Plots near perennial and seasonal rivers showed significantly higher Shannon diversity of bat species compared to plots on crests (both p = 0.001). Interestingly, Tukey HSD method did not identify any significant contrasts among landsystems, despite their overall significant effect on the Shannon diversity. Similarly to species richness, plots in the south of KNP showed higher Shannon diversity of bat species compared to plots in the north, and the same holds for the plots without mopane compared to plots with mopane present (Tables 2 and 3 ). The type of habitat, distance to streams, and PCO2 (second ordination axis representing the vegetation gradient) had significant effects on the Pielou evenness of bat species (p = 0.006, p = 0.014, and p = 0.017, respectively). The effect of PCO2 contained a significant quadratic term. Similarly to bat species richness and Shannon diversity, plots near perennial rivers showed a significantly higher Pielou evenness compared to plots on crests (p = 0.007). Concerning the total abundance of bats as a response variable, the effects of habitat and distance to rivers were highly significant (both p < 0.001), as was that of waterbody area (waterSum) (p = 0.001), with a significant quadratic term. There was also a significant interaction between bedrock and distance to rivers (p = 0.03). Plots near perennial rivers showed a higher total abundance of bats compared to plots near seasonal rivers and on crests (p = 0.017 and p < 0.001, respectively). Similarly to plots near perennial rivers, those near seasonal rivers also showed a higher abundance of bats compared to plots on crests (p < 0.001). Concerning the effects on the species composition of bat communities, all three groups of predictors (categorical, vegetation, abiotic environment) had significant overall effects (p = 0.024, p = 0.006 and p = 0.028, respectively). Vegetation characteristics explained most variability (47.6%), followed by environmental context (45.6%) and categorical predictors (12.5%; Table 2 ). Of vegetation parameters, the forward selection procedure identified the richness of shrubs (p = 0.008), vegetation PCO1 (p = 0.014), vegetation PCO3 (p = 0.032), grass cover (p = 0.044), and richness of herbs (p = 0.046) as significant predictors; Fig. 4 ). These vegetation parameters explained 14.2%, 12.7%, 11.4%, 7.7% and 7.4% of variability, respectively. Variability in precipitation (expressed as its standard deviation), distance to the park boundary, and latitude were the most influential abiotic environmental predictors (p = 0.002, p = 0.002 and p = 0.034, respectively), explaining 24.8%, 15.4% and 8.9% of variability, respectively; Table 3 ). Interestingly, no significant effects were found for the categorical predictors, even though their overall effect was significant. Also, we did not find a significant effect of insect supply (weight of insects in light traps taken synchronously with bat recording at a plot) on the above-surveyed community variables or the abundance of any single OTU. The regression tree analyses (Supplementary File S3) revealed a key effect of triplet specifities upon either species richness, Shannon diversity, Pielou evenness, and abundance, predominating over those of other predictors (including water availability, mopane cover, distance to camps, herb diversity or bedrock). None of the environmental predictors, except water availability, significantly affected the variation in the quantitative representation of individual OTUs in particular community samples, (comp. also Pielou evenness in Table 3 ). Further comparison dealing with abundances and dominances of individual OTUs (Supplementary File S2.3) showed a mosaic of weakly pronounced effects of few environmental predictors, in most instances restricted to one or few OTUs, except for the following factors: season (valid for 10 OTUs), habitat type (access to water) for eight OTUs, triplet (microgeographic variation) for seven OTUs, distance to the nearest campsite (valid for five OTUs), and species richness of herbs (valid for three OTUs; Supplementary File S2.3). A comparison of the community structure in plots delimited by contrasting states of the two most influencing environmental variables (habitat, mopane; Supplementary File S2.4) showed only shallow differences, at least for most of the core OTUs. Variation in community structure. In all community samples, the two major foraging guilds, i.e., the clutter-edge foragers (Vespertilionidae) and open-air foragers (Molossidae), were represented with nearly equal abundances. Their abundance ratio (Vespertilionidae/Molossidae) varied from 0.13 to 3.47 around a mean value of 1.09 ± 0.80 (mean ± S.D.); in plots with high abundance (n > 2,000) it was 0.24–3.47, with the mean of 1.10 ± 0.89, and in those with low abundances (n < 1000) the corresponding values were 0.13–2.91 and 1.18 ± 0.76, respectively. The correspondence analysis (Fig. 5 ) revealed nearly equidistant dispersal of community members in the component space of abundance distributions and distinct divergence between the two major guilds. The projection of DC2 and DC3 axes (Fig. 5 B) showed the extensive similarity among all molossid OTUs, the dissimilarity of EPTHOT to other vespertilionids (supporting an alternative taxonomic affiliation of that acoustic parataxon in Fig. 5 CD; see Discussion), and distant position of an emballonurid (TAPMAU). The 3D reconstruction of phenetic packing (Fig. 6 ) also suggested a nearly equidistant distribution of particular community members with the concentration of core elements along the community centroid except for a marginal position of NEONAN ( Afroromicia nana ) and PIPRUS ( Pipistrellus rusticus ) that markedly differed from others in response to particular environmental predictors (compare Fig. 4 and Supplementary File S2.3). The comparison of species composition and dominance across all 120 primary plot records revealed a strong signal of the overall homogeneity of the community structure. The Jaccard index as a measure of similarity in species composition ranged from 0.306–0.737, with a mean 0.634 ± 0.153, while values of Bray-Curtis index of dissimilarity in dominance structure were low, BC = 0.089–0.414, with a mean of 0.317 ± 0.176. Moreover, a detailed comparison of between-plot similarities in species composition and dominance structure among geographic or environmental units (merged seasonal records, Table 4 ) revealed a high degree of homogeneity: all samples exhibited about 80% overlap in species composition and 50–60% in dominance structure, without any marked differences between particular local (triplet-related) and subregional (landsystem-related) units. Table 4 The similarity in species composition (measured by the Jaccard index) and in dominance structure (Renkonnen index) calculated for within- and between-community data representing particular spatial units or those assembled by habitat type, i.e. access to surface water (C, P, S). species composition dominance structure Jaccard Renkonen avg min max SD avg min max SD within units all 0.802 0.438 1.000 0.098 0.544 0.180 0.876 0.128 Crests 0.815 0.500 1.000 0.100 0.526 0.180 0.847 0.144 Perennial 0.810 0.474 1.000 0.093 0.566 0.308 0.824 0.122 Seasonal 0.784 0.688 0.882 0.071 0.547 0.230 0.836 0.109 South 0.836 0.625 1.000 0.076 0.558 0.284 0.868 0.119 North 0.800 0.467 1.000 0.109 0.551 0.206 0.876 0.143 East 0.827 0.467 1.000 0.071 0.554 0.215 0.868 0.132 West 0.820 0.533 1.000 0.099 0.582 0.314 0.876 0.123 SAT 0.859 0.722 1.000 0.064 0.593 0.284 0.868 0.113 SK 0.843 0.647 1.000 0.087 0.556 0.314 0.820 0.119 LET 0.799 0.467 1.000 0.113 0.520 0.215 0.846 0.156 PH 0.798 0.533 1.000 0.105 0.609 0.330 0.876 0.122 between units South/North 0.783 0.438 1.000 0.097 0.533 0.180 0.853 0.123 East/West 0.814 0.500 1.000 0.094 0.538 0.206 0.841 0.129 SAT / PH 0.754 0.474 0.944 0.098 0.572 0.309 0.781 0.103 SAT / LET 0.785 0.556 1.000 0.090 0.526 0.276 0.792 0.108 SAT / SK 0.823 0.625 1.000 0.071 0.543 0.292 0.841 0.118 SK / PH 0.780 0.438 0.941 0.097 0.552 0.273 0.853 0.129 SK / LET 0.810 0.533 1.000 0.095 0.484 0.180 0.763 0.131 LET / PH 0.802 0.500 1.000 0.110 0.535 0.206 0.847 0.138 Although the vegetation cover differed among the three habitat types and microgeographic units (triplets), it had no significant effect on the abundance of particular OTUs or on other community variables (Supplementary File S2.3). The habitat preference of all OTUs, reflecting association with the cover of herbs, shrubs and trees (Fig. 7 ), resulted in their compact aggregation around centroid values that were almost identical for both molossid and vespertilionid OTUs in the February and November period and for the sites with exceptionally high bat species Shannon H’ diversity. Effect of spatial scale. To examine the trade-off between community variables and spatial scaling, we compared diversity measures in subsets representing different spatial scales (Table 5 ). We found no essential between-scale differences in alpha diversity and generally low beta diversity, including beta turnover (beta T according to Wilson and Shmida 52 ). Except for the plot- and local scales, the beta diversity was negligible in contrast to non-spatially scaled clusters of plots grouped by habitat. This suggests that at each spatial scale, the whole set of community samples exhibited characteristics of a single entity. This picture is further illustrated by (i) a high degree of community nestedness, (ii) abundance relations among the community members demonstrated by a detrended correspondence analysis (Fig. 6 ), and (iii) the patterns of its phenetic packing (Fig. 7 ). The nestedness analyzed by using metrics based on overlaps and decreasing fills (NODF) and matrix temperatures (MT) yielded mean values of 81.4 and 15.5, respectively, with variation at different spatial scales from 75.1 to 85.7 (NODF) and 8.1 to 19.2 (MT) (Fig. 8 , Supplementary File S4). It conforms to a very high degree of incipient integrity of the community structure over the whole studied area. The contribution of particular OTUs to the nestedness pattern was generally scaled by their abundances. Two notable exceptions are the peak role of Scotophilus dinganii in establishing the nestedness pattern and the lowest degree of nestedness contribution by Afronycteris nana . The central position of SCODIN ( Scotophilus dingani) , contrasting with the marginal position of NEONAN ( Afronycteris nana ), resembling their positions in the phenotype setting of the community (Fig. 6 ) is here particularly worth mentioning. Table 5 Comparison of community characteristics: sp – species richness; N – species abundances; H’ - Shannon alpha diversity; Simpson 1-D alpha diversity; beta diversity W 68 , beta T 52 in sample sets representing different spatial scales: point, local (triplets), landsystems, subregional and regional (with mean grid size in km 2 ) and those assembled by habitat type, i.e. access to surface water (C, P, S). Scale: point local (triplets) landsystem subregional subregional regional C P S n = 60 n = 20 n = 4 n = 2: N-S n = 2: E-W n = 1: all n = 20 n = 20 n = 20 km2 0.001 100 4,000 10,000 10,000 20,000 sp avg 14.9 16.7 18.3 18.5 19.0 19 14.3 15.5 15.7 min 10 15 17 18 19 10 12 13 max 19 19 19 19 19 19 18 18 N avg 1,886.5 5,659.5 28,297.5 56,595.0 56,595.0 113,190.0 1,025.5 3,030.1 2,042.0 min 122 1,181 21,727 54,218 43,532 255 1,101 299 max 8,563 13,954 37,245 58,972 69,658 3,094 8,563 5,789 H* avg 2.009 2.147 2.267 2.312 2.316 2.332 1.917 1.967 2.089 min 1.252 1.603 1.976 2.175 2.241 1.253 1.372 1.455 max 2.501 2.427 2.435 2.440 2.391 2.277 2.420 2.501 Simpson 1-D avg 0.816 0.840 0.862 0.870 0.869 0.874 0.792 0.803 0.835 min 0.602 0.711 0.805 0.847 0.855 0.631 0.602 0.692 max 0.908 0.897 0.890 0.892 0.883 0.881 0.891 0.908 Beta W 0.274 0.138 0.041 0.027 0.000 0.326 0.228 0.208 Beta T 3.291 2.048 0.781 0.514 0.000 3.844 3.030 2.898 Discussion Due to the methodical restrictions of acoustic recordings, we collected relevant data only for two of the four guilds of local bat communities, i.e., aerial foragers (Molossidae, Emballonuridae) and clutter-edge forages (Vespertilionidae). For obvious reasons, fruit bats are not represented in our sample, and records of clutter foragers (Rhinolophidae, Hipposideridae) and ground gleaners (Nycteridae) were restricted to accidental recordings and thus omitted from most comparisons. Nevertheless, as also live captures suggest 30 , 53 , the essential contribution of the former two guilds to bat communities of the region, both in terms of their species richness and abundance, is a real phenomenon. Our results can thus be assumed to provide relevant information on the bat community structure and the patterns of its variation in KNP. Yet, the applied method is incipiently associated with putative uncertainties in the actual taxonomic identity of particular acoustic parataxa 54 . Although the suggested taxonomical identity was robustly supported in most OTUs included in our study (Supplementary File S1), two items are worth discussing. One is MOLOSSID45 – the calls resembling molossid bats by a flat frequency sweep (which alternatively could be interpreted as atypical search flight calls of an unidentified vespertilionid). The other case is more noteworthy, as it concerns one of the core OTUs: EPTHOT. Here, the post-hoc comparisons reveal several discrepancies: (i) Based on capturing records, Eptesicus hottentotus appears to be a rare bat with quite a sparse occurrence in the region 30 , 53 . (ii) The sonographic patterns conforming to the search echolocation calls of E. hottentotus (see Supplementary File SF1) might appear at the approach stage of call sequences of some molossid bats, first of all Mops condylurus , under conditions of a high prey concentration. (iii) There is obvious discrepancy between a large number of roost records of M. condylurus and relatively low abundances of MOPCON in the acoustic record (the low-frequency search calls of M. condylurus ; see Supplementary File S1). Thus, in regard to the fact that based on phonologic traits, it seems impossible to distinguish high-frequency FM calls of molossid from those of Eptesicus hottentotus , it appears reasonable to consider most EPTHOT records as belonging to M. condylurus . The post-hoc comparisons of alternative taxonomic interpretations of EPTHOT (Figs. 4 – 6 ) seem to support it quite convincingly. Nevertheless, despite the above-mentioned uncertainties on the real taxonomic identity of some OTUs, the standardized identification procedure of particular acoustic parataxa ensured formal consistency of input community records, which is a key prerequisite of further comparative analyses. Moreover, thanks to the identical identification procedure applied in both studies, the community characteristics that we revealed can be directly compared to the acoustic monitoring undertaken in 2017 and 2018 at 26 sites in the northernmost part of KNP and neighboring regions of Limpopo province 30 (see Supplementary File S5 for details). Both datasets show a great agreement not only in species composition, which is identical, but also in the overall dominance of particular OTUs (Wilkinson pair test, p = 0.756). Notwithstanding particular differences (discussed in Supplementary File S5), the correspondence between both datasets convincingly supports one of the most exciting results of our study: a surprisingly high degree of homogeneity in bat community structure, extending beyond the spatial margins of the area under study. Even more surprising is that just quite a few of the numerous environmental and vegetation-related predictors we examined exhibited certain effects on abundance, diversity, and variation in community structure. Those whose effects were significant, i.e., access to water (obviously related to increased concentration of insects and higher plant diversity along rivers 7 ), the distance to the nearest camp (a source of multiple roosting opportunities), and overall similarity among neighboring plots within a triplet, all refer to trivial relationships with explicit meaning – local food and roost availability and consistency of resident occurrence at sublocal scales. Moreover, the homogeneity in species composition and dominance structure, as well as the lack of large-scale effects of environmental or vegetation predictors (notwithstanding those explaining microgeographic variations), appeared consistently at all spatial scales, as did the balanced contribution of two major foraging guilds, both with a nearly equal number of species and overall abundances. Within the whole set of recorded OTUs, we found no distinct between-species differences in their habitat preferences – all taxon-specific mean values were densely clustered along the centroid identical with the centroid of habitat variation of all sites under study, i.e., the mean state of habitat conditions available in the study area. All these unexpected observations point unequivocally to the picture of a single community exhibiting at large the patterns of steady coexistence dynamics. The drivers supporting the coexistence dynamics as identified by modern coexistence theory 55 – 59 can be subdivided into two classes: extrinsic factors buffering effects of interspecific competition, and intrinsic factors stabilizing coexistence disposition of particular community members. Among the former class, the dynamics of environmental variation affecting temporarily the advances of individual competitors is to be mentioned first. The intermediate disturbance hypothesis 58 , 60 predicts that local species diversity is maximized when ecological disturbances are sufficient to affect particular community members yet not too strong to force their extinctions, and, at the same time, they are neither too rare nor too frequent. Obviously, the dynamics of the savanna biome, with a mosaic of fire events and habitat structure resulting in fluctuations in the impact of large herbivores 8 , fit the picture of steady intermediate disturbance dynamics to the full extent. Among the intrinsic factors, stabilizing niche differences and relative fitness differences 59 or fitness inequalities 55 , 56 are essential. The niche differences stabilized via resource partitioning increase the role of intraspecific over interspecific competition and accelerate adaptive rearrangements scaled by capacities of locally available resources. In the case of bats, the data required for detailed analyses of resource partitioning are, for obvious reasons, mostly not available. Yet, in some cases 61 , they can be adequately estimated with indirect evidence, typically addressing the taxon-specific combination of the phenotype traits directly related to foraging dispositions (wing design, echolocation). The multivariate comparisons of these traits in all community members, the phenetic packing of a community 50 , can serve as reliable information, namely if supported by detailed data on individual co-occurring species 36 , 51 , 62 – 64 . Our analyses demonstrated a nearly equidistant distribution of individual species and a high degree of phenetic packing of the bat community studied, indicating a well-balanced resource partitioning with reduced effects on between-taxa competition. It is worth mentioning that the corresponding picture was also obtained by scaling the abundance of particular community members, the variable addressing the topic of fitness differences or fitness inequalities. The theory predicts a stable coexistence when stabilizing niche differences of species are greater than their relative fitness differences 59 . Also, this condition seems to be met in bats. All are strongly pronounced K-strategists with nearly equal population growth rates notwithstanding their temporal variation due to different responses of particular species to diverse disturbance cues. Moreover, as the theory predicts 55 , the stabilizing niche difference promotes an increase in low-density growth rate, another factor contributing to coexistence dynamics. In spatial regards, the savanna community studied was characterized by a greatly pronounced nestedness pattern retained at all temporal and spatial scales. Previously, the nestedness of spatial organization was demonstrated to be an independent factor that effectively reduces interspecific competition and enhances the number of coexisting species 65 . To summarize, our results suggest that the bat community of the savanna biome is to be considered a distinct entity integrated by multiple mechanisms promoting the coexistence of all community members. The stabilized niche differences among them arose, besides incipient ecological specificities of particular clades, from ongoing local interactions under shared common conditions provided by the savanna biome throughout its long history (including steady impacts of herbivores and fire maintaining the habitat mosaic 27 ). As a result, all the community members prefer essentially just the mean conditions that the dynamics of the savanna biome provide. The differences in their spatial dispersal seem to be restricted to the fine variation manifested at the microgeographic scale (within- and between-triplet differences). At the local, subregional, and regional scales, respectively, the variation in KNP bat community takes a form of a homogeneous densely patched mosaic that stays beyond the direct control of any large-scale environmental predictors. In these regards, the bat communities illustrate distinctive features of the savanna biome, the unique element of global biodiversity worth thorough conservation interest. Material and methods Collecting field data. The data collected within the MOSAIK project (Fig. 1 ) come from 60 fixed point sites 50 × 50 m in size (plots) representing four subregional systems (landsystems – LET: Letaba, PHA: Phalaborwa, SAT: Satara, SKU: Skukuza), each with five local triplets of neighboring plots (6–8 km from each other) representing distinct habitats differing in water availability (P – perennial river with water available all year round, S – seasonal river with water available in rainy season, C – dry crest with no access to water). Correspondingly, the formal designation of individual sites, followed in this paper, consists of the abbreviations of the territorial unit (SKU, SAT, LET, PHA), the triplet number (1–5), and the type of site (S, P, C). All sites were controlled twice during 2018–2020, in spring, representing the rainy season (November, December – XI), and late summer, representing the onset of the dry season (February, March – II). For more detailed explanations of the MOSAIK project settings, see Hejda et al. 7 . Bats were recorded by using an automated recorder Song Meter SM4BAT FS (Wildlife Acoustics) with ultrasonic microphone SMM-U2 (Wildlife Acoustics) placed (typically in a centre of the plot) by the stick at a height of 4 m (to reduce noise from stridulation of ground insects). Recording started automatically at sunset and ended at sunrise. Each recording was accompanied by site identification, date, and time automatically provided by the device. The sample rate was set to 384 kHz, and the recording sensitivity to 10 kHz to minimize noise. This limit was below the minimum characteristic frequency of the lowest echolocating species in the region, Otomops martiensseni 66 . The recordings were made in WAV format and stored on SD memory cards. The recordings in the plots of the same triplet always took place during one night. Within the plot, the devices were placed in open places near bushes or trees, depending on the condition of the site, but at a sufficient distance from the canopy to prevent the microphone from being covered by leaves. The present paper is based on acoustic records obtained from the standard plots of the MOSAIK project. The dataset contains 94,310 valid recordings representing 130,888 individual bat records: 116,623 sequences of echolocation calls and 14,261 social calls. Here, it should be emphasized that the community record was biased by essential restrictions of the applied method. It provided no information on non-echolocating bats (Pteropodidae), and ground gleaners with weak echolocation (Nycteridae) and, correspondingly, just scarce records on the foliage gleaners detectable at short distances only (Hipposideridae, Rhinolophidae, Rhinonycteridae). Hence, the representation of these groups in our sample was considered rather incidental and not suitable for quantitative comparisons. This data was excluded from analyses. In contrast, quite robust data were obtained for the remaining groups, i.e., Molossidae, Vespertilionidae, and Emballonuridae, which represent (both in terms of species richness and abundance) the true core of local bat communities and compose 97% of all records. The vast majority of our analyses are thus restricted to them. For each plot or its close surroundings, a series of variables were recorded. Besides retrieving the remote sensing data (NDVI, EVI, MODIS, etc. mostly referring to a 4 km 2 grid cell; see Hejda et al. 7 for details), the MOSAIK team performed an extensive on-ground recording of vegetation cover, plant species diversity, birds, large mammals, insects, and bats. Insects were collected by using night light traps; in the present paper, we used the total mass of insects collected per night as a rough estimate of food resource richness for bats. A complete list of all environmental variables and their abbreviations is provided bellow. Additional whole-night recordings, not used in analyses, were undertaken in campsites in Skukuza, Letaba, Mopane, Shingwendzi, Phalaborwa, and Punda Maria. A number of day-time bat roosts were repeatedly censused, mostly in campsites (including colonies of Hipposideros caffer , Nycteris grandis , Taphozous mauritianus , Afronycteris nana , Pipistrellus hesperidus , Pipistrellus rusticus , Scotophilus dinganii , Chaerephon pumilus , Mops condylurus , Epomophorus wahlbergi ). Acoustic analyses and OTUs. The post hoc voice identification and visualization of acoustic records were undertaken by using Kaleidoscope Pro software (Wildlife Acoustics). The identification procedure applied in this project was based on the cluster analysis method, which classifies a voice record using multivariate comparisons with previously recorded data and a manually edited database of echolocation parameters. We used a database of sonar variables compiled for bats of Kruger National Park (KNP-specific classifier) that was developed in the laboratory of P. Taylor 30 , further supplemented with own recordings of reliably identified bats either mist-netted or those leaving roosts in camps. Recordings of echolocation or social calls that contained at least two cries were considered valid; the invalid recordings (biased by noise from insects, wind, birds, or poor resolution) were excluded. Species determination was based essentially on the acoustic characteristics of the search phase of the echolocation sequence. Since the automated identification procedure (including the cluster method) does not allow the identification of multiple bat species on a single recording, recordings with multiple species were always manually edited. The cluster analysis method of the Kaleidoscope program splits the recordings with multiple call sequences into separate files, often representing a duplication of calls from a single individual only – then a given individual was counted only once. The distribution of records by cluster analysis was double-checked, and any misidentification was manually corrected. Using the above procedures, the echolocation voice records were categorized into 31 acoustic parataxa, taken as operation taxonomic units (OTUs) of subsequent analyses. Twenty-two of them were coidentified with real local species at a high probability level. Nine OTUs with uncertain species affiliations (mostly represented by scarce records only) were treated as separate entities under original working labels (Table 1 , and below). Detailed data concerning echolocation characteristics of particular OTUs and post-hoc estimates of possible identification bias are shown in Supplementary File S1. Statistical analysis: community patterns. The resulting dataset consisting of the records of particular OTUs in 120 plots representing basic spatiotemporal units (site, date, and season XI or II, respectively) was aggregated to partial subsets by topographic units, the northern area (N: LET and PHA) and the southern area (S: SKU and SAT), spring (IX) and autumn (II) seasons, and habitat (perennial, seasonal, crest). For each, we computed the basic statistics (mean, minimum, maximum, SD, CV (= SD/mean), skewness) and the standard community parameters: abundance (n, N), dominance (D), and frequency (F) of each OTUs, Shannon diversity index (H*) and Pielou evenness (E*), calculated as H’ = Ʃ P i *ln(P i ), where P i represents the relative abundance of species i in a given community, and E = H’/log(S), where S represents the number of species within a community 67 ). Between-sample comparisons were further supplemented by computation of Simpson 1-D alpha diversity (D = 1- Ʃ n i (n i -1)/N(N-1)); beta diversity W 68 , and beta T 52 . Further, we calculated the summary abundances of the families Vespertilionidae and Molossidae and their abundance ratios. The dominance rank of particular species was expressed in terms of Tischler’s scale 69 : eudominant (D > 10%), dominant (10% > D > 5%), subdominant (5% > D > 2%), recedent (2% > D > 1%), subrecedent (1% > D). Conceptual issues concerning core and satellite community elements 70 were taken into account. The individual samples and the above-mentioned subsets were mutually compared using a series of univariate, bivariate, and multivariate techniques. Null hypotheses were tested by paired Wilcoxon test, Mann-Whitney U/z test for mean values, F statistics for variance, and Kolmogorov-Smirnov test (D) for distribution (with Monte-Carlo permutation). Paired Wilcoxon tests and Mann-Whitney tests were used to test the differences in individual OTUs. Statistical significances of differences between compared subsets were expressed as respective probabilities (in most instances, for the sake of simplicity, the values of particular test criteria were omitted). The between-sample similarity of species composition was quantified by the Jaccard index (Jc = the ratio of the number of species present in both samples to the sum of the species represented in a single sample only), the Bray-Curtis index was used as a measure of dissimilarity in dominance structure (BC = 1–2c ij /(n i +n j ), where c is the sum of the smaller values of the abundance of species common to samples i and j and n are the total abundances of the samples). Similarities among individual samples, presence, abundance, dominance, and distributional pattern of individual species were further analyzed by a series of multivariate techniques. Regarding the log-normal distribution of abundance data, for most computations, we applied the logarithmic tranSupplementary FileSFormation of abundance data. The matrices of presence, abundance, and dominance records, correlation matrices of abundance data, and similarity matrices of the Jaccard index and Bray-Curtis index were further examined by cluster analyses, principal component analysis (PCA), and correspondence analyses (CA, DCA). Patterns of variation in community structure were further examined by using nestedness analysis followed by the concepts proposed by Atmar and Patterson 71 (for a recent review, see e.g. Payrató-Borràs et al. 72 ). Using open software module at http://ecosoft.alwaysdata.net , we computed NODF (nestedness metrics based on overlaps and decreasing fills 73 , MT (matrix temperature), and extracted graphical outlines of packed community matrix. The matrix temperature 71 expresses a degree of nestedness order in thermodynamic terms, i.e. within a continuum from a perfectly ordered system, absent of all randomness (maximally “cold” state: 0 0 ), to the absence of any order (maximally “hot” state:100 0 ). NODF (with an inverse scaling), is a measure responding the objections against the uncorrected use of some other nestedness measures, which is considered to be much less prone to effects of taxon and area specificities 73 . Statistical analysis: effects of environmental variables. We examined habitat preferences of particular species regarding vegetation cover (percentages of herb, shrub and tree cover) of the sites where the species was recorded. For each OTU, we computed a mean value of its habitat preferences in terms of ternary representation of herb, shrub, and tree cover (as a sum of site-specific values of vegetation cover multiplied by abundances of given OTU in respective sites, divided by its total abundance). The respective characteristic was expressed as a mean value of the vegetation cover (computed from a sum of particular site values multiplied by species abundance at these sites divided by the total abundance of the species). In the same way, we also computed the distribution of species records in particular site types (crest, seasonal, perennial river). Using a dataset of particular categorical variables and large-scale predictors of environmental, climatic, and vegetation conditions at each point (surveyed in detail by Hejda et al. 7 and briefly listed in the Methods), we tested their effects on bat species richness, abundance, Shannon diversity, Pielou evenness, and species composition. For that purpose, the predictors were assembled in three sets. The first set, further termed as "categorical predictors" included variables reflecting the basic habitat characteristics: habitat type (close to perennial rivers, close to seasonal rivers and on crests, at least 5 km from any waterbody), bedrock (granite, basalt) and landsystem (Skukuza, Satara, Phalaborwa, Lethaba), North versus South of KNP and presence/absence of Mopane as a dominant woody species. The second set of predictors, further termed as “vegetation parameters”, included basic characteristics of vegetation, such as herbal, grass, shrub and tree species richness and cover. Further, we included a share of typical “savanna” species with potential conservation value, separately for herbs, shrubs, and trees. To account for the gradients in plant composition, we calculated an indirect gradient ordination analysis (PCA) and then included the first three ordination axis (PCO1, PCO2 and PCO3 – see below for explanation) among the predictors representing the character of vegetation. The third set of predictors represented a broader environmental context and included variables such as distance to park boundaries, distances to roads, tracks and camps, distance to rivers and streams, climatic characteristics (temperature, precipitation), and fire history and frequency. See below for the list of all included predictors. The data were analyzed using both univariate and multivariate models. In particular, the data with the bat species richness, Shannon diversity, Pielou evenness, and total abundances as a response variable were analyzed using univariate methods, while data on the bat species composition were analyzed using multivariate direct gradient ordination models (RDA). We tested the effects of all categorical predictors, along with their interactions with “habitat type”. We applied regression tree models and indirect gradient ordination analysis as exploratory tools to select potentially important predictors of species richness, diversity, evenness and abundance among all continuous variables (vegetation parameters, environmental context). The categorical predictors were also included in the regression tree models to show possible interactions with continuous predictors. In regression tree models, the categorical variables, vegetation parameters and variables of environmental context were set as predictors, while bat species richness, abundance, diversity and evenness were set as response variables in individual regression tree models. The effects of predictors, identified as important by the regression tree models, were then tested by linear mixed-effect models. The regression tree models were created in the R software, using the package “rpart” 74 . Further, the relations between predictors (categorical variables, vegetation parameters, environmental context) and univariate response variables (bat species richness, abundance, diversity and evenness) were explored using indirect gradient ordination models, where all variables of interest were used as individual response variables in the multi-dimensional space. Ordination plots were used to visually estimate the relations between different variables and to identify predictors related to bat species richness, diversity, evenness and abundance. Similarly to the regression tree models, the effects of variables (categorical, vegetation, environmental context) that were identified as closely related to bat species richness, abundance, diversity and evenness were then tested by linear mixed effect models. The exploratory indirect gradient ordination analyses were performed using the CANOCO5 software 75 . In the mixed effect models, the predictors of interest (categorical, vegetation parameters, environmental context) were set as fixed effects, while the identity of triplets (representing the hierarchy of the sampling design) was set as a random effect. The normality of individual response variables (bat species richness, abundance, diversity and evenness) as well as that of the predictors (continuous variables among vegetation parameters and environmental context) was tested using the Shapiro-Wilk tests. Square-root, log and arcsine tranformations were applied to improve normality in case the Shapiro-Wilk tests revealed significant deviations from it. The significance of individual main effects and their interactions was estimated using the deletion tests, which compare the explanatory power of models with and without a particular term or interaction 76 . The significance of quadratic terms of all continuous predictors was also tested by deletion tests. The quality of the parsimonous model (= with only significant terms and interactions) was checked by testing the normality of residuals and also visually, by inspecting the normal-probability plots. The linear mixed-effect models were created in the R software, using the package "nlme" 77 . The differences between individual levels of all categorical predictors were tested by the Tukey HSD method, using the package “emmeans” of the R software. The data on the composition of bats were examined using direct gradient ordination methods (RDA and CCA), with the abundances of individual bat species set as response variables in the multivariate ordination space and the categorical variables, variables expressing the vegetation parameters and environmental context as predictors. The significance of predictors was tested using the Monte-Carlo permutation tests with 499 permutations. To account for the possible autocorrelations given by the hierarchical sampling desing, a split-plot permutation scheme was applied. The triplets were set as whole-plots, while individual sampling sites (plots) were set as split-plots. Both whole-plot and split-plot level were permuted freely, as some of the predictors (habitat: perennial rivers, seasonal rivers, crests; mopane cover etc.) were defined at the split-plot level, while others (bedrock, landsystem, north-south) were defined at the whole-plot level. First, complex ordination models with the whole group of predictors (categorical, vegetation paremeters, environmental context) were created to estimate the overal explanatory power of these three groups of predictors. Further, forward selection procedure was applied to identify the strongest predictors within each group. The results were visualized using ordination plots. The choice between a linear and a unimodal model (RDA or CCA, resp.) was decided based on the main gradient within the data. All direct gradient ordination analyses were performed in the CANOCO 5 software 75 . The between-season differences and the season related effects were not analysed in the present paper – these topics will be surveyed separately elsewhere. Database operations were performed in Microsoft Excel, the basic statistics were computed in STATISTICA 13, and/or in Past, and SAM software, LME and regression tree analyses were computed in Rstudio, the ordination analyses mostly in CANOCO. List of the environmental predictors used in the present study and Abbreviations We largerly used the dataset assembled by the MOSAIK project team and adopted completely all particular variables and their values. For detailed explanation of all particular variables (including their primary sources etc.) see Hejda et al. 7 . Categorial factors N, S : Northern vs. Southern part of the region, * Bedrock : granite vs. basalt, * Habitat type or plot /s ite category (C – crest with no direct acces per 4 km 2 grid cell ter, S – seasonal river, P – perennial river), * triplet : neighbouring plots within the same triplet vs. distant plots, * season : II (February) vs. XI (November) Large-scale predictors of environmental context (referred mostly to mean values within 4km 2 grid cell surrounding the site point) fireSum : Fire frequency, total number of fires from 2000 to 2019; * fireMean : Mean number of fires, Pixel-wise mean of the number of fires recorded over the long-term. * fireSD : StdDev of number of fires * eviSum : Sum EVI, Long-term sum of Enhanced Vegetation Index (EVI). * eviMean : Mean EVI, Long-term mean of Enhanced Vegetation Index (EVI). * eviSD : StdDev EVI, Long-term standard deviation of Enhanced Vegetation Index (EVI). * rainSum : Sum rainfall, Long-term sum of all rainfall. * rainMean : Mean rainfall, Long-term mean of all rainfall. * rainSD : StdDev of rainfall, Long-term standard deviation of all rainfall. * tempMean : Mean temperature, Long-term mean of temperature (°C). * tempSD : StdDev of temperature, Long-term standard deviation of temperature (°C). * tempMin : Minimum temperature, Long-term mean of minimum temperature (°C). * tempMax : Maximum temperature, Long-term mean of maximum temperature (°C). * waterSum : Sum surface water occurrence density, Long-term sum of surface water occurrence, expressed as kernel density estimates (KDE). Annual density of water extent, i.e. weighed by % re-occurrence of water each year; * WaterMean : Mean surface water occurrence density, Long-term mean of surface water occurrence, expressed as kernel density estimates (KDE). Annual density of water extent, i.e. weighed by % re-occurrence of water each year; * waterSD : StdDev Long-term standard deviation surface water occurrence, expressed as kernel density estimates (KDE). Annual density of water extent, i.e. weighed by % re-occurrence of water each year; * distBnd : Distance to KNP boundary. * distCamp : Distance to all campsite. * distTar : Distance to all tarred roads. * distDirt : Distance to all gravel roads. * distRiv : Distance to all major rivers. * distStrm : Distance to all rivers and streams. * Latitude : Latitudinal GPS coordinates of the site * Longitude : Longitudinal GPS coordinates of the site. * CampDist : Distance to the nearest campsite Large scale predictors of vegetation patterns (based on ground analyses by the MOSAIK team) vegPCO1, vegPCO2, veg PCO3 - loading values of first three components of multivariate factor analysis of all vegetation variables: PCO1 - poor, open-ground vegetation, mostly on granite substrate, PCO2 - shaded grounds with deep soil, high productivity, "sweet veld", PCO3 – temporarly disturbed, largely mesic vegetation, related to seasonal rivers, * vegCoverHerbs, *vegCoverShrub *vegCoverTrees *vegCoverGrass - cover of particular vegetation guilds, vegSPtotal - total species richness at the site, *vegSPHerbs, *vegSPShrub *vegSP Trees - species richness of particular vegetation guilds, vegH* - Shannon diversity of vegetation at the site, vegHherbs *vegHshrubs *vegHtrees - Shannon diversity of particular vegetation guilds, *vegAlien - number of alien plant species, *savHerb *savShrub *savTree - number of typical savanna elements, * Mopane - cover of mopane (Colophospermum mopane). List of OTUs CHAANS – Chaerephon ansrogei * CHAPUM – Chaerephon pumilus * EPTHOT – Eptesicus hottentotus (but see Discussion and Supplementary File S1) * MINNAT – Miniopterus natalensis * Molossid45 – parataxon not afiliated to any real species, with echolocation characteristics resembling those in molosid, yet with fpeak 45 kHz * MOPCON – Mops condylurus * MOPMID – Mops midas * MyotisSp – parataxon not afiliated to any real species, with echolocation characteristics of the genus Myotis * MYOTRI – Myotis tricolor * NEOCAP – Laephotis capensis * NEONAN – Afronycteris nana * NYCSCH – Nycticeinops schlieffeni * OTOMAR – Otomops martiensseni * PIPHES – Pipistrellus hesperidus * PIPRUS – Pipistrellus rusticus * SCODIN – Scotophilus dinganii * TADAEG – Tadarida aegyptiaca * TAPMAU – Taphozous mauritianus * RHICAP – Rhinolophus darlingi * RHI105 – Rhinolophus landeri * RHIFUM – Rhinolophus fumigatus * RHISIM – Rhinolophus simulator * RHIRHO – Rhinolphus swinnyi * HIPCAF (RHICAF) – Hipposideros caffer * UNK75 (75) – parataxon not afiliated to any real species, with fpeak around 75 kHz. * 37 – parataxon not affiliated to any real species, with fpeak around 37 kHz. * 35 – parataxon not affiliated to any real species, with fpeak around 35 kHz. * species1 – parataxon not affiliated to any real species, with fpeak around 20 kHz. * NoID – unidentified echolocation signals Abbreviations avg average * BC Bray-Curtis measure of dissimilarity in dominance structure * C, S, P sites : crest, seasonal river, perennial river * CF - constant frequency * CV coffeficient of variation (=SD / avg) * D - dominance (D=100(n/N)* E* evenness * FM -frequency modulated * GLM generalized linear model * GLZ generalized parametrized linear models * H* Shannon diversity index * Jc Jaccard index * KNP - Kruger National Park * LET Letaba landsystem * n - abundance (number of species records) *N - abundance (total number of all records) * p probability (p<0.05 taken as significant) * PH Phalaborwa landsystem * qCF - quasi-constant frequency * r Pearson’s coefficient of correlation * R2 squared coefficient of correlation * SAT Satara landsystem * SD statistical deviation * SK Skukuza landsystem Declarations Acknowledgements The MOSAIK project was supported by grant no. 18-18495S (Czech Science Foundation), long-term research development project RVO 67985939 (Czech Academy of Sciences), and internal project no. UNCE204069 (Charles University). The project was registered as PYSK1432 with SANParks. Support from SANParks during the field visits was greatly appreciated. We thank our guards Obert Mathebula, Thomas Rikombe, Desmond Mabaso, Herman Ntimane, Annoit Mashele, Isaac Sedibe, Priska Rikombe, and Velly Ndlovu for keeping us safe in the field. Authors contributions IH, PP, KP, DS, LCF and SMF conceived the idea, MS, IH and SD collected the bat data, ERB, DMP, PJT, SMW provided a comparative database of bat echolocation calls, SD, JČ, MH, KP, PP, and RT collected the data on the environmental variables and other contextual information, MS, IH and MH analyzed the data, IH and MS wrote the first draft of the paper, and all authors commented on the manuscript and gave final approval for its publication. Competing interests The authors declare no competing interests. Data Availability Statement The data are available in a public depository: https://github.com/IvanHoracek/KNP.git. References Sankaran, M. et al. Determinants of woody cover in African savannas. Nature 438 , 846–849 (2005). Staver, A. C., Archibald, S. & Levin, S. A. The global extent and determinants of savanna and forest as alternative biome states. Science 334 , 230–232 (2011). Mucina, L. & Rutherford, M. C. The Vegetation of South Africa, Lesotho and Swaziland (South African National Biodiversity Institute, Pretoria, 2006). Devine, A. P., McDonald, R. A., Quaife, T. & Maclean, I. Determinants of woody encroachment and cover in African savannas. Oecologia 183 , 939–951 (2017). Pyšek, P. et al. Into the great wide open: Do alien plants spread from rivers to dry savanna in the Kruger National Park? NeoBiota 60 , 61–77 (2020). Delabye, S. et al. Thirteen moth species (Lepidoptera, Erebidae, Noctuidae) newly recorded in South Africa, with comments on their distribution. Biodiv. Data J. 10 , e89729 (2022). Hejda, M. et al. Water availability, bedrock, disturbance by herbivores, and climate determine plant diversity in South-African savanna. Sci. Rep. 12 , 1–19 (2022). Čuda, J. et al. Habitat modifies the relationship between grass and herbivore species richness in a South African savanna. Ecol. Evol. 14 , e11167 (2024). Du Toit, J. T. Large herbivores and savanna heterogeneity. In The Kruger Experience: Ecology and Management of Savanna Heterogeneity (eds du Toit, J. T., Rogers, K. H. & Biggs, H. C.) 292–309 (Island Press, Washington, 2003). Bucini, G., Saatchi, S., Hanan, N., Boone, R. B. & Smit, I. Woody cover and heterogeneity in the savannas of the Kruger National Park, South Africa. IEEE 4 , 334–337 (2009). Smit, I. P. & Prins, H. H. Predicting the effects of woody encroachment on mammal communities, grazing biomass and fire frequency in African savannas. PLoS ONE 10 , e0137857 (2015). Loggins, A. A., Shrader, A. M., Monadjem, A. & McCleery, R. A. Shrub cover homogenizes small mammals’ activity and perceived predation risk. Sci. Rep. 9 , 1–11 (2019). Wessels, K. J. et al. Relationship between herbaceous biomass and 1km 2 advanced very high resolution radiometer (AVHRR) NDVI in Kruger National Park, South Africa. Int. J. Remote Sens. 27 , 951–973 (2006). Urban, M. et al. Surface moisture and vegetation cover analysis for drought monitoring in the southern Kruger National Park using Sentinel-1, Sentinel-2, and Landsat-8. Remote Sens. 10 , 1482 (2018). MacFadyen, S., Hui, C., Verburg, P. H. & Van Teeffelen, A. J. Quantifying spatiotemporal drivers of environmental heterogeneity in Kruger National Park, South Africa. Landsc. Ecol. 31 , 2013–2029 (2016). MacFadyen, S. Linking long-term patterns of landscape heterogeneity to changing ecosystem processes in the Kruger National Park, South Africa (Doctoral dissertation, Stellenbosch University) http://hdl.handle.net/10019.1/105154 (2018). MacFadyen, S., Hui, C., Verburg, P. H. & Van Teeffelen, A. J. Spatiotemporal distribution dynamics of elephants in response to density, rainfall, rivers and fire in Kruger National Park, South Africa. Diversity Distrib. 25 , 880–894 (2019). Malherbe, J., Smit, I. P., Wessels, K. J. & Beukes, P. J. Recent droughts in the Kruger National Park as reflected in the extreme climate index. Afr. J. Range For. Sci. 37 , 1–17 (2020). Cumming, D. H. et al. Elephants, woodlands and biodiversity in southern Africa. S. Afr. J. Sci . 93 , 231–236 (1997). Fenton, M. B. et al. Bats and the loss tree canopy in African Woodlands. Conserv. Biol. 12 , 399–407 (1998). Wilkinson, D. M., Midgley, J. J. & Cunningham, A. B. Constraints, crashes and conservation: Were historical African savanna elephants Loxodonta africana densities relatively high or lower than those seen in protected areas today? Plant Ecol. Divers. 15 , 1–2, 1–11 (2022). Rautenbach, I. L., Schlitter, D. A. & Braack, L. E. O. New distributional records of bats for the Republic of South Africa, with special reference to the Kruger National Park. Koedoe 27 , 131–135 (1984). Rautenbach, I. L., Fenton, M. B. & Braack, L. E. O. First records of five species of insectivorous bats from the Kruger National Park. Koedoe 28 , 73–80 (1985). Van der Merwe, M., Van der Merwe, N. J. & Penzhorn, B. L. Aspects of reproduction in the seasonally breeding African yellow bat, Scotophilus dinganii (A. Smith, 1833). Afr. Zool . 41 , 67–74 (2006). Rautenbach, I. L., Whiting, M. J. & Fenton, M. B. Bats in riverine forests and woodlands: A latitudinal transect in southern Africa. Can. J. Zool. 74 , 312–322 (1996). Taylor, P. J., Schoeman, M. C. & Monadjem, A. Diversity of bats in the Soutpansberg and Blouberg Mountains of northern South Africa: Complementarity of acoustic and non-acoustic survey methods. S. Afr. J. Wildl. Res . 43 , 12–26 (2013). Taylor, P. J., Nelufule, M., Parker, D. M., Toussaint, D. C. & Weier, S. M. The Limpopo River exerts a powerful but spatially limited effect on bat communities in a semiarid region of South Africa. Acta Chiropterol. 22 , 75–86 (2020). Monadjem, A., Shapiro, J. T., Mtsetfwa, F., Reside, A. E. & McCleery, R. A. Acoustic call library and detection distances for bats of Swaziland. Acta Chiropterol. 19 , 175–187 (2017). Monadjem, A. et al. Cryptic diversity in the genus Miniopterus with the description of a new species from southern Africa. Acta Chiropterol. 22 , 1–19 (2020). Brinkley, E. R., Weier, S. M., Parker, D. M. & Taylor, P. J. Three decades later in the northern Kruger National Park: Multiple acoustic and capture surveys may underestimate the true local richness of bats based on historical collections. Hystrix It. J. Mamm . 32 , 109–117 (2021). Schoeman, M. C., Cotterill, F. P. D., Taylor, P. J. & Monadjem, A. Using potential distributions to explore environmental correlates of bat species richness in southern Africa: Effects of model selection and taxonomy. Curr. Zool. 59 , 279–293 (2013). Cooper-Bohannon, R. et al. Predicting bat distributions and diversity hotspots in southern Africa. Hystrix It. J. Mamm. 27 , 1–11 (2016). Herkt, K. M. B., Barnikel, G., Skidmore, A. K. & Fahr, J. A high-resolution model of bat diversity and endemism for continental Africa. Ecol. Modell . 320 , 9–28 (2016). Smith, A. et al. Synergistic effects of climate and land-use change on representation of African bats in priority conservation areas. Ecol. Indic. 69 , 276–283 (2016). Monadjem, A., Conenna, I., Taylor, P. J. & Schoeman, M. C. Species richness patterns and functional traits of the bat fauna of arid southern Africa. Hystrix It. J. Mamm . 29 , 19–24 (2018). Schoeman, M. C. & Monadjem, A. Community structure of bats in the savannas of southern Africa: Influence of scale and human land use. Hystrix It. J. Mamm . 29 , 3–10 (2018). Medellin, R. A. & Redford, K. H. The role of mammals in neotropical forest-savanna boundaries. In Nature and Dynamics of Forest-Savanna Boundaries (eds Furley, P. A., Proctor, J. & Ratter, J. A.) 519–548 (Chapman and Hall, London, 1992). Aguirre, L. F. Structure of a Neotropical savanna bat community. J. Mammal. 83 , 775–784 (2002). Aguirre, L. F., Lens, L., Van Damme, R. & Matthysen, E. Consistency and variation in the bat assemblages inhabiting two forest islands within a Neotropical savanna in Bolivia. J. Trop. Ecol . 19 , 367–374 (2003). Bernard, E. & Fenton, M. B. Bats in a fragmented landscape: Species composition, diversity and habitat interactions in savannas of Santarém, Central Amazonia, Brazil. Biol. Conserv. 134 , 332–343 (2007). Larrea-Alcázar, D. M. et al. Spatial patterns of biological diversity in a neotropical lowland savanna of northeastern Bolivia. Biodivers. Conserv. 20 , 1167–1182 (2011). de Oliveira, H. F., de Camargo, N. F., Gager, Y. & Aguiar, L. M. The response of bats (Mammalia: Chiroptera) to habitat modification in a Neotropical Savannah. Tropic. Conserv. Sci . 10 , 1–14 (2017). Lima, C. S., Varzinczak, L. H. & Passos, F. C. Richness, diversity and abundance of bats from a savanna landscape in central Brazil. Mammalia 81 , 33–40 (2017). Morales-Martínez, D. M., Rodríguez-Posada, M. E., Fernández-Rodríguez, C., Calderón-Capote, M. C. & Gutiérrez-Sanabria, D. R. Spatial variation of bat diversity between three floodplain-savanna ecosystems of the Colombian Llanos. Therya 9 , 41–52 (2018). Gelderblom, C., Bronner, G., Lombard, A. & Taylor, P. J. Patterns of distribution and current protection status of the Carnivora, Chiroptera and Insectivora in South Africa. S. Afr. J. Zool. 30 , 103–114 (1995). Monadjem, A. & Reside, A. The influence of riparian vegetation on the distribution and abundance of bats in an African savanna. Acta Chiropterol . 10 , 339–348 (2008). Schoeman, M. C. & Jacobs, D. S. The relative influence of competition and prey defenses on the phenotypic structure of insectivorous bat ensembles in southern Africa. PLoS ONE 3 , e3715 (2008). Schoeman, M. C. & Waddington, K. J. Do deterministic processes influence the phenotypic patterns of animalivorous bat ensembles at urban rivers? Afr. Zool . 46 , 288–301 (2011). Arita, H. T. & Fenton, M. B. Flight and echolocation in the ecology and evolution of bats. Trends Ecol. Evol . 12 , 53–58 (1997). Findley, J. S. Phenetic packing as a measure of faunal diversity. Am. Nat. 107 , 580–584 (1973). Patterson, B. D., Willig, M. R. & Stevens, R. D. Trophic strategies, niche partitioning, and patterns of ecological organization. In Bat Ecology (eds Kunz, T. H. & Fenton, M. B.) 536–557 (University Chicago Press, 2003). Wilson, M. V. & Shmida, A. Measuring beta-diversity with presence absence data. J. Ecol. 72 , 1055–1064 (1984). Monadjem, A., Taylor, P. J. & Schoeman, M. C. Bats of Southern and Central Africa: A Biogeographic and Taxonomic Synthesis (Wits University Press, 2020). Russo, D. & Voigt, C. C. The use of automated identification of bat echolocation calls in acoustic monitoring: A cautionary note for a sound analysis. Ecol. Indic. 66 , 598–602 (2016). Chesson, P. General theory of competitive coexistence in spatially-varying environments. Theor. Popul. Biol. 58 , 211–237 (2000a). Chesson, P. Mechanisms of maintenance of species diversity. Annu. Rev. Ecol. Evol. Syst. 31 , 343–366 (2000b). Kneitel, J. M. & Chase, J. M. Trade‐offs in community ecology: Linking spatial scales and species coexistence. Ecol. Lett. 7 , 69–80 (2004). Kricher, J. C. Tropical Ecology (Princeton University Press, 2011). HilleRisLambers, J., Adler, P. B., Harpole, W. S., Levine, J. M. & Mayfield, M. M. Rethinking community assembly through the lens of coexistence theory. Annu. Rev. Ecol. Evol. Syst . 43 , 227–248 (2012). Wilkinson, D. M. The disturbing history of intermediate disturbance. Oikos 84 , 145–147 (1999). Fenton, M. B. & Rautenbach, I. L. A comparison of the roosting and foraging behaviour of three species of African insectivorous bats (Rhinolophidae, Vespertilionidae, and Molossidae). Can. J. Zool . 64 , 2860–2867 (1986). Findley, J. S. & Black, H. Morphological and dietary structuring of a Zambian insectivorous bat community. Ecology 64 , 625–630 (1983). Taylor, P. J., Monadjem, A. & Nicolaas Steyn, J. Seasonal patterns of habitat use by insectivorous bats in a subtropical African agro-ecosystem dominated by macadamia orchards. Afr. J. Ecol . 51 , 552–561 (2013). Monadjem, A. et al. Morphology and stable isotope analysis demonstrate different structuring of bat communities in rainforest and savannah habitats. R. Soc. Open Sci . 5 , 180849 (2016). Bastolla, U. et al. The architecture of mutualistic networks minimizes competition and increases biodiversity. Nature 458 , 1018–1020 (2009). Adams, R. A., Bonaccorso, F. J. & Winkelmann, J. R. Revised distribution for Otomops martiensseni (Chiroptera: Molossidae) in southern Africa. Glob. Ecol. Conserv . 3 , 707–714 (2015). Magurran, A. E. Measuring Biological Diversity (Blackwell, Oxford, 2004). Whittaker, R. H. Vegetation of the Siskiyou Mountains, Oregon and California. Ecol. Monogr . 30 , 279–338 (1960). Tischler, W. Einführung in die Ökologie . 4th Ed. (Gustav Fischer, Stuttgart, 1993). Hanski, I. & Gyllenberg, M. Two general metapopulation models and the core-satellite species hypothesis. Am. Nat . 142 , 17–41 (1993). Atmar, W. & Patterson, B. D. The measure of order and disorder in the distribution of species in fragmented habitat. Oecologia 96 , 373–382 (1993). Payrató-Borràs, C., Hernández, L. & Moreno, Y. Measuring nestedness: A comparative study of the performance of different metrics. Ecol. Evol . 10 , 11906–11921 (2020). Almeida-Neto, M., Guimaraes, P., Guimaraes Jr, P. R., Loyola, R. D. & Ulrich, W. A consistent metric for nestedness analysis in ecological systems: Reconciling concept and measurement. Oikos 117 , 1227–1239 (2008). R Development Core Team. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, Vienna, 2013). Šmilauer, P. & Lepš, J. Multivariate Analysis of Ecological Data Using CANOCO 5 (Cambridge University Press, 2014). Crawley, M. J. The R Book (John Wiley & Sons, Chichester, 2007). Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D. R. & R Core Team. Nlme: Linear and Nonlinear Mixed Effects Models . R package Version 3.1-152. https://CRAN.R-project.org/package=nlme (2021). Aldridge, H. D. J. N. & Rautenbach, I. L. Morphology, echolocation and resource partitioning in insectivorous bats. J. Anim. Ecol . 56 , 763–778 (1987). Wilson, D. E. & Mittermeier, R. A. Handbook of the Mammals of the World. Vol. 9. Bats (Lynx Edicions, Barcelona, 2019). Waghiiwimbom, M. D., Eric-Moise, B. F., Jules, A. P., Aimé, T. K. J. & Tamesse, J. L. Diversity and community structure of bats (Chiroptera) in the centre region of Cameroon. Afr. J. Ecol . 58 , 211–226 (2020). Table 1 and 3 Table 1 and 3 are available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files Table1and3.docx knpSupplementaryInformationSF15OK.pdf Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 16 Apr, 2025 Reviews received at journal 05 Apr, 2025 Reviewers agreed at journal 27 Mar, 2025 Reviews received at journal 25 Jan, 2025 Reviewers agreed at journal 02 Jan, 2025 Reviewers agreed at journal 01 Jan, 2025 Reviewers invited by journal 03 Dec, 2024 Editor assigned by journal 02 Dec, 2024 Editor invited by journal 07 Nov, 2024 Submission checks completed at journal 04 Nov, 2024 First submitted to journal 19 Oct, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5295291","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":377579056,"identity":"2d0f34cc-53ea-4d50-9734-c0d13013c368","order_by":0,"name":"Markéta Staňková","email":"","orcid":"","institution":"Charles University","correspondingAuthor":false,"prefix":"","firstName":"Markéta","middleName":"","lastName":"Staňková","suffix":""},{"id":377579057,"identity":"6bb82a97-246d-425b-801e-8cf52c146923","order_by":1,"name":"Martin Hejda","email":"","orcid":"","institution":"Institute of Botany","correspondingAuthor":false,"prefix":"","firstName":"Martin","middleName":"","lastName":"Hejda","suffix":""},{"id":377579058,"identity":"1f3c6dcb-db07-45ee-81a1-33eacafcaeb7","order_by":2,"name":"Erin R. Brinkley","email":"","orcid":"","institution":"Rhodes University","correspondingAuthor":false,"prefix":"","firstName":"Erin","middleName":"R.","lastName":"Brinkley","suffix":""},{"id":377579059,"identity":"8051fdc3-e772-43a7-bf0b-a518f9ad4110","order_by":3,"name":"Jan Čuda","email":"","orcid":"","institution":"Institute of Botany","correspondingAuthor":false,"prefix":"","firstName":"Jan","middleName":"","lastName":"Čuda","suffix":""},{"id":377579060,"identity":"d3c3f014-0e03-48b8-bb71-b0ccbac51ee7","order_by":4,"name":"Sylvain Delabye","email":"","orcid":"","institution":"Charles University","correspondingAuthor":false,"prefix":"","firstName":"Sylvain","middleName":"","lastName":"Delabye","suffix":""},{"id":377579061,"identity":"ab518c8a-478d-41cd-b9fe-6b8c5d3e11d8","order_by":5,"name":"Llewellyn C. Foxcroft","email":"","orcid":"","institution":"South African National Parks","correspondingAuthor":false,"prefix":"","firstName":"Llewellyn","middleName":"C.","lastName":"Foxcroft","suffix":""},{"id":377579062,"identity":"ebe3c5fb-a7db-4fbe-bea0-7969feaee5fc","order_by":6,"name":"Sandra MacFadyen","email":"","orcid":"","institution":"Stellenbosch University","correspondingAuthor":false,"prefix":"","firstName":"Sandra","middleName":"","lastName":"MacFadyen","suffix":""},{"id":377579063,"identity":"311abc40-f87e-44ff-a48b-4df2276ecc47","order_by":7,"name":"Daniel M. Parker","email":"","orcid":"","institution":"Rhodes University","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"M.","lastName":"Parker","suffix":""},{"id":377579064,"identity":"3dfed83b-6fd6-41b4-b6a4-eeb58e30df1b","order_by":8,"name":"Petr Pyšek","email":"","orcid":"","institution":"Institute of Botany","correspondingAuthor":false,"prefix":"","firstName":"Petr","middleName":"","lastName":"Pyšek","suffix":""},{"id":377579065,"identity":"fe06eea4-6545-48b1-a1b6-c477ea416af3","order_by":9,"name":"Klára Pyšková","email":"","orcid":"","institution":"Institute of Botany","correspondingAuthor":false,"prefix":"","firstName":"Klára","middleName":"","lastName":"Pyšková","suffix":""},{"id":377579072,"identity":"95dfaaf2-6811-457c-a6dc-9d46940a935d","order_by":10,"name":"Peter J. Taylor","email":"","orcid":"","institution":"University of KwaZulu-Natal","correspondingAuthor":false,"prefix":"","firstName":"Peter","middleName":"J.","lastName":"Taylor","suffix":""},{"id":377579073,"identity":"9665c79a-a621-461e-abdf-1bedb81d4685","order_by":11,"name":"Robert Tropek","email":"","orcid":"","institution":"Institute of Entomology","correspondingAuthor":false,"prefix":"","firstName":"Robert","middleName":"","lastName":"Tropek","suffix":""},{"id":377579074,"identity":"aa7cb46f-b5be-4dc4-96f3-19c9c6863d1a","order_by":12,"name":"Sina M. Weier","email":"","orcid":"","institution":"University of KwaZulu-Natal","correspondingAuthor":false,"prefix":"","firstName":"Sina","middleName":"M.","lastName":"Weier","suffix":""},{"id":377579075,"identity":"2cf643b2-cd3a-4bbc-9d21-73d80440b312","order_by":13,"name":"Ivan Horáček","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1UlEQVRIiWNgGAWjYBACAwiVIIcQYiZSizFCKbFaEhsIKoUBc/behx9+VKSlb2c/f0y6gsEucTs7A+PHH3i0WPYcN5bsOZOTu7MnmU3yDENy4s5mBmYJCXwOu5HGIM3YVpG74QBQSwPDgcQNhxnYYO7FpYX5N+O/inSD84+RtCTg18ImzdiQk2BwA9mWA/i0nDnGZtlzLM1ww43HxpYNBsnGGw4zNgP14tFyvI35xo+aZHmD84kPbzZU2MluOH/4IN4QQzcBRDDis2MUjIJRMApGATEAALzhSemDhYTQAAAAAElFTkSuQmCC","orcid":"","institution":"Charles University","correspondingAuthor":true,"prefix":"","firstName":"Ivan","middleName":"","lastName":"Horáček","suffix":""}],"badges":[],"createdAt":"2024-10-19 15:38:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5295291/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5295291/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":69103140,"identity":"1d8095f1-6106-4f30-baa7-ca028194e3ed","added_by":"auto","created_at":"2024-11-15 16:31:40","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":476660,"visible":true,"origin":"","legend":"\u003cp\u003eGeographic setting of the monitored point sites (plots) in Kruger National Park with delimitation of the geographic landsystems (LET, PH, SH, SAT) and basic climatic characteristics of the region: averaged course of mean month temperatures in the southern and northern part and monthly sums of precipitation during the study period (monitoring terms indicted by blue strips).\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-5295291/v1/51f36f5c3cbc3a55eda09e1b.png"},{"id":69103141,"identity":"52efa643-4e03-4baa-8949-f09b4c9698b5","added_by":"auto","created_at":"2024-11-15 16:31:40","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":263280,"visible":true,"origin":"","legend":"\u003cp\u003eA synoptic survey of bat abundances in particular sites. A. Geographic (latitude/longitude) location of plots, their mean bat abundances and between-season differences in sample abundances, and mopane distribution. B. Total abundances of particular OTUs (bottom) and graphical outline of their appearance in particular plots (both seasons combined).\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-5295291/v1/88079e413c249b9a0f409c82.png"},{"id":69103142,"identity":"f16da427-615b-44b0-8e49-76672e060496","added_by":"auto","created_at":"2024-11-15 16:31:40","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":175160,"visible":true,"origin":"","legend":"\u003cp\u003e(A) Variation in abundances of bat communities foraging in different habitat types (plots at perennial rivers, seasonal rivers and crests) and (B) their dominance structure (n = 40 per habitat, merged across seasons).\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-5295291/v1/546c4f33e7d4f5c598aedf20.png"},{"id":69103878,"identity":"b2bfbfd3-84f1-42ec-88ab-a4b4a28029e2","added_by":"auto","created_at":"2024-11-15 16:39:40","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":158481,"visible":true,"origin":"","legend":"\u003cp\u003eOrdination plot (RDA) visualizing the relations of particular OTUs abundances to statistically significant environmental predictors related to abiotic factors (A, B) and vegetation characteristics (C). Note the distant position of PIPRUS, NEONAN, CHAPUM and EPTHOT in (C): the former two being related to a shrub and herb vegetation richness (comp. also the preference for central areas of the region in NEONAN in A), the latter two to open ground habitats suitable for mobile aerial foragers. Position of ETHOT supports the alternative taxonomic intepretation of this OTU suggested in Discussion.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-5295291/v1/997a17b2f27a83c0e0ed46a4.png"},{"id":69103879,"identity":"17a558d0-e9d0-42bc-aee4-66565d9de9e1","added_by":"auto","created_at":"2024-11-15 16:39:40","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":308432,"visible":true,"origin":"","legend":"\u003cp\u003eDetrended correspondence analysis of abundances of clutter-edge foragers (Vespertilionidae, blue) and open-air foragers (Molossidae + Emballonuridae, red). Note the distinct separation of both guilds by the factor axes 1 and 2 (A, C) contrasting to the extensive similarity among molossid parataxa but considerable differences among core elements of vespertilionids and distant position of an emballonurid (TAPMAU) at plot of factors 2 and 3 (B, D). Separate position of parataxon \"Molossid45\" supports doubts on its belonging to the latter group (comp. Supplementary File1). Centroid values of vespertilionids (blue), molossid (red) and total sample (white) indicated by stars. A, B - EPTHOT interpreted as \u003cem\u003eEptesicus hottentotus, \u003c/em\u003eC, D - EPTHOT interpreted as \u003cem\u003eMops condylurus \u003c/em\u003e(see Discussion).\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-5295291/v1/427fdbc7c3d98700ee5422e9.png"},{"id":69103147,"identity":"6fcdae9d-4cf1-46ba-a920-94234eb5a10e","added_by":"auto","created_at":"2024-11-15 16:31:40","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":257622,"visible":true,"origin":"","legend":"\u003cp\u003ePosition of particular OTUs in phenotype 3D morphospace of mean body size, wing and echolocation properties with the indication of their abundances in the total sample. Body size and wing loading data taken from Aldridge and Rautenbach\u003csup\u003e78\u003c/sup\u003e, Schoeman and Jacobs\u003csup\u003e47\u003c/sup\u003e, Monadjem et al.\u003csup\u003e 53\u003c/sup\u003e, Wilson and Mittermeier\u003csup\u003e79\u003c/sup\u003e, Waghiiwimbom et al.\u003csup\u003e80\u003c/sup\u003e, echolocation properties are represented by PC1 values of all echolocation variables based on our own record (supplementary File S1). Stars refer to centroids computed from input mean values (grey) and those weighted by OTUs abundances (yellow). (A) EPTHOT as \u003cem\u003eEptesicus hottentotus, \u003c/em\u003e(B) EPTHOT as \u003cem\u003eMops condylurus\u003c/em\u003e. (see Discussion)\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-5295291/v1/57338819629a5ed2c2d79a8b.png"},{"id":69103145,"identity":"e9e90580-7f16-4137-bed9-2f9c7f725b4f","added_by":"auto","created_at":"2024-11-15 16:31:40","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":234793,"visible":true,"origin":"","legend":"\u003cp\u003eTernary plot of vegetation-related characteristics: cover of herbs, shrubs and trees. Left: values for all plots studied (n = 60) and the centroid (star). Right: Corresponding diagrams for plots with the highest values of Shannon diversity (H*\u0026gt;2.0) of bat communities in February and November periods, and mean values for individual OTUs of Vespertilionidae and Molossidae, weighted by OTUs' abundances in particular plots, and centroids of respective groups (stars).\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-5295291/v1/9a0248143f5f88728efe847f.png"},{"id":69103146,"identity":"c9ae6be8-4f12-4d3a-acc0-022aa606873b","added_by":"auto","created_at":"2024-11-15 16:31:40","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":177436,"visible":true,"origin":"","legend":"\u003cp\u003eMatrix plot of nestedness pattern of the studied bat communities, OTUs (columns) and sites (rows) arranged according to their contribution to the nestedness pattern, which is roughly scaled by abundances of particular OTUs except for SCODIN and NEONAN (indicated by arrowheads).\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-5295291/v1/af799a4aa3e1555027a268d2.png"},{"id":69104116,"identity":"3a5dd129-f54c-4145-b650-1765be7aac59","added_by":"auto","created_at":"2024-11-15 16:47:45","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3405914,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5295291/v1/147c700b-6e0e-4090-b418-0ac50d25fb6e.pdf"},{"id":69103139,"identity":"038bddfb-e344-469f-987d-b98f5872011c","added_by":"auto","created_at":"2024-11-15 16:31:40","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":56711,"visible":true,"origin":"","legend":"","description":"","filename":"Table1and3.docx","url":"https://assets-eu.researchsquare.com/files/rs-5295291/v1/9e7856e59972d12334f15a60.docx"},{"id":69103148,"identity":"6f3c7520-a692-4ec3-94d3-80aed8140b2c","added_by":"auto","created_at":"2024-11-15 16:31:40","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":6747759,"visible":true,"origin":"","legend":"","description":"","filename":"knpSupplementaryInformationSF15OK.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5295291/v1/05e7484f674c286009cbdd90.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Bat communities of savanna biome in the Kruger National Park, South Africa","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe savanna biome represents an essential component of global biodiversity that occupies a fifth of the earth\u0026rsquo;s land surface\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e and hosts a huge variety of biota that are rare in other habitats. This is particularly valid for southern Africa, where the savanna biome covers more than 54% of the total area\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. The savanna biome is often presented as a dynamic mixture of semi-open habitats interfacing tropical forest and pure grassland biomes, with a climate characterized by low to intermediate rainfall and mild seasonality combined with tropic to subtropic thermal conditions\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Sankaran et al.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e demonstrated that in Africa, the arid and semiarid savanna represents a \u0026ldquo;stable\u0026rdquo; formation in areas with mean annual precipitation of \u0026lt;\u0026thinsp;650 mm, while areas with higher precipitation represent an \u0026ldquo;unstable\u0026rdquo; system where the coexistence of herbs and trees depends on steady effects of disturbing factors (fire, herbivores).\u003c/p\u003e \u003cp\u003eYet, responses of particular biotic communities of savanna to extrinsic drivers promoting the habitat mosaic differ\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e, similar to responses of diverse community members to variation in vegetation cover and other variables of environmental context. The multidisciplinary project MOSAIK (Monitoring Savanna Biodiversity in Kruger National Park) was established to examine the patterns of spatial and temporal variation in syntopic savanna communities (plants, insects, birds, large mammals, and bats) by using a standardized monitoring design that consisted of 20 triplets of fixed plots (typically 6\u0026ndash;8 km apart) each covering three main habitat types defined with regard to water availability (perennial rivers, seasonal rivers, and dry crests), disturbance levels, and landsystems of the region (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). This design enabled to attribute the local, subregional and regional scales to patterns of within-plot variation\u003csup\u003e\u003cspan additionalcitationids=\"CR6 CR7\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe Kruger National Park (KNP) represents a suitable model area for such studies. It covers a large landscape unit of lowland semiarid savanna (19,175 km\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e, mean elevation 342 m a.s.l., mean annual precipitation 511 mm), protected for 120 years from anthropogenic influence. With all that, it provides an exceptional opportunity to analyze the organization of the savanna biome by natural drivers to the full extent. A plethora of previous research projects undertaken in KNP\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan additionalcitationids=\"CR10 CR11\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e provide a robust contextual background. Among others, this concerns detailed climatic and environmental data from high-resolution remote sensing\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e and diverse records of large-scale programs of long-term ground monitoring\u003csup\u003e\u003cspan additionalcitationids=\"CR16\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Thus, for KNP, comprehensive information is available on local variation in the biotic and abiotic drivers of landscape heterogeneity and its spatiotemporal dynamics\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e, including currently changing trends\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e such as, e.g., impact of elephants on large-scale pattern of habitat mosaic\u003csup\u003e\u003cspan additionalcitationids=\"CR20\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe present study deals with bats, a group for which KNP represents one of the pertinent hotspots of African diversity; inferring from more than 50 literary sources focused on bats in this area, the park is also one of the best-investigated regional units in Africa. Compared to former reports surveying mostly records obtained with the aid of mist netting\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e or details of reproduction biology of several local species\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e, the more recent studies on bat communities in Kruger NP and close surroundings are patterned by the influx of innovative field techniques, first of all, the application of ultrasonic detectors\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e and automated acoustic recordings. Detailed data on interspecific variation in echolocation calls of particular species and investigations on the complementarity of acoustic and non-acoustic survey methods\u003csup\u003e\u003cspan additionalcitationids=\"CR27 CR28 CR29\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e created a robust methodological platform enabling to integrate results of acoustic monitoring into macroecological analyses addressing the large-scale patterns of bat diversity in South Africa\u003csup\u003e\u003cspan additionalcitationids=\"CR32 CR33 CR34 CR35\" citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. Yet, in contrast to robust information about the natural drivers associated with regional and supraregional variation in bat communities, the patterns of microgeographic variation at subregional and local scales and the feedback loop between local and regional dynamics still have not been adequately comprehended\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eBats differ from other animal groups studied within MOSAIK by extreme mobility (not constrained by territoriality as in birds), disposing them to respond rapidly to spatial variation in dispersal and capacity of their feeding resources, notwithstanding some unique features of their biology: small body size, temporal aggregation in day-time resting colonies, dependence upon the availability of taxon-specific roosts, etc. Studies in the Neotropical region revealed that the abundances and species richness of bat communities in the savanna biome correspond to those of communities in the rainforest habitats\u003csup\u003e\u003cspan additionalcitationids=\"CR38 CR39 CR40 CR41 CR42 CR43\" citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e and this obviously holds also for the Ethiopian region. Numerous studies addressing this topic in southern and central Africa were comprehensively reviewed by Schoeman and Monadjem\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. Here, the peak richness of savanna bats occurs in northeastern southern Africa, where at local and regional scales, savanna bat assemblages include more than 30 species\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan additionalcitationids=\"CR46 CR47\" citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. Schoeman and Monadjem\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e stress that in terms of alpha and gamma diversity, bats exceed any other group of mammals inhabiting the savanna subregion and raise the question of whether these species-rich assemblages are structured in any way.\u003c/p\u003e \u003cp\u003eThus, the central questions for any study of savanna bat communities are whether they represent (i) mere \u003cem\u003ead hoc\u003c/em\u003e constituted assemblages of incidentally co-occurring species attracted by the local patches of the habitats to which they are primarily adapted\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e or (ii) integrated entities established by a long-term coexistence and balanced partitioning of the resources inherently provided by the savanna biome and its dynamics\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e,\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e? Our paper, based on an extensive dataset of automated whole-night acoustic recordings of foraging bat communities collected under standardized setting of the MOSAIK project, is intended to contribute to these topics by testing the effects of diverse natural drivers, analyzing variation in community patterns at point, local, and regional scales and revealing the commonalities across these scales.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eStructure of the dataset.\u003c/strong\u003e Within the set of all 120 primary records (plot/night), the total abundances (number of individual bat passes recorded during one night) ranged from 22 to 7,003 (mean\u0026thinsp;\u0026plusmn;\u0026thinsp;S.D. = 1090\u0026thinsp;\u0026plusmn;\u0026thinsp;1199), the number of recorded OTUs (acoustic parataxa) varied from 4 to 19 (mean 12.3\u0026thinsp;\u0026plusmn;\u0026thinsp;3.22). For the sets of 60 plot records (February and November samples merged), the respective values were abundances of 122 to 8,565 (mean 1887\u0026thinsp;\u0026plusmn;\u0026thinsp;1642, for spatial pattern see Fig.\u0026nbsp;\u003cspan\u003e2\u003c/span\u003eA), and OTUs from 10 to 22 (mean 15.0\u0026thinsp;\u0026plusmn;\u0026thinsp;2.32).\u003c/p\u003e\n\u003cdiv\u003e\n\u003c/div\u003e\n\u003cp\u003eThe rank distributions (RAD) of abundances and species richness for seasonal plot records and all plots merged fit well to a log-normal scaling with both variables mutually evenly scaled (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.358 and 0.228, respectively) except for a set of 11 samples with excessively high abundances (Supplementary File S2.1). In total, 93.7% of all samples were composed exclusively of OTUs belonging to Vespertilionidae, Molossidae, and Emballonuridae; representation of other groups was more or less incidental, and they were omitted from the present analysis (compare Supplementary File S 1 and Table\u0026nbsp;\u003cspan\u003e1\u003c/span\u003e). The rank distribution of total abundances of individual OTUs (Supplementary File S2.1) showed a nearly ideal log-normal distribution (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.87).\u003c/p\u003e\n\u003cp\u003eThe core of community samples in all plots in both seasons was composed of four eudominant (d\u0026thinsp;\u0026gt;\u0026thinsp;10%) and three dominant (10%\u0026lt;d\u0026thinsp;\u0026gt;\u0026thinsp;5%) OTUs (CHAPUM \u0026ndash; \u003cem\u003eChaerephon pumilus\u003c/em\u003e, TADAEG \u0026ndash; \u003cem\u003eTadarida aegyptiaca\u003c/em\u003e, EPTHOT \u0026ndash; \u003cem\u003eEptesicus hottentotus\u003c/em\u003e (but see below), NEONAN \u0026ndash; \u003cem\u003eAfronycteris nana\u003c/em\u003e, and CHAANS \u0026ndash; \u003cem\u003eChaerephon ansorgei\u003c/em\u003e, NEOCAP \u0026ndash; \u003cem\u003eLaephoptis capensis\u003c/em\u003e, SCODIN \u0026ndash; \u003cem\u003eScotophilus dinganii\u003c/em\u003e, respectively) with slight differences in their abundance ranks among particular geographic units (and particularly between February and November samples). Together with other five, these OTUs appeared in more than 90% of the samples; Table\u0026nbsp;\u003cspan\u003e1\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan\u003e2\u003c/span\u003eB).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEffect of environmental variables on bat species richness, diversity, abundance, and composition.\u003c/strong\u003e Habitat type (perennial rivers, seasonal rivers, crests) had a highly significant effect (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) on bat species richness. Presence/absence of mopane (\u003cem\u003eColophospermum mopane\u003c/em\u003e, a dominant tree species at some plots; p\u0026thinsp;=\u0026thinsp;0.001), landsystem (p\u0026thinsp;=\u0026thinsp;0.002), north versus south (p\u0026thinsp;=\u0026thinsp;0.004), longitude (p\u0026thinsp;=\u0026thinsp;0.04), and distance to camps p\u0026thinsp;=\u0026thinsp;0.042) also significantly affected bat species richness. The relation between distance to camps and bat species richness contained a significant quadratic term. The effect of mopane was also significant as a continuous variable based on a visual estimate of its cover (p\u0026thinsp;=\u0026thinsp;0.039; Fig.\u0026nbsp;\u003cspan\u003e3\u003c/span\u003e). Further, there were significant interactions between the north/south location of plots and habitat as well as between landsystem and habitat (p\u0026thinsp;=\u0026thinsp;0.006 and p\u0026thinsp;=\u0026thinsp;0.033, respectively). In terms of the direction of these effects, plots near perennial and seasonal rivers harboured significantly more species compared to plots on crests (p\u0026thinsp;=\u0026thinsp;0.002 and p\u0026thinsp;=\u0026thinsp;0.001, respectively). Further, plots in the Satara landsystem harboured more bat species compared to plots in the Phalaborwa and Letaba landsystems (p\u0026thinsp;=\u0026thinsp;0.007 and p\u0026thinsp;=\u0026thinsp;0.014, respectively). Plots in the south of KNP harboured more species compared to plots in the north. Similarly, plots without dominant mopane harboured more species than plots with mopane present (Tables\u0026nbsp;\u003cspan\u003e2\u003c/span\u003e and \u003cspan\u003e3\u003c/span\u003e).\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv\u003eTable 2\u003c/div\u003e\n \u003cdiv\u003e\n \u003cp\u003eDirect ordination analysis (RDA) on the effects of environmental variables on the dominance structure of bat community in particular sites (as a response variable, n\u0026thinsp;=\u0026thinsp;120). A list of the variables with significant effects revealed by forward selection is presented.\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"3\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" colspan=\"3\"\u003e\n \u003cp\u003ecategorical\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eexplained vairability (%)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ep-value\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eall predictors\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e12.5\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e0.024\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"3\"\u003e\n \u003cp\u003e\u003cstrong\u003evegetation parameters\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eall predictors\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e47.6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.006\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003erichness of shrubs\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e14.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.008\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePCO1 - vegetation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.014\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePCO3 - vegetation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.032\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003egrass cover\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.044\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003erichness of herbs\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.046\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"3\"\u003e\n \u003cp\u003e\u003cstrong\u003eenvironmental context\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eall predictors\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e45.6\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.028\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003evariability in precipitation (S.D.)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e24.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003edistance to parkboundary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e15.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003elattitude\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.034\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cdiv\u003e\n\u003c/div\u003e\n\u003cp\u003eHabitat type had a highly significant effect on the Shannon diversity of bats (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and there were also significant effects of mopane presence/absence (p\u0026thinsp;=\u0026thinsp;0.002), mopane cover (p\u0026thinsp;=\u0026thinsp;0.003), landsystem (p\u0026thinsp;=\u0026thinsp;0.007), north/south location (p\u0026thinsp;=\u0026thinsp;0.012), and distance to streams (p\u0026thinsp;=\u0026thinsp;0.039). The relation between Shannon diversity and distance to streams contained a significant quadratic term. There was also a significant interaction between landsystem and habitat, and between north/south and habitat (p\u0026thinsp;=\u0026thinsp;0.009 and p\u0026thinsp;=\u0026thinsp;0.01, respectively). Plots near perennial and seasonal rivers showed significantly higher Shannon diversity of bat species compared to plots on crests (both p\u0026thinsp;=\u0026thinsp;0.001). Interestingly, Tukey HSD method did not identify any significant contrasts among landsystems, despite their overall significant effect on the Shannon diversity. Similarly to species richness, plots in the south of KNP showed higher Shannon diversity of bat species compared to plots in the north, and the same holds for the plots without mopane compared to plots with mopane present (Tables\u0026nbsp;\u003cspan\u003e2\u003c/span\u003e and \u003cspan\u003e3\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eThe type of habitat, distance to streams, and PCO2 (second ordination axis representing the vegetation gradient) had significant effects on the Pielou evenness of bat species (p\u0026thinsp;=\u0026thinsp;0.006, p\u0026thinsp;=\u0026thinsp;0.014, and p\u0026thinsp;=\u0026thinsp;0.017, respectively). The effect of PCO2 contained a significant quadratic term. Similarly to bat species richness and Shannon diversity, plots near perennial rivers showed a significantly higher Pielou evenness compared to plots on crests (p\u0026thinsp;=\u0026thinsp;0.007).\u003c/p\u003e\n\u003cp\u003eConcerning the total abundance of bats as a response variable, the effects of habitat and distance to rivers were highly significant (both p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), as was that of waterbody area (waterSum) (p\u0026thinsp;=\u0026thinsp;0.001), with a significant quadratic term. There was also a significant interaction between bedrock and distance to rivers (p\u0026thinsp;=\u0026thinsp;0.03). Plots near perennial rivers showed a higher total abundance of bats compared to plots near seasonal rivers and on crests (p\u0026thinsp;=\u0026thinsp;0.017 and p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, respectively). Similarly to plots near perennial rivers, those near seasonal rivers also showed a higher abundance of bats compared to plots on crests (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e\n\u003cp\u003eConcerning the effects on the species composition of bat communities, all three groups of predictors (categorical, vegetation, abiotic environment) had significant overall effects (p\u0026thinsp;=\u0026thinsp;0.024, p\u0026thinsp;=\u0026thinsp;0.006 and p\u0026thinsp;=\u0026thinsp;0.028, respectively). Vegetation characteristics explained most variability (47.6%), followed by environmental context (45.6%) and categorical predictors (12.5%; Table\u0026nbsp;\u003cspan\u003e2\u003c/span\u003e). Of vegetation parameters, the forward selection procedure identified the richness of shrubs (p\u0026thinsp;=\u0026thinsp;0.008), vegetation PCO1 (p\u0026thinsp;=\u0026thinsp;0.014), vegetation PCO3 (p\u0026thinsp;=\u0026thinsp;0.032), grass cover (p\u0026thinsp;=\u0026thinsp;0.044), and richness of herbs (p\u0026thinsp;=\u0026thinsp;0.046) as significant predictors; Fig.\u0026nbsp;\u003cspan\u003e4\u003c/span\u003e). These vegetation parameters explained 14.2%, 12.7%, 11.4%, 7.7% and 7.4% of variability, respectively. Variability in precipitation (expressed as its standard deviation), distance to the park boundary, and latitude were the most influential abiotic environmental predictors (p\u0026thinsp;=\u0026thinsp;0.002, p\u0026thinsp;=\u0026thinsp;0.002 and p\u0026thinsp;=\u0026thinsp;0.034, respectively), explaining 24.8%, 15.4% and 8.9% of variability, respectively; Table\u0026nbsp;\u003cspan\u003e3\u003c/span\u003e). Interestingly, no significant effects were found for the categorical predictors, even though their overall effect was significant. Also, we did not find a significant effect of insect supply (weight of insects in light traps taken synchronously with bat recording at a plot) on the above-surveyed community variables or the abundance of any single OTU. The regression tree analyses (Supplementary File S3) revealed a key effect of triplet specifities upon either species richness, Shannon diversity, Pielou evenness, and abundance, predominating over those of other predictors (including water availability, mopane cover, distance to camps, herb diversity or bedrock).\u003c/p\u003e\n\u003cp\u003eNone of the environmental predictors, except water availability, significantly affected the variation in the quantitative representation of individual OTUs in particular community samples, (comp. also Pielou evenness in Table\u0026nbsp;\u003cspan\u003e3\u003c/span\u003e). Further comparison dealing with abundances and dominances of individual OTUs (Supplementary File S2.3) showed a mosaic of weakly pronounced effects of few environmental predictors, in most instances restricted to one or few OTUs, except for the following factors: season (valid for 10 OTUs), habitat type (access to water) for eight OTUs, triplet (microgeographic variation) for seven OTUs, distance to the nearest campsite (valid for five OTUs), and species richness of herbs (valid for three OTUs; Supplementary File S2.3). A comparison of the community structure in plots delimited by contrasting states of the two most influencing environmental variables (habitat, mopane; Supplementary File S2.4) showed only shallow differences, at least for most of the core OTUs.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eVariation in community structure.\u003c/strong\u003e In all community samples, the two major foraging guilds, i.e., the clutter-edge foragers (Vespertilionidae) and open-air foragers (Molossidae), were represented with nearly equal abundances. Their abundance ratio (Vespertilionidae/Molossidae) varied from 0.13 to 3.47 around a mean value of 1.09\u0026thinsp;\u0026plusmn;\u0026thinsp;0.80 (mean\u0026thinsp;\u0026plusmn;\u0026thinsp;S.D.); in plots with high abundance (n\u0026thinsp;\u0026gt;\u0026thinsp;2,000) it was 0.24\u0026ndash;3.47, with the mean of 1.10\u0026thinsp;\u0026plusmn;\u0026thinsp;0.89, and in those with low abundances (n\u0026thinsp;\u0026lt;\u0026thinsp;1000) the corresponding values were 0.13\u0026ndash;2.91 and 1.18\u0026thinsp;\u0026plusmn;\u0026thinsp;0.76, respectively.\u003c/p\u003e\n\u003cp\u003eThe correspondence analysis (Fig.\u0026nbsp;\u003cspan\u003e5\u003c/span\u003e) revealed nearly equidistant dispersal of community members in the component space of abundance distributions and distinct divergence between the two major guilds. The projection of DC2 and DC3 axes (Fig.\u0026nbsp;\u003cspan\u003e5\u003c/span\u003eB) showed the extensive similarity among all molossid OTUs, the dissimilarity of EPTHOT to other vespertilionids (supporting an alternative taxonomic affiliation of that acoustic parataxon in Fig.\u0026nbsp;\u003cspan\u003e5\u003c/span\u003eCD; see Discussion), and distant position of an emballonurid (TAPMAU).\u003c/p\u003e\n\u003cp\u003eThe 3D reconstruction of phenetic packing (Fig.\u0026nbsp;\u003cspan\u003e6\u003c/span\u003e) also suggested a nearly equidistant distribution of particular community members with the concentration of core elements along the community centroid except for a marginal position of NEONAN (\u003cem\u003eAfroromicia nana\u003c/em\u003e) and PIPRUS (\u003cem\u003ePipistrellus rusticus\u003c/em\u003e) that markedly differed from others in response to particular environmental predictors (compare Fig.\u0026nbsp;\u003cspan\u003e4\u003c/span\u003e and Supplementary File S2.3).\u003c/p\u003e\n\u003cp\u003eThe comparison of species composition and dominance across all 120 primary plot records revealed a strong signal of the overall homogeneity of the community structure. The Jaccard index as a measure of similarity in species composition ranged from 0.306\u0026ndash;0.737, with a mean 0.634\u0026thinsp;\u0026plusmn;\u0026thinsp;0.153, while values of Bray-Curtis index of dissimilarity in dominance structure were low, BC\u0026thinsp;=\u0026thinsp;0.089\u0026ndash;0.414, with a mean of 0.317\u0026thinsp;\u0026plusmn;\u0026thinsp;0.176. Moreover, a detailed comparison of between-plot similarities in species composition and dominance structure among geographic or environmental units (merged seasonal records, Table\u0026nbsp;\u003cspan\u003e4\u003c/span\u003e) revealed a high degree of homogeneity: all samples exhibited about 80% overlap in species composition and 50\u0026ndash;60% in dominance structure, without any marked differences between particular local (triplet-related) and subregional (landsystem-related) units.\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable id=\"Tab4\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv\u003eTable 4\u003c/div\u003e\n \u003cdiv\u003e\n \u003cp\u003eThe similarity in species composition (measured by the Jaccard index) and in dominance structure (Renkonnen index) calculated for within- and between-community data representing particular spatial units or those assembled by habitat type, i.e. access to surface water (C, P, S).\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"11\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003especies composition\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003edominance structure\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eJaccard\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"4\"\u003e\n \u003cp\u003eRenkonen\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eavg\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003emin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003emax\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eavg\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003emin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003emax\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSD\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"12\"\u003e\n \u003cp\u003ewithin units\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.802\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.438\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.098\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.544\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.180\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.876\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.128\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCrests\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.815\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.526\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.180\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.847\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.144\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePerennial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.810\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.474\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.093\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.566\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.308\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.824\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.122\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSeasonal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.784\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.688\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.882\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.071\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.547\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.230\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.836\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.109\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSouth\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.836\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.625\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.076\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.558\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.284\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.868\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.119\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNorth\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.800\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.467\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.109\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.551\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.206\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.876\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.143\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEast\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.827\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.467\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.071\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.554\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.215\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.868\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.132\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eWest\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.820\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.533\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.099\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.582\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.314\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.876\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.123\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSAT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.859\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.722\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.064\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.593\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.284\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.868\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.113\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSK\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.843\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.647\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.087\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.556\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.314\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.820\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.119\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLET\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.799\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.467\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.113\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.520\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.215\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.846\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.156\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.798\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.533\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.105\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.609\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.330\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.876\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.122\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" rowspan=\"8\"\u003e\n \u003cp\u003ebetween units\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSouth/North\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.783\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.438\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.097\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.533\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.180\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.853\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.123\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEast/West\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.814\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.094\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.538\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.206\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.841\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.129\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSAT / PH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.754\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.474\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.944\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.098\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.572\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.309\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.781\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.103\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSAT / LET\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.785\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.556\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.090\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.526\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.276\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.792\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.108\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSAT / SK\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.823\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.625\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.071\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.543\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.292\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.841\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.118\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSK / PH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.780\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.438\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.941\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.097\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.552\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.273\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.853\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.129\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSK / LET\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.810\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.533\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.095\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.484\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.180\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.763\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.131\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLET / PH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.802\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.110\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.535\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.206\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.847\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.138\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eAlthough the vegetation cover differed among the three habitat types and microgeographic units (triplets), it had no significant effect on the abundance of particular OTUs or on other community variables (Supplementary File S2.3). The habitat preference of all OTUs, reflecting association with the cover of herbs, shrubs and trees (Fig.\u0026nbsp;\u003cspan\u003e7\u003c/span\u003e), resulted in their compact aggregation around centroid values that were almost identical for both molossid and vespertilionid OTUs in the February and November period and for the sites with exceptionally high bat species Shannon H\u0026rsquo; diversity.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEffect of spatial scale.\u003c/strong\u003e To examine the trade-off between community variables and spatial scaling, we compared diversity measures in subsets representing different spatial scales (Table\u0026nbsp;\u003cspan\u003e5\u003c/span\u003e). We found no essential between-scale differences in alpha diversity and generally low beta diversity, including beta turnover (beta T according to Wilson and Shmida\u003csup\u003e\u003cspan\u003e52\u003c/span\u003e\u003c/sup\u003e). Except for the plot- and local scales, the beta diversity was negligible in contrast to non-spatially scaled clusters of plots grouped by habitat. This suggests that at each spatial scale, the whole set of community samples exhibited characteristics of a single entity. This picture is further illustrated by (i) a high degree of community nestedness, (ii) abundance relations among the community members demonstrated by a detrended correspondence analysis (Fig.\u0026nbsp;\u003cspan\u003e6\u003c/span\u003e), and (iii) the patterns of its phenetic packing (Fig.\u0026nbsp;\u003cspan\u003e7\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eThe nestedness analyzed by using metrics based on overlaps and decreasing fills (NODF) and matrix temperatures (MT) yielded mean values of 81.4 and 15.5, respectively, with variation at different spatial scales from 75.1 to 85.7 (NODF) and 8.1 to 19.2 (MT) (Fig.\u0026nbsp;\u003cspan\u003e8\u003c/span\u003e, Supplementary File S4). It conforms to a very high degree of incipient integrity of the community structure over the whole studied area. The contribution of particular OTUs to the nestedness pattern was generally scaled by their abundances. Two notable exceptions are the peak role of \u003cem\u003eScotophilus dinganii\u003c/em\u003e in establishing the nestedness pattern and the lowest degree of nestedness contribution by \u003cem\u003eAfronycteris nana\u003c/em\u003e. The central position of SCODIN (\u003cem\u003eScotophilus dingani)\u003c/em\u003e, contrasting with the marginal position of NEONAN (\u003cem\u003eAfronycteris nana\u003c/em\u003e), resembling their positions in the phenotype setting of the community (Fig.\u0026nbsp;\u003cspan\u003e6\u003c/span\u003e) is here particularly worth mentioning.\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable id=\"Tab5\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv\u003eTable 5\u003c/div\u003e\n \u003cdiv\u003e\n \u003cp\u003eComparison of community characteristics: sp \u0026ndash; species richness; N \u0026ndash; species abundances; H\u0026rsquo; - Shannon alpha diversity; Simpson 1-D alpha diversity; beta diversity W\u003csup\u003e\u003cspan\u003e68\u003c/span\u003e\u003c/sup\u003e, beta T\u003csup\u003e\u003cspan\u003e52\u003c/span\u003e\u003c/sup\u003e in sample sets representing different spatial scales: point, local (triplets), landsystems, subregional and regional (with mean grid size in km\u003csup\u003e\u003cspan\u003e2\u003c/span\u003e\u003c/sup\u003e) and those assembled by habitat type, i.e. access to surface water (C, P, S).\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"13\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eScale:\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003epoint\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003elocal (triplets)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003elandsystem\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003esubregional\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003esubregional\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eregional\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eC\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eS\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003en\u0026thinsp;=\u0026thinsp;60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003en\u0026thinsp;=\u0026thinsp;20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003en\u0026thinsp;=\u0026thinsp;4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003en\u0026thinsp;=\u0026thinsp;2: N-S\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003en\u0026thinsp;=\u0026thinsp;2: E-W\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003en\u0026thinsp;=\u0026thinsp;1: all\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003en\u0026thinsp;=\u0026thinsp;20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003en\u0026thinsp;=\u0026thinsp;20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003en\u0026thinsp;=\u0026thinsp;20\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ekm2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4,000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e10,000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e10,000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e20,000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003esp\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eavg\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e14.9\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e16.7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e18.3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e18.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e19.0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e19\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e14.3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e15.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e15.7\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003emin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003emax\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eN\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eavg\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e1,886.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e5,659.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e28,297.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e56,595.0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e56,595.0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e113,190.0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e1,025.5\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e3,030.1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e2,042.0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003emin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e122\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1,181\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e21,727\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e54,218\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e43,532\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e255\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1,101\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e299\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003emax\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8,563\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13,954\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e37,245\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e58,972\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e69,658\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3,094\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8,563\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5,789\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eH*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eavg\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e2.009\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e2.147\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e2.267\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e2.312\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e2.316\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e2.332\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.917\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.967\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e2.089\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003emin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.252\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.603\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.976\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.175\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.241\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.253\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.372\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.455\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003emax\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.501\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.427\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.435\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.440\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.391\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.277\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.420\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.501\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eSimpson 1-D\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eavg\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.816\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.840\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.862\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.870\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.869\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.874\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.792\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.803\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.835\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003emin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.602\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.711\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.805\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.847\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.855\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.631\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.602\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.692\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003emax\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.908\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.897\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.890\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.892\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.883\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.881\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.891\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.908\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eBeta W\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.274\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.138\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.041\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.027\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.000\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.326\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.228\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.208\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eBeta T\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e3.291\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e2.048\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.781\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.514\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.000\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e3.844\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e3.030\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003e2.898\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eDue to the methodical restrictions of acoustic recordings, we collected relevant data only for two of the four guilds of local bat communities, i.e., aerial foragers (Molossidae, Emballonuridae) and clutter-edge forages (Vespertilionidae). For obvious reasons, fruit bats are not represented in our sample, and records of clutter foragers (Rhinolophidae, Hipposideridae) and ground gleaners (Nycteridae) were restricted to accidental recordings and thus omitted from most comparisons. Nevertheless, as also live captures suggest\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e, the essential contribution of the former two guilds to bat communities of the region, both in terms of their species richness and abundance, is a real phenomenon. Our results can thus be assumed to provide relevant information on the bat community structure and the patterns of its variation in KNP.\u003c/p\u003e \u003cp\u003eYet, the applied method is incipiently associated with putative uncertainties in the actual taxonomic identity of particular acoustic parataxa\u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. Although the suggested taxonomical identity was robustly supported in most OTUs included in our study (Supplementary File S1), two items are worth discussing. One is MOLOSSID45 \u0026ndash; the calls resembling molossid bats by a flat frequency sweep (which alternatively could be interpreted as atypical search flight calls of an unidentified vespertilionid). The other case is more noteworthy, as it concerns one of the core OTUs: EPTHOT. Here, the post-hoc comparisons reveal several discrepancies: (i) Based on capturing records, \u003cem\u003eEptesicus hottentotus\u003c/em\u003e appears to be a rare bat with quite a sparse occurrence in the region\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e. (ii) The sonographic patterns conforming to the search echolocation calls of \u003cem\u003eE. hottentotus\u003c/em\u003e (see Supplementary File SF1) might appear at the approach stage of call sequences of some molossid bats, first of all \u003cem\u003eMops condylurus\u003c/em\u003e, under conditions of a high prey concentration. (iii) There is obvious discrepancy between a large number of roost records of \u003cem\u003eM. condylurus\u003c/em\u003e and relatively low abundances of MOPCON in the acoustic record (the low-frequency search calls of \u003cem\u003eM. condylurus\u003c/em\u003e; see Supplementary File S1). Thus, in regard to the fact that based on phonologic traits, it seems impossible to distinguish high-frequency FM calls of molossid from those of \u003cem\u003eEptesicus hottentotus\u003c/em\u003e, it appears reasonable to consider most EPTHOT records as belonging to \u003cem\u003eM. condylurus\u003c/em\u003e. The post-hoc comparisons of alternative taxonomic interpretations of EPTHOT (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) seem to support it quite convincingly.\u003c/p\u003e \u003cp\u003eNevertheless, despite the above-mentioned uncertainties on the real taxonomic identity of some OTUs, the standardized identification procedure of particular acoustic parataxa ensured formal consistency of input community records, which is a key prerequisite of further comparative analyses. Moreover, thanks to the identical identification procedure applied in both studies, the community characteristics that we revealed can be directly compared to the acoustic monitoring undertaken in 2017 and 2018 at 26 sites in the northernmost part of KNP and neighboring regions of Limpopo province\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e (see Supplementary File S5 for details). Both datasets show a great agreement not only in species composition, which is identical, but also in the overall dominance of particular OTUs (Wilkinson pair test, p\u0026thinsp;=\u0026thinsp;0.756). Notwithstanding particular differences (discussed in Supplementary File S5), the correspondence between both datasets convincingly supports one of the most exciting results of our study: a surprisingly high degree of homogeneity in bat community structure, extending beyond the spatial margins of the area under study.\u003c/p\u003e \u003cp\u003eEven more surprising is that just quite a few of the numerous environmental and vegetation-related predictors we examined exhibited certain effects on abundance, diversity, and variation in community structure. Those whose effects were significant, i.e., access to water (obviously related to increased concentration of insects and higher plant diversity along rivers\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e), the distance to the nearest camp (a source of multiple roosting opportunities), and overall similarity among neighboring plots within a triplet, all refer to trivial relationships with explicit meaning \u0026ndash; local food and roost availability and consistency of resident occurrence at sublocal scales.\u003c/p\u003e \u003cp\u003eMoreover, the homogeneity in species composition and dominance structure, as well as the lack of large-scale effects of environmental or vegetation predictors (notwithstanding those explaining microgeographic variations), appeared consistently at all spatial scales, as did the balanced contribution of two major foraging guilds, both with a nearly equal number of species and overall abundances. Within the whole set of recorded OTUs, we found no distinct between-species differences in their habitat preferences \u0026ndash; all taxon-specific mean values were densely clustered along the centroid identical with the centroid of habitat variation of all sites under study, i.e., the mean state of habitat conditions available in the study area.\u003c/p\u003e \u003cp\u003eAll these unexpected observations point unequivocally to the picture of a single community exhibiting at large the patterns of steady coexistence dynamics. The drivers supporting the coexistence dynamics as identified by modern coexistence theory\u003csup\u003e\u003cspan additionalcitationids=\"CR56 CR57 CR58\" citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e can be subdivided into two classes: extrinsic factors buffering effects of interspecific competition, and intrinsic factors stabilizing coexistence disposition of particular community members. Among the former class, the dynamics of environmental variation affecting temporarily the advances of individual competitors is to be mentioned first. The intermediate disturbance hypothesis\u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e,\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e predicts that local species diversity is maximized when ecological disturbances are sufficient to affect particular community members yet not too strong to force their extinctions, and, at the same time, they are neither too rare nor too frequent. Obviously, the dynamics of the savanna biome, with a mosaic of fire events and habitat structure resulting in fluctuations in the impact of large herbivores\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e, fit the picture of steady intermediate disturbance dynamics to the full extent. Among the intrinsic factors, stabilizing niche differences and relative fitness differences\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e or fitness inequalities\u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e,\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e are essential. The niche differences stabilized via resource partitioning increase the role of intraspecific over interspecific competition and accelerate adaptive rearrangements scaled by capacities of locally available resources. In the case of bats, the data required for detailed analyses of resource partitioning are, for obvious reasons, mostly not available. Yet, in some cases\u003csup\u003e\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e, they can be adequately estimated with indirect evidence, typically addressing the taxon-specific combination of the phenotype traits directly related to foraging dispositions (wing design, echolocation). The multivariate comparisons of these traits in all community members, the phenetic packing of a community\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e, can serve as reliable information, namely if supported by detailed data on individual co-occurring species\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e,\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e,\u003cspan additionalcitationids=\"CR63\" citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e\u003c/sup\u003e. Our analyses demonstrated a nearly equidistant distribution of individual species and a high degree of phenetic packing of the bat community studied, indicating a well-balanced resource partitioning with reduced effects on between-taxa competition. It is worth mentioning that the corresponding picture was also obtained by scaling the abundance of particular community members, the variable addressing the topic of fitness differences or fitness inequalities. The theory predicts a stable coexistence when stabilizing niche differences of species are greater than their relative fitness differences\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e. Also, this condition seems to be met in bats. All are strongly pronounced K-strategists with nearly equal population growth rates notwithstanding their temporal variation due to different responses of particular species to diverse disturbance cues. Moreover, as the theory predicts\u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e, the stabilizing niche difference promotes an increase in low-density growth rate, another factor contributing to coexistence dynamics.\u003c/p\u003e \u003cp\u003eIn spatial regards, the savanna community studied was characterized by a greatly pronounced nestedness pattern retained at all temporal and spatial scales. Previously, the nestedness of spatial organization was demonstrated to be an independent factor that effectively reduces interspecific competition and enhances the number of coexisting species\u003csup\u003e\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eTo summarize, our results suggest that the bat community of the savanna biome is to be considered a distinct entity integrated by multiple mechanisms promoting the coexistence of all community members. The stabilized niche differences among them arose, besides incipient ecological specificities of particular clades, from ongoing local interactions under shared common conditions provided by the savanna biome throughout its long history (including steady impacts of herbivores and fire maintaining the habitat mosaic\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e). As a result, all the community members prefer essentially just the mean conditions that the dynamics of the savanna biome provide. The differences in their spatial dispersal seem to be restricted to the fine variation manifested at the microgeographic scale (within- and between-triplet differences). At the local, subregional, and regional scales, respectively, the variation in KNP bat community takes a form of a homogeneous densely patched mosaic that stays beyond the direct control of any large-scale environmental predictors. In these regards, the bat communities illustrate distinctive features of the savanna biome, the unique element of global biodiversity worth thorough conservation interest.\u003c/p\u003e"},{"header":"Material and methods","content":"\u003cp\u003e \u003cb\u003eCollecting field data.\u003c/b\u003e The data collected within the MOSAIK project (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) come from 60 fixed point sites 50 × 50 m in size (plots) representing four subregional systems (landsystems – LET: Letaba, PHA: Phalaborwa, SAT: Satara, SKU: Skukuza), each with five local triplets of neighboring plots (6–8 km from each other) representing distinct habitats differing in water availability (P – perennial river with water available all year round, S – seasonal river with water available in rainy season, C – dry crest with no access to water). Correspondingly, the formal designation of individual sites, followed in this paper, consists of the abbreviations of the territorial unit (SKU, SAT, LET, PHA), the triplet number (1–5), and the type of site (S, P, C). All sites were controlled twice during 2018–2020, in spring, representing the rainy season (November, December – XI), and late summer, representing the onset of the dry season (February, March – II). For more detailed explanations of the MOSAIK project settings, see Hejda et al.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eBats were recorded by using an automated recorder Song Meter SM4BAT FS (Wildlife Acoustics) with ultrasonic microphone SMM-U2 (Wildlife Acoustics) placed (typically in a centre of the plot) by the stick at a height of 4 m (to reduce noise from stridulation of ground insects). Recording started automatically at sunset and ended at sunrise. Each recording was accompanied by site identification, date, and time automatically provided by the device. The sample rate was set to 384 kHz, and the recording sensitivity to 10 kHz to minimize noise. This limit was below the minimum characteristic frequency of the lowest echolocating species in the region, \u003cem\u003eOtomops martiensseni\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e. The recordings were made in WAV format and stored on SD memory cards. The recordings in the plots of the same triplet always took place during one night. Within the plot, the devices were placed in open places near bushes or trees, depending on the condition of the site, but at a sufficient distance from the canopy to prevent the microphone from being covered by leaves.\u003c/p\u003e \u003cp\u003eThe present paper is based on acoustic records obtained from the standard plots of the MOSAIK project. The dataset contains 94,310 valid recordings representing 130,888 individual bat records: 116,623 sequences of echolocation calls and 14,261 social calls. Here, it should be emphasized that the community record was biased by essential restrictions of the applied method. It provided no information on non-echolocating bats (Pteropodidae), and ground gleaners with weak echolocation (Nycteridae) and, correspondingly, just scarce records on the foliage gleaners detectable at short distances only (Hipposideridae, Rhinolophidae, Rhinonycteridae). Hence, the representation of these groups in our sample was considered rather incidental and not suitable for quantitative comparisons. This data was excluded from analyses. In contrast, quite robust data were obtained for the remaining groups, i.e., Molossidae, Vespertilionidae, and Emballonuridae, which represent (both in terms of species richness and abundance) the true core of local bat communities and compose 97% of all records. The vast majority of our analyses are thus restricted to them.\u003c/p\u003e \u003cp\u003eFor each plot or its close surroundings, a series of variables were recorded. Besides retrieving the remote sensing data (NDVI, EVI, MODIS, etc. mostly referring to a 4 km\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e grid cell; see Hejda et al.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e for details), the MOSAIK team performed an extensive on-ground recording of vegetation cover, plant species diversity, birds, large mammals, insects, and bats. Insects were collected by using night light traps; in the present paper, we used the total mass of insects collected per night as a rough estimate of food resource richness for bats. A complete list of all environmental variables and their abbreviations is provided bellow.\u003c/p\u003e \u003cp\u003eAdditional whole-night recordings, not used in analyses, were undertaken in campsites in Skukuza, Letaba, Mopane, Shingwendzi, Phalaborwa, and Punda Maria. A number of day-time bat roosts were repeatedly censused, mostly in campsites (including colonies of \u003cem\u003eHipposideros caffer\u003c/em\u003e, \u003cem\u003eNycteris grandis\u003c/em\u003e, \u003cem\u003eTaphozous mauritianus\u003c/em\u003e, \u003cem\u003eAfronycteris nana\u003c/em\u003e, \u003cem\u003ePipistrellus hesperidus\u003c/em\u003e, \u003cem\u003ePipistrellus rusticus\u003c/em\u003e, \u003cem\u003eScotophilus dinganii\u003c/em\u003e, \u003cem\u003eChaerephon pumilus\u003c/em\u003e, \u003cem\u003eMops condylurus\u003c/em\u003e, \u003cem\u003eEpomophorus wahlbergi\u003c/em\u003e).\u003c/p\u003e \u003cp\u003e \u003cb\u003eAcoustic analyses and OTUs.\u003c/b\u003e The \u003cem\u003epost hoc\u003c/em\u003e voice identification and visualization of acoustic records were undertaken by using Kaleidoscope Pro software (Wildlife Acoustics). The\u003c/p\u003e \u003cp\u003eidentification procedure applied in this project was based on the cluster analysis method, which classifies a voice record using multivariate comparisons with previously recorded data and a manually edited database of echolocation parameters. We used a database of sonar variables compiled for bats of Kruger National Park (KNP-specific classifier) that was developed in the laboratory of P. Taylor\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e, further supplemented with own recordings of reliably identified bats either mist-netted or those leaving roosts in camps.\u003c/p\u003e \u003cp\u003eRecordings of echolocation or social calls that contained at least two cries were considered valid; the invalid recordings (biased by noise from insects, wind, birds, or poor resolution) were excluded. Species determination was based essentially on the acoustic characteristics of the search phase of the echolocation sequence. Since the automated identification procedure (including the cluster method) does not allow the identification of multiple bat species on a single recording, recordings with multiple species were always manually edited. The cluster analysis method of the Kaleidoscope program splits the recordings with multiple call sequences into separate files, often representing a duplication of calls from a single individual only – then a given individual was counted only once. The distribution of records by cluster analysis was double-checked, and any misidentification was manually corrected.\u003c/p\u003e \u003cp\u003eUsing the above procedures, the echolocation voice records were categorized into 31 acoustic parataxa, taken as operation taxonomic units (OTUs) of subsequent analyses. Twenty-two of them were coidentified with real local species at a high probability level. Nine OTUs with uncertain species affiliations (mostly represented by scarce records only) were treated as separate entities under original working labels (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, and below). Detailed data concerning echolocation characteristics of particular OTUs and post-hoc estimates of possible identification bias are shown in Supplementary File S1.\u003c/p\u003e \u003cp\u003e \u003cb\u003eStatistical analysis: community patterns.\u003c/b\u003e The resulting dataset consisting of the records of particular OTUs in 120 plots representing basic spatiotemporal units (site, date, and season XI or II, respectively) was aggregated to partial subsets by topographic units, the northern area (N: LET and PHA) and the southern area (S: SKU and SAT), spring (IX) and autumn (II) seasons, and habitat (perennial, seasonal, crest). For each, we computed the basic statistics (mean, minimum, maximum, SD, CV (= SD/mean), skewness) and the standard community parameters: abundance (n, N), dominance (D), and frequency (F) of each OTUs, Shannon diversity index (H*) and Pielou evenness (E*), calculated as H’ = Ʃ P\u003csub\u003ei\u003c/sub\u003e*ln(P\u003csub\u003ei\u003c/sub\u003e), where P\u003csub\u003ei\u003c/sub\u003e represents the relative abundance of species i in a given community, and E = H’/log(S), where S represents the number of species within a community\u003csup\u003e\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e). Between-sample comparisons were further supplemented by computation of Simpson 1-D alpha diversity (D = 1- Ʃ n\u003csub\u003ei\u003c/sub\u003e(n\u003csub\u003ei\u003c/sub\u003e-1)/N(N-1)); beta diversity W\u003csup\u003e\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e, and beta T\u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eFurther, we calculated the summary abundances of the families Vespertilionidae and Molossidae and their abundance ratios. The dominance rank of particular species was expressed in terms of Tischler’s scale\u003csup\u003e\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e: eudominant (D \u0026gt; 10%), dominant (10% \u0026gt; D \u0026gt; 5%), subdominant (5% \u0026gt; D \u0026gt; 2%), recedent (2% \u0026gt; D \u0026gt; 1%), subrecedent (1% \u0026gt; D). Conceptual issues concerning core and satellite community elements\u003csup\u003e\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e were taken into account.\u003c/p\u003e \u003cp\u003eThe individual samples and the above-mentioned subsets were mutually compared using a series of univariate, bivariate, and multivariate techniques. Null hypotheses were tested by paired Wilcoxon test, Mann-Whitney U/z test for mean values, F statistics for variance, and Kolmogorov-Smirnov test (D) for distribution (with Monte-Carlo permutation). Paired Wilcoxon tests and Mann-Whitney tests were used to test the differences in individual OTUs. Statistical significances of differences between compared subsets were expressed as respective probabilities (in most instances, for the sake of simplicity, the values of particular test criteria were omitted).\u003c/p\u003e \u003cp\u003eThe between-sample similarity of species composition was quantified by the Jaccard index (Jc = the ratio of the number of species present in both samples to the sum of the species represented in a single sample only), the Bray-Curtis index was used as a measure of dissimilarity in dominance structure (BC = 1–2c\u003csub\u003eij\u003c/sub\u003e/(n\u003csub\u003ei\u003c/sub\u003e+n\u003csub\u003ej\u003c/sub\u003e), where c is the sum of the smaller values of the abundance of species common to samples i and j and n are the total abundances of the samples). Similarities among individual samples, presence, abundance, dominance, and distributional pattern of individual species were further analyzed by a series of multivariate techniques. Regarding the log-normal distribution of abundance data, for most computations, we applied the logarithmic tranSupplementary FileSFormation of abundance data. The matrices of presence, abundance, and dominance records, correlation matrices of abundance data, and similarity matrices of the Jaccard index and Bray-Curtis index were further examined by cluster analyses, principal component analysis (PCA), and correspondence analyses (CA, DCA).\u003c/p\u003e \u003cp\u003ePatterns of variation in community structure were further examined by using nestedness analysis followed by the concepts proposed by Atmar and Patterson\u003csup\u003e\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e (for a recent review, see e.g. Payrató-Borràs et al.\u003csup\u003e\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e\u003c/sup\u003e). Using open software module at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://ecosoft.alwaysdata.net\u003c/span\u003e\u003cspan address=\"http://ecosoft.alwaysdata.net\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, we computed NODF (nestedness metrics based on overlaps and decreasing fills\u003csup\u003e\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e\u003c/sup\u003e, MT (matrix temperature), and extracted graphical outlines of packed community matrix. The matrix temperature\u003csup\u003e\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e expresses a degree of nestedness order in thermodynamic terms, i.e. within a continuum from a perfectly ordered system, absent of all randomness (maximally “cold” state: 0\u003csup\u003e0\u003c/sup\u003e), to the absence of any order (maximally “hot” state:100\u003csup\u003e0\u003c/sup\u003e). NODF (with an inverse scaling), is a measure responding the objections against the uncorrected use of some other nestedness measures, which is considered to be much less prone to effects of taxon and area specificities\u003csup\u003e\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003cb\u003eStatistical analysis: effects of environmental variables.\u003c/b\u003e We examined habitat preferences of particular species regarding vegetation cover (percentages of herb, shrub and tree cover) of the sites where the species was recorded. For each OTU, we computed a mean value of its habitat preferences in terms of ternary representation of herb, shrub, and tree cover (as a sum of site-specific values of vegetation cover multiplied by abundances of given OTU in respective sites, divided by its total abundance). The respective characteristic was expressed as a mean value of the vegetation cover (computed from a sum of particular site values multiplied by species abundance at these sites divided by the total abundance of the species). In the same way, we also computed the distribution of species records in particular site types (crest, seasonal, perennial river).\u003c/p\u003e \u003cp\u003eUsing a dataset of particular categorical variables and large-scale predictors of environmental, climatic, and vegetation conditions at each point (surveyed in detail by Hejda et al.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e and briefly listed in the Methods), we tested their effects on bat species richness, abundance, Shannon diversity, Pielou evenness, and species composition. For that purpose, the predictors were assembled in three sets. The first set, further termed as \"categorical predictors\" included variables reflecting the basic habitat characteristics: habitat type (close to perennial rivers, close to seasonal rivers and on crests, at least 5 km from any waterbody), bedrock (granite, basalt) and landsystem (Skukuza, Satara, Phalaborwa, Lethaba), North versus South of KNP and presence/absence of Mopane as a dominant woody species. The second set of predictors, further termed as “vegetation parameters”, included basic characteristics of vegetation, such as herbal, grass, shrub and tree species richness and cover. Further, we included a share of typical “savanna” species with potential conservation value, separately for herbs, shrubs, and trees. To account for the gradients in plant composition, we calculated an indirect gradient ordination analysis (PCA) and then included the first three ordination axis (PCO1, PCO2 and PCO3 – see below for explanation) among the predictors representing the character of vegetation. The third set of predictors represented a broader environmental context and included variables such as distance to park boundaries, distances to roads, tracks and camps, distance to rivers and streams, climatic characteristics (temperature, precipitation), and fire history and frequency. See below for the list of all included predictors.\u003c/p\u003e \u003cp\u003eThe data were analyzed using both univariate and multivariate models. In particular, the data with the bat species richness, Shannon diversity, Pielou evenness, and total abundances as a response variable were analyzed using univariate methods, while data on the bat species composition were analyzed using multivariate direct gradient ordination models (RDA).\u003c/p\u003e \u003cp\u003eWe tested the effects of all categorical predictors, along with their interactions with “habitat type”. We applied regression tree models and indirect gradient ordination analysis as exploratory tools to select potentially important predictors of species richness, diversity, evenness and abundance among all continuous variables (vegetation parameters, environmental context). The categorical predictors were also included in the regression tree models to show possible interactions with continuous predictors. In regression tree models, the categorical variables, vegetation parameters and variables of environmental context were set as predictors, while bat species richness, abundance, diversity and evenness were set as response variables in individual regression tree models. The effects of predictors, identified as important by the regression tree models, were then tested by linear mixed-effect models. The regression tree models were created in the R software, using the package “rpart”\u003csup\u003e\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eFurther, the relations between predictors (categorical variables, vegetation parameters, environmental context) and univariate response variables (bat species richness, abundance, diversity and evenness) were explored using indirect gradient ordination models, where all variables of interest were used as individual response variables in the multi-dimensional space. Ordination plots were used to visually estimate the relations between different variables and to identify predictors related to bat species richness, diversity, evenness and abundance. Similarly to the regression tree models, the effects of variables (categorical, vegetation, environmental context) that were identified as closely related to bat species richness, abundance, diversity and evenness were then tested by linear mixed effect models. The exploratory indirect gradient ordination analyses were performed using the CANOCO5 software\u003csup\u003e\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn the mixed effect models, the predictors of interest (categorical, vegetation parameters, environmental context) were set as fixed effects, while the identity of triplets (representing the hierarchy of the sampling design) was set as a random effect. The normality of individual response variables (bat species richness, abundance, diversity and evenness) as well as that of the predictors (continuous variables among vegetation parameters and environmental context) was tested using the Shapiro-Wilk tests. Square-root, log and arcsine tranformations were applied to improve normality in case the Shapiro-Wilk tests revealed significant deviations from it. The significance of individual main effects and their interactions was estimated using the deletion tests, which compare the explanatory power of models with and without a particular term or interaction\u003csup\u003e\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e\u003c/sup\u003e. The significance of quadratic terms of all continuous predictors was also tested by deletion tests. The quality of the parsimonous model (= with only significant terms and interactions) was checked by testing the normality of residuals and also visually, by inspecting the normal-probability plots. The linear mixed-effect models were created in the R software, using the package \"nlme\"\u003csup\u003e\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e\u003c/sup\u003e. The differences between individual levels of all categorical predictors were tested by the Tukey HSD method, using the package “emmeans” of the R software.\u003c/p\u003e \u003cp\u003eThe data on the composition of bats were examined using direct gradient ordination methods (RDA and CCA), with the abundances of individual bat species set as response variables in the multivariate ordination space and the categorical variables, variables expressing the vegetation parameters and environmental context as predictors. The significance of predictors was tested using the Monte-Carlo permutation tests with 499 permutations. To account for the possible autocorrelations given by the hierarchical sampling desing, a split-plot permutation scheme was applied. The triplets were set as whole-plots, while individual sampling sites (plots) were set as split-plots. Both whole-plot and split-plot level were permuted freely, as some of the predictors (habitat: perennial rivers, seasonal rivers, crests; mopane cover etc.) were defined at the split-plot level, while others (bedrock, landsystem, north-south) were defined at the whole-plot level.\u003c/p\u003e \u003cp\u003eFirst, complex ordination models with the whole group of predictors (categorical, vegetation paremeters, environmental context) were created to estimate the overal explanatory power of these three groups of predictors. Further, forward selection procedure was applied to identify the strongest predictors within each group. The results were visualized using ordination plots. The choice between a linear and a unimodal model (RDA or CCA, resp.) was decided based on the main gradient within the data. All direct gradient ordination analyses were performed in the CANOCO 5 software\u003csup\u003e\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe between-season differences and the season related effects were not analysed in the present paper – these topics will be surveyed separately elsewhere.\u003c/p\u003e \u003cp\u003eDatabase operations were performed in Microsoft Excel, the basic statistics were computed in STATISTICA 13, and/or in Past, and SAM software, LME and regression tree analyses were computed in Rstudio, the ordination analyses mostly in CANOCO.\u003c/p\u003e\n\n\n\n\n\n \n\n"},{"header":"List of the environmental predictors used in the present study and Abbreviations","content":"\u003cp\u003eWe largerly used the dataset assembled by the MOSAIK project team and adopted completely all particular variables and their values. For detailed explanation of all particular variables (including their primary sources etc.) see Hejda et al.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003ch3\u003eCategorial factors\u003c/h3\u003e\u003cp\u003e \u003cb\u003eN, S\u003c/b\u003e: Northern vs. Southern part of the region, * \u003cb\u003eBedrock\u003c/b\u003e: granite vs. basalt, * \u003cb\u003eHabitat type or plot\u003c/b\u003e/s\u003cb\u003eite category\u003c/b\u003e (C – crest with no direct acces per 4 km\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e grid cell ter, S – seasonal river, P – perennial river), * \u003cb\u003etriplet\u003c/b\u003e: neighbouring plots within the same triplet vs. distant plots, * \u003cb\u003eseason\u003c/b\u003e: II (February) vs. XI (November)\u003c/p\u003e\u003ch3\u003eLarge-scale predictors of environmental context\u003c/h3\u003e\u003cp\u003e(referred mostly to mean values within 4km\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e grid cell surrounding the site point)\u003c/p\u003e\u003cp\u003e \u003cb\u003efireSum\u003c/b\u003e: Fire frequency, total number of fires from 2000 to 2019; \u003cb\u003e* fireMean\u003c/b\u003e: Mean number of fires, Pixel-wise mean of the number of fires recorded over the long-term. * \u003cb\u003efireSD\u003c/b\u003e: StdDev of number of fires * \u003cb\u003eeviSum\u003c/b\u003e: Sum EVI, Long-term sum of Enhanced Vegetation Index (EVI). * \u003cb\u003eeviMean\u003c/b\u003e: Mean EVI, Long-term mean of Enhanced Vegetation Index (EVI). * \u003cb\u003eeviSD\u003c/b\u003e: StdDev EVI, Long-term standard deviation of Enhanced Vegetation Index (EVI). * \u003cb\u003erainSum\u003c/b\u003e: Sum rainfall, Long-term sum of all rainfall. * \u003cb\u003erainMean\u003c/b\u003e: Mean rainfall, Long-term mean of all rainfall. * \u003cb\u003erainSD\u003c/b\u003e: StdDev of rainfall, Long-term standard deviation of all rainfall. *\u003cb\u003etempMean\u003c/b\u003e: Mean temperature, Long-term mean of temperature (°C). * \u003cb\u003etempSD\u003c/b\u003e: StdDev of temperature, Long-term standard deviation of temperature (°C). * \u003cb\u003etempMin\u003c/b\u003e: Minimum temperature, Long-term mean of minimum temperature (°C). * \u003cb\u003etempMax\u003c/b\u003e: Maximum temperature, Long-term mean of maximum temperature (°C). * \u003cb\u003ewaterSum\u003c/b\u003e: Sum surface water occurrence density, Long-term sum of surface water occurrence, expressed as kernel density estimates (KDE). Annual density of water extent, i.e. weighed by % re-occurrence of water each year; * \u003cb\u003eWaterMean\u003c/b\u003e: Mean surface water occurrence density, Long-term mean of surface water occurrence, expressed as kernel density estimates (KDE). Annual density of water extent, i.e. weighed by % re-occurrence of water each year; * \u003cb\u003ewaterSD\u003c/b\u003e: StdDev Long-term standard deviation surface water occurrence, expressed as kernel density estimates (KDE). Annual density of water extent, i.e. weighed by % re-occurrence of water each year; * \u003cb\u003edistBnd\u003c/b\u003e: Distance to KNP boundary. * \u003cb\u003edistCamp\u003c/b\u003e: Distance to all campsite. * \u003cb\u003edistTar\u003c/b\u003e: Distance to all tarred roads. * \u003cb\u003edistDirt\u003c/b\u003e: Distance to all gravel roads. * \u003cb\u003edistRiv\u003c/b\u003e: Distance to all major rivers. * \u003cb\u003edistStrm\u003c/b\u003e: Distance to all rivers and streams. * \u003cb\u003eLatitude\u003c/b\u003e: Latitudinal GPS coordinates of the site * \u003cb\u003eLongitude\u003c/b\u003e: Longitudinal GPS coordinates of the site. * \u003cb\u003eCampDist\u003c/b\u003e: Distance to the nearest campsite\u003c/p\u003e\u003ch2\u003eLarge scale predictors of vegetation patterns\u003c/h2\u003e\u003cp\u003e(based on ground analyses by the MOSAIK team)\u003c/p\u003e\u003cp\u003e \u003cb\u003evegPCO1, vegPCO2, veg PCO3\u003c/b\u003e - loading values of first three components of multivariate factor analysis of all vegetation variables: \u003cb\u003ePCO1 -\u003c/b\u003e poor, open-ground vegetation, mostly on granite substrate, \u003cb\u003ePCO2 -\u003c/b\u003e shaded grounds with deep soil, high productivity, \"sweet veld\", \u003cb\u003ePCO3 –\u003c/b\u003e temporarly disturbed, largely mesic vegetation, related to seasonal rivers, * \u003cb\u003evegCoverHerbs, *vegCoverShrub *vegCoverTrees *vegCoverGrass -\u003c/b\u003e cover of particular vegetation guilds, \u003cb\u003evegSPtotal -\u003c/b\u003e total species richness at the site, \u003cb\u003e*vegSPHerbs, *vegSPShrub *vegSP Trees -\u003c/b\u003e species richness of particular vegetation guilds, \u003cb\u003evegH* -\u003c/b\u003e Shannon diversity of vegetation at the site, \u003cb\u003evegHherbs *vegHshrubs *vegHtrees -\u003c/b\u003e Shannon diversity of particular vegetation guilds, \u003cb\u003e*vegAlien\u003c/b\u003e - number of alien plant species, \u003cb\u003e*savHerb *savShrub *savTree\u003c/b\u003e - number of typical savanna elements, \u003cb\u003e* Mopane -\u003c/b\u003e cover of mopane \u003cem\u003e(Colophospermum mopane).\u003c/em\u003e\u003c/p\u003e\u003ch3\u003eList of OTUs\u003c/h3\u003e\u003cp\u003e \u003cb\u003eCHAANS\u003c/b\u003e – \u003cem\u003eChaerephon ansrogei *\u003c/em\u003e \u003cb\u003eCHAPUM\u003c/b\u003e – \u003cem\u003eChaerephon pumilus *\u003c/em\u003e \u003cb\u003eEPTHOT\u003c/b\u003e – \u003cem\u003eEptesicus hottentotus\u003c/em\u003e (but see Discussion and Supplementary File S1) \u003cem\u003e*\u003c/em\u003e \u003cb\u003eMINNAT\u003c/b\u003e – \u003cem\u003eMiniopterus natalensis *\u003c/em\u003e \u003cb\u003eMolossid45\u003c/b\u003e – parataxon not afiliated to any real species, with echolocation characteristics resembling those in molosid, yet with fpeak 45 kHz * \u003cb\u003eMOPCON\u003c/b\u003e – \u003cem\u003eMops condylurus *\u003c/em\u003e \u003cb\u003eMOPMID\u003c/b\u003e – \u003cem\u003eMops midas *\u003c/em\u003e \u003cb\u003eMyotisSp\u003c/b\u003e – parataxon not afiliated to any real species, with echolocation characteristics of the genus \u003cem\u003eMyotis *\u003c/em\u003e \u003cb\u003eMYOTRI\u003c/b\u003e – \u003cem\u003eMyotis tricolor *\u003c/em\u003e \u003cb\u003eNEOCAP\u003c/b\u003e – \u003cem\u003eLaephotis capensis *\u003c/em\u003e \u003cb\u003eNEONAN\u003c/b\u003e \u003cem\u003e– Afronycteris nana *\u003c/em\u003e \u003cb\u003eNYCSCH\u003c/b\u003e – \u003cem\u003eNycticeinops schlieffeni *\u003c/em\u003e \u003cb\u003eOTOMAR\u003c/b\u003e – \u003cem\u003eOtomops martiensseni *\u003c/em\u003e \u003cb\u003ePIPHES\u003c/b\u003e – \u003cem\u003ePipistrellus hesperidus *\u003c/em\u003e \u003cb\u003ePIPRUS\u003c/b\u003e – \u003cem\u003ePipistrellus rusticus *\u003c/em\u003e \u003cb\u003eSCODIN\u003c/b\u003e – \u003cem\u003eScotophilus dinganii *\u003c/em\u003e \u003cb\u003eTADAEG\u003c/b\u003e – \u003cem\u003eTadarida aegyptiaca *\u003c/em\u003e \u003cb\u003eTAPMAU\u003c/b\u003e – \u003cem\u003eTaphozous mauritianus *\u003c/em\u003e \u003cb\u003eRHICAP\u003c/b\u003e – \u003cem\u003eRhinolophus darlingi *\u003c/em\u003e \u003cb\u003eRHI105\u003c/b\u003e – \u003cem\u003eRhinolophus landeri *\u003c/em\u003e \u003cb\u003eRHIFUM\u003c/b\u003e – \u003cem\u003eRhinolophus fumigatus *\u003c/em\u003e \u003cb\u003eRHISIM\u003c/b\u003e – \u003cem\u003eRhinolophus simulator *\u003c/em\u003e \u003cb\u003eRHIRHO\u003c/b\u003e – \u003cem\u003eRhinolphus swinnyi *\u003c/em\u003e \u003cb\u003eHIPCAF (RHICAF)\u003c/b\u003e – \u003cem\u003eHipposideros caffer *\u003c/em\u003e \u003cb\u003eUNK75 (75)\u003c/b\u003e – parataxon not afiliated to any real species, with fpeak around 75 kHz. * \u003cb\u003e37\u003c/b\u003e – parataxon not affiliated to any real species, with fpeak around 37 kHz. * \u003cb\u003e35\u003c/b\u003e – parataxon not affiliated to any real species, with fpeak around 35 kHz. * \u003cb\u003especies1\u003c/b\u003e – parataxon not affiliated to any real species, with fpeak around 20 kHz. * \u003cb\u003eNoID\u003c/b\u003e – unidentified echolocation signals\u003c/p\u003e\n\u003ch3\u003eAbbreviations\u003c/h3\u003e\n\u003cp\u003e\u003cstrong\u003eavg\u003c/strong\u003e average\u0026nbsp; \u0026nbsp; \u0026nbsp;* \u003cstrong\u003eBC\u003c/strong\u003e Bray-Curtis measure of dissimilarity in dominance structure * \u003cstrong\u003eC, S, P sites\u003c/strong\u003e: crest, seasonal river, perennial river * \u003cstrong\u003eCF\u003c/strong\u003e - constant frequency * \u003cstrong\u003eCV\u003c/strong\u003e coffeficient of variation (=SD / avg) * \u003cstrong\u003eD\u003c/strong\u003e - dominance (D=100(n/N)* \u003cstrong\u003eE*\u003c/strong\u003e evenness * \u003cstrong\u003eFM\u003c/strong\u003e -frequency modulated * \u003cstrong\u003eGLM\u003c/strong\u003e generalized linear model * \u003cstrong\u003eGLZ\u003c/strong\u003e generalized parametrized linear models * \u003cstrong\u003eH*\u003c/strong\u003e Shannon diversity index * \u003cstrong\u003eJc\u003c/strong\u003e Jaccard index * \u003cstrong\u003eKNP\u003c/strong\u003e - \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Kruger National Park * \u003cstrong\u003eLET\u003c/strong\u003e Letaba landsystem * \u003cstrong\u003en\u003c/strong\u003e - abundance (number of species records) *N - abundance (total number of all records) * \u003cstrong\u003ep\u003c/strong\u003e probability (p\u0026lt;0.05 taken as significant) * \u003cstrong\u003ePH\u003c/strong\u003e Phalaborwa landsystem * \u003cstrong\u003eqCF\u003c/strong\u003e - quasi-constant frequency * \u003cstrong\u003er\u003c/strong\u003e Pearson\u0026rsquo;s coefficient of correlation * \u003cstrong\u003eR2\u003c/strong\u003e squared coefficient of correlation * \u003cstrong\u003eSAT\u003c/strong\u003e Satara landsystem * \u003cstrong\u003eSD\u003c/strong\u003e statistical deviation * \u003cstrong\u003eSK\u003c/strong\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Skukuza landsystem\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\n\u003cp\u003eThe MOSAIK project was supported by grant no. 18-18495S (Czech Science Foundation), long-term research development project RVO 67985939 (Czech Academy of Sciences), and internal project no. UNCE204069 (Charles University). The project was registered as PYSK1432 with SANParks. Support from SANParks during the field visits was greatly appreciated. We thank our guards Obert Mathebula, Thomas Rikombe, Desmond Mabaso, Herman Ntimane, Annoit Mashele, Isaac Sedibe, Priska Rikombe, and Velly Ndlovu for keeping us safe in the field. \u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eAuthors contributions\u003c/strong\u003e\u003c/p\u003e\n\n\u003cp\u003eIH, PP, KP, DS, LCF and SMF conceived the idea, MS, IH and SD collected the bat data, ERB, DMP, PJT, SMW provided a comparative database of bat echolocation calls, SD, JČ, MH, KP, PP, and RT collected the data on the environmental variables and other contextual information, MS, IH and MH analyzed the data, IH and MS wrote the first draft of the paper, and all authors commented on the manuscript and gave final approval for its publication.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eCompeting interests \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eData Availability Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data are available in a public depository: https://github.com/IvanHoracek/KNP.git.\u003c/p\u003e\n\n"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSankaran, M. et al. Determinants of woody cover in African savannas. \u003cem\u003eNature\u003c/em\u003e\u003cstrong\u003e438\u003c/strong\u003e, 846\u0026ndash;849 (2005).\u003c/li\u003e\n\u003cli\u003eStaver, A. C., Archibald, S. \u0026amp; Levin, S. A. The global extent and determinants of savanna and forest as alternative biome states. \u003cem\u003eScience\u003c/em\u003e\u003cstrong\u003e334\u003c/strong\u003e, 230\u0026ndash;232 (2011). \u003c/li\u003e\n\u003cli\u003eMucina, L. \u0026amp; Rutherford, M. C. \u003cem\u003eThe Vegetation of South Africa, Lesotho and Swaziland\u003c/em\u003e (South African National Biodiversity Institute, Pretoria, 2006).\u003c/li\u003e\n\u003cli\u003eDevine, A. P., McDonald, R. A., Quaife, T. \u0026amp; Maclean, I. Determinants of woody encroachment and cover in African savannas. \u003cem\u003eOecologia\u003c/em\u003e\u003cstrong\u003e183\u003c/strong\u003e, 939\u0026ndash;951 (2017). \u003c/li\u003e\n\u003cli\u003ePy\u0026scaron;ek, P. et al. Into the great wide open: Do alien plants spread from rivers to dry savanna in the Kruger National Park? \u003cem\u003eNeoBiota\u003c/em\u003e\u003cstrong\u003e60\u003c/strong\u003e, 61\u0026ndash;77 (2020). \u003c/li\u003e\n\u003cli\u003eDelabye, S. et al. Thirteen moth species (Lepidoptera, Erebidae, Noctuidae) newly recorded in South Africa, with comments on their distribution. \u003cem\u003eBiodiv. Data J.\u003c/em\u003e\u003cstrong\u003e10\u003c/strong\u003e, e89729 (2022). \u003c/li\u003e\n\u003cli\u003eHejda, M. et al. Water availability, bedrock, disturbance by herbivores, and climate determine plant diversity in South-African savanna. \u003cem\u003eSci. Rep.\u003c/em\u003e\u003cstrong\u003e12\u003c/strong\u003e, 1\u0026ndash;19 (2022). \u003c/li\u003e\n\u003cli\u003eČuda, J. et al. Habitat modifies the relationship between grass and herbivore species richness in a South African savanna. \u003cem\u003eEcol. Evol.\u003c/em\u003e\u003cstrong\u003e14\u003c/strong\u003e, e11167 (2024). \u003c/li\u003e\n\u003cli\u003eDu Toit, J. T. Large herbivores and savanna heterogeneity. In \u003cem\u003eThe Kruger Experience: Ecology and Management of Savanna Heterogeneity\u003c/em\u003e (eds du Toit, J. T., Rogers, K. H. \u0026amp; Biggs, H. C.) 292\u0026ndash;309 (Island Press, Washington, 2003).\u003c/li\u003e\n\u003cli\u003eBucini, G., Saatchi, S., Hanan, N., Boone, R. B. \u0026amp; Smit, I. Woody cover and heterogeneity in the savannas of the Kruger National Park, South Africa. \u003cem\u003eIEEE\u003c/em\u003e\u003cstrong\u003e4\u003c/strong\u003e, 334\u0026ndash;337 (2009). \u003c/li\u003e\n\u003cli\u003eSmit, I. P. \u0026amp; Prins, H. H. Predicting the effects of woody encroachment on mammal communities, grazing biomass and fire frequency in African savannas. \u003cem\u003ePLoS ONE\u003c/em\u003e\u003cstrong\u003e10\u003c/strong\u003e, e0137857 (2015). \u003c/li\u003e\n\u003cli\u003eLoggins, A. A., Shrader, A. M., Monadjem, A. \u0026amp; McCleery, R. A. Shrub cover homogenizes small mammals\u0026rsquo; activity and perceived predation risk. \u003cem\u003eSci. Rep.\u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 1\u0026ndash;11 (2019). \u003c/li\u003e\n\u003cli\u003eWessels, K. J. et al. Relationship between herbaceous biomass and 1km\u003csup\u003e2\u003c/sup\u003e advanced very high resolution radiometer (AVHRR) NDVI in Kruger National Park, South Africa. \u003cem\u003eInt. J. Remote Sens.\u003c/em\u003e\u003cstrong\u003e27\u003c/strong\u003e, 951\u0026ndash;973 (2006). \u003c/li\u003e\n\u003cli\u003eUrban, M. et al. Surface moisture and vegetation cover analysis for drought monitoring in the southern Kruger National Park using Sentinel-1, Sentinel-2, and Landsat-8. \u003cem\u003eRemote Sens.\u003c/em\u003e\u003cstrong\u003e10\u003c/strong\u003e, 1482 (2018). \u003c/li\u003e\n\u003cli\u003eMacFadyen, S., Hui, C., Verburg, P. H. \u0026amp; Van Teeffelen, A. J. Quantifying spatiotemporal drivers of environmental heterogeneity in Kruger National Park, South Africa. \u003cem\u003eLandsc. Ecol.\u003c/em\u003e\u003cstrong\u003e31\u003c/strong\u003e, 2013\u0026ndash;2029 (2016). \u003c/li\u003e\n\u003cli\u003eMacFadyen, S. Linking long-term patterns of landscape heterogeneity to changing ecosystem processes in the Kruger National Park, South Africa (Doctoral dissertation, Stellenbosch University) http://hdl.handle.net/10019.1/105154 (2018).\u003c/li\u003e\n\u003cli\u003eMacFadyen, S., Hui, C., Verburg, P. H. \u0026amp; Van Teeffelen, A. J. Spatiotemporal distribution dynamics of elephants in response to density, rainfall, rivers and fire in Kruger National Park, South Africa. \u003cem\u003eDiversity Distrib.\u003c/em\u003e\u003cstrong\u003e25\u003c/strong\u003e, 880\u0026ndash;894 (2019). \u003c/li\u003e\n\u003cli\u003eMalherbe, J., Smit, I. P., Wessels, K. J. \u0026amp; Beukes, P. J. Recent droughts in the Kruger National Park as reflected in the extreme climate index. \u003cem\u003eAfr. J. Range For. Sci.\u003c/em\u003e\u003cstrong\u003e37\u003c/strong\u003e, 1\u0026ndash;17 (2020).\u003c/li\u003e\n\u003cli\u003eCumming, D. H. et al. Elephants, woodlands and biodiversity in southern Africa. \u003cem\u003eS. Afr. J. Sci\u003c/em\u003e. \u003cstrong\u003e93\u003c/strong\u003e, 231\u0026ndash;236 (1997).\u003c/li\u003e\n\u003cli\u003eFenton, M. B. et al. Bats and the loss tree canopy in African Woodlands. \u003cem\u003eConserv. Biol.\u003c/em\u003e\u003cstrong\u003e12\u003c/strong\u003e, 399\u0026ndash;407 (1998).\u003c/li\u003e\n\u003cli\u003eWilkinson, D. M., Midgley, J. J. \u0026amp; Cunningham, A. B. Constraints, crashes and conservation: Were historical African savanna elephants \u003cem\u003eLoxodonta africana\u003c/em\u003e densities relatively high or lower than those seen in protected areas today? \u003cem\u003ePlant Ecol. Divers.\u003c/em\u003e\u003cstrong\u003e15\u003c/strong\u003e, 1\u0026ndash;2, 1\u0026ndash;11 (2022). \u003c/li\u003e\n\u003cli\u003eRautenbach, I. L., Schlitter, D. A. \u0026amp; Braack, L. E. O. New distributional records of bats for the Republic of South Africa, with special reference to the Kruger National Park. \u003cem\u003eKoedoe\u003c/em\u003e\u003cstrong\u003e27\u003c/strong\u003e, 131\u0026ndash;135 (1984). \u003c/li\u003e\n\u003cli\u003eRautenbach, I. L., Fenton, M. B. \u0026amp; Braack, L. E. O. First records of five species of insectivorous bats from the Kruger National Park. \u003cem\u003eKoedoe\u003c/em\u003e\u003cstrong\u003e28\u003c/strong\u003e, 73\u0026ndash;80 (1985). \u003c/li\u003e\n\u003cli\u003eVan der Merwe, M., Van der Merwe, N. J. \u0026amp; Penzhorn, B. L. Aspects of reproduction in the seasonally breeding African yellow bat, \u003cem\u003eScotophilus dinganii\u003c/em\u003e (A. Smith, 1833). \u003cem\u003eAfr. Zool\u003c/em\u003e. \u003cstrong\u003e41\u003c/strong\u003e, 67\u0026ndash;74 (2006). \u003c/li\u003e\n\u003cli\u003eRautenbach, I. L., Whiting, M. J. \u0026amp; Fenton, M. B. Bats in riverine forests and woodlands: A latitudinal transect in southern Africa. \u003cem\u003eCan. J. Zool.\u003c/em\u003e\u003cstrong\u003e74\u003c/strong\u003e, 312\u0026ndash;322 (1996). \u003c/li\u003e\n\u003cli\u003eTaylor, P. J., Schoeman, M. C. \u0026amp; Monadjem, A. Diversity of bats in the Soutpansberg and Blouberg Mountains of northern South Africa: Complementarity of acoustic and non-acoustic survey methods. \u003cem\u003eS. Afr. J. Wildl. Res\u003c/em\u003e. \u003cstrong\u003e43\u003c/strong\u003e, 12\u0026ndash;26 (2013). \u003c/li\u003e\n\u003cli\u003eTaylor, P. J., Nelufule, M., Parker, D. M., Toussaint, D. C. \u0026amp; Weier, S. M. The Limpopo River exerts a powerful but spatially limited effect on bat communities in a semiarid region of South Africa. \u003cem\u003eActa Chiropterol.\u003c/em\u003e\u003cstrong\u003e22\u003c/strong\u003e, 75\u0026ndash;86 (2020). \u003c/li\u003e\n\u003cli\u003eMonadjem, A., Shapiro, J. T., Mtsetfwa, F., Reside, A. E. \u0026amp; McCleery, R. A. Acoustic call library and detection distances for bats of Swaziland. \u003cem\u003eActa Chiropterol.\u003c/em\u003e\u003cstrong\u003e19\u003c/strong\u003e, 175\u0026ndash;187 (2017). \u003c/li\u003e\n\u003cli\u003eMonadjem, A. et al. Cryptic diversity in the genus Miniopterus with the description of a new species from southern Africa. \u003cem\u003eActa Chiropterol.\u003c/em\u003e\u003cstrong\u003e22\u003c/strong\u003e, 1\u0026ndash;19 (2020). \u003c/li\u003e\n\u003cli\u003eBrinkley, E. R., Weier, S. M., Parker, D. M. \u0026amp; Taylor, P. J. Three decades later in the northern Kruger National Park: Multiple acoustic and capture surveys may underestimate the true local richness of bats based on historical collections. \u003cem\u003eHystrix It. J. Mamm\u003c/em\u003e. \u003cstrong\u003e32\u003c/strong\u003e, 109\u0026ndash;117 (2021). \u003c/li\u003e\n\u003cli\u003eSchoeman, M. C., Cotterill, F. P. D., Taylor, P. J. \u0026amp; Monadjem, A. Using potential distributions to explore environmental correlates of bat species richness in southern Africa: Effects of model selection and taxonomy. \u003cem\u003eCurr. Zool.\u003c/em\u003e\u003cstrong\u003e59\u003c/strong\u003e, 279\u0026ndash;293 (2013). \u003c/li\u003e\n\u003cli\u003eCooper-Bohannon, R. et al. Predicting bat distributions and diversity hotspots in southern Africa. \u003cem\u003eHystrix It. J. Mamm.\u003c/em\u003e\u003cstrong\u003e27\u003c/strong\u003e, 1\u0026ndash;11 (2016).\u003c/li\u003e\n\u003cli\u003eHerkt, K. M. B., Barnikel, G., Skidmore, A. K. \u0026amp; Fahr, J. A high-resolution model of bat diversity and endemism for continental Africa. \u003cem\u003eEcol. Modell\u003c/em\u003e. \u003cstrong\u003e320\u003c/strong\u003e, 9\u0026ndash;28 (2016). \u003c/li\u003e\n\u003cli\u003eSmith, A. et al. Synergistic effects of climate and land-use change on representation of African bats in priority conservation areas. \u003cem\u003eEcol. Indic.\u003c/em\u003e\u003cstrong\u003e69\u003c/strong\u003e, 276\u0026ndash;283 (2016).\u003c/li\u003e\n\u003cli\u003eMonadjem, A., Conenna, I., Taylor, P. J. \u0026amp; Schoeman, M. C. Species richness patterns and functional traits of the bat fauna of arid southern Africa. \u003cem\u003eHystrix It. J. Mamm\u003c/em\u003e. \u003cstrong\u003e29\u003c/strong\u003e, 19\u0026ndash;24 (2018).\u003c/li\u003e\n\u003cli\u003eSchoeman, M. C. \u0026amp; Monadjem, A. Community structure of bats in the savannas of southern Africa: Influence of scale and human land use. \u003cem\u003eHystrix It. J. Mamm\u003c/em\u003e. \u003cstrong\u003e29\u003c/strong\u003e, 3\u0026ndash;10 (2018). \u003c/li\u003e\n\u003cli\u003eMedellin, R. A. \u0026amp; Redford, K. H. The role of mammals in neotropical forest-savanna boundaries. In \u003cem\u003eNature and Dynamics of Forest-Savanna Boundaries\u003c/em\u003e (eds Furley, P. A., Proctor, J. \u0026amp; Ratter, J. A.) 519\u0026ndash;548 (Chapman and Hall, London, 1992).\u003c/li\u003e\n\u003cli\u003eAguirre, L. F. Structure of a Neotropical savanna bat community. \u003cem\u003eJ. Mammal. \u003c/em\u003e\u003cstrong\u003e83\u003c/strong\u003e, 775\u0026ndash;784 (2002).\u003c/li\u003e\n\u003cli\u003eAguirre, L. F., Lens, L., Van Damme, R. \u0026amp; Matthysen, E. Consistency and variation in the bat assemblages inhabiting two forest islands within a Neotropical savanna in Bolivia. \u003cem\u003eJ. Trop. Ecol\u003c/em\u003e. \u003cstrong\u003e19\u003c/strong\u003e, 367\u0026ndash;374 (2003). \u003c/li\u003e\n\u003cli\u003eBernard, E. \u0026amp; Fenton, M. B. Bats in a fragmented landscape: Species composition, diversity and habitat interactions in savannas of Santar\u0026eacute;m, Central Amazonia, Brazil. \u003cem\u003eBiol. Conserv.\u003c/em\u003e\u003cstrong\u003e134\u003c/strong\u003e, 332\u0026ndash;343 (2007). \u003c/li\u003e\n\u003cli\u003eLarrea-Alc\u0026aacute;zar, D. M. et al. Spatial patterns of biological diversity in a neotropical lowland savanna of northeastern Bolivia. \u003cem\u003eBiodivers. Conserv.\u003c/em\u003e\u003cstrong\u003e20\u003c/strong\u003e, 1167\u0026ndash;1182 (2011). \u003c/li\u003e\n\u003cli\u003ede Oliveira, H. F., de Camargo, N. F., Gager, Y. \u0026amp; Aguiar, L. M. The response of bats (Mammalia: Chiroptera) to habitat modification in a Neotropical Savannah. \u003cem\u003eTropic. Conserv. Sci\u003c/em\u003e. \u003cstrong\u003e10\u003c/strong\u003e, 1\u0026ndash;14 (2017). \u003c/li\u003e\n\u003cli\u003eLima, C. S., Varzinczak, L. H. \u0026amp; Passos, F. C. Richness, diversity and abundance of bats from a savanna landscape in central Brazil. \u003cem\u003eMammalia\u003c/em\u003e\u003cstrong\u003e81\u003c/strong\u003e, 33\u0026ndash;40 (2017). \u003c/li\u003e\n\u003cli\u003eMorales-Mart\u0026iacute;nez, D. M., Rodr\u0026iacute;guez-Posada, M. E., Fern\u0026aacute;ndez-Rodr\u0026iacute;guez, C., Calder\u0026oacute;n-Capote, M. C. \u0026amp; Guti\u0026eacute;rrez-Sanabria, D. R. Spatial variation of bat diversity between three floodplain-savanna ecosystems of the Colombian Llanos. \u003cem\u003eTherya\u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 41\u0026ndash;52 (2018). \u003c/li\u003e\n\u003cli\u003eGelderblom, C., Bronner, G., Lombard, A. \u0026amp; Taylor, P. J. Patterns of distribution and current protection status of the Carnivora, Chiroptera and Insectivora in South Africa. \u003cem\u003eS. Afr. J. Zool.\u003c/em\u003e\u003cstrong\u003e30\u003c/strong\u003e, 103\u0026ndash;114 (1995).\u003c/li\u003e\n\u003cli\u003eMonadjem, A. \u0026amp; Reside, A. The influence of riparian vegetation on the distribution and abundance of bats in an African savanna. \u003cem\u003eActa Chiropterol\u003c/em\u003e. \u003cstrong\u003e10\u003c/strong\u003e, 339\u0026ndash;348 (2008). \u003c/li\u003e\n\u003cli\u003eSchoeman, M. C. \u0026amp; Jacobs, D. S. The relative influence of competition and prey defenses on the phenotypic structure of insectivorous bat ensembles in southern Africa. \u003cem\u003ePLoS ONE\u003c/em\u003e\u003cstrong\u003e3\u003c/strong\u003e, e3715 (2008). \u003c/li\u003e\n\u003cli\u003eSchoeman, M. C. \u0026amp; Waddington, K. J. Do deterministic processes influence the phenotypic patterns of animalivorous bat ensembles at urban rivers? \u003cem\u003eAfr. Zool\u003c/em\u003e. \u003cstrong\u003e46\u003c/strong\u003e, 288\u0026ndash;301 (2011).\u003c/li\u003e\n\u003cli\u003eArita, H. T. \u0026amp; Fenton, M. B. Flight and echolocation in the ecology and evolution of bats. \u003cem\u003eTrends Ecol. Evol\u003c/em\u003e. \u003cstrong\u003e12\u003c/strong\u003e, 53\u0026ndash;58 (1997). \u003c/li\u003e\n\u003cli\u003eFindley, J. S. Phenetic packing as a measure of faunal diversity. \u003cem\u003eAm. Nat.\u003c/em\u003e\u003cstrong\u003e107\u003c/strong\u003e, 580\u0026ndash;584 (1973). \u003c/li\u003e\n\u003cli\u003ePatterson, B. D., Willig, M. R. \u0026amp; Stevens, R. D. Trophic strategies, niche partitioning, and patterns of ecological organization. In \u003cem\u003eBat Ecology\u003c/em\u003e (eds Kunz, T. H. \u0026amp; Fenton, M. B.) 536\u0026ndash;557 (University Chicago Press, 2003).\u003c/li\u003e\n\u003cli\u003eWilson, M. V. \u0026amp; Shmida, A. Measuring beta-diversity with presence absence data. \u003cem\u003eJ. Ecol.\u003c/em\u003e\u003cstrong\u003e72\u003c/strong\u003e, 1055\u0026ndash;1064 (1984).\u003c/li\u003e\n\u003cli\u003eMonadjem, A., Taylor, P. J. \u0026amp; Schoeman, M. C. \u003cem\u003eBats of Southern and Central Africa: A Biogeographic and Taxonomic Synthesis\u003c/em\u003e (Wits University Press, 2020).\u003c/li\u003e\n\u003cli\u003eRusso, D. \u0026amp; Voigt, C. C. The use of automated identification of bat echolocation calls in acoustic monitoring: A cautionary note for a sound analysis. \u003cem\u003eEcol. Indic.\u003c/em\u003e\u003cstrong\u003e66\u003c/strong\u003e, 598\u0026ndash;602 (2016). \u003c/li\u003e\n\u003cli\u003eChesson, P. General theory of competitive coexistence in spatially-varying environments. \u003cem\u003eTheor. Popul. Biol.\u003c/em\u003e\u003cstrong\u003e58\u003c/strong\u003e, 211\u0026ndash;237 (2000a).\u003c/li\u003e\n\u003cli\u003eChesson, P. Mechanisms of maintenance of species diversity. \u003cem\u003eAnnu. Rev. Ecol. Evol. Syst.\u003c/em\u003e\u003cstrong\u003e31\u003c/strong\u003e, 343\u0026ndash;366 (2000b).\u003c/li\u003e\n\u003cli\u003eKneitel, J. M. \u0026amp; Chase, J. M. Trade‐offs in community ecology: Linking spatial scales and species coexistence. \u003cem\u003eEcol. Lett.\u003c/em\u003e\u003cstrong\u003e7\u003c/strong\u003e, 69\u0026ndash;80 (2004).\u003c/li\u003e\n\u003cli\u003eKricher, J. C. \u003cem\u003eTropical Ecology\u003c/em\u003e (Princeton University Press, 2011).\u003c/li\u003e\n\u003cli\u003eHilleRisLambers, J., Adler, P. B., Harpole, W. S., Levine, J. M. \u0026amp; Mayfield, M. M. Rethinking community assembly through the lens of coexistence theory. \u003cem\u003eAnnu. Rev. Ecol. Evol. Syst\u003c/em\u003e. \u003cstrong\u003e43\u003c/strong\u003e, 227\u0026ndash;248 (2012).\u003c/li\u003e\n\u003cli\u003eWilkinson, D. M. The disturbing history of intermediate disturbance. \u003cem\u003eOikos\u003c/em\u003e\u003cstrong\u003e84\u003c/strong\u003e, 145\u0026ndash;147 (1999). \u003c/li\u003e\n\u003cli\u003eFenton, M. B. \u0026amp; Rautenbach, I. L. A comparison of the roosting and foraging behaviour of three species of African insectivorous bats (Rhinolophidae, Vespertilionidae, and Molossidae). \u003cem\u003eCan. J. Zool\u003c/em\u003e. \u003cstrong\u003e64\u003c/strong\u003e, 2860\u0026ndash;2867 (1986).\u003c/li\u003e\n\u003cli\u003eFindley, J. S. \u0026amp; Black, H. Morphological and dietary structuring of a Zambian insectivorous bat community. \u003cem\u003eEcology\u003c/em\u003e\u003cstrong\u003e64\u003c/strong\u003e, 625\u0026ndash;630 (1983).\u003c/li\u003e\n\u003cli\u003eTaylor, P. J., Monadjem, A. \u0026amp; Nicolaas Steyn, J. Seasonal patterns of habitat use by insectivorous bats in a subtropical African agro-ecosystem dominated by macadamia orchards. \u003cem\u003eAfr. J. Ecol\u003c/em\u003e. \u003cstrong\u003e51\u003c/strong\u003e, 552\u0026ndash;561 (2013). \u003c/li\u003e\n\u003cli\u003eMonadjem, A. et al. Morphology and stable isotope analysis demonstrate different structuring of bat communities in rainforest and savannah habitats. \u003cem\u003eR. Soc. Open Sci\u003c/em\u003e. \u003cstrong\u003e5\u003c/strong\u003e, 180849 (2016). \u003c/li\u003e\n\u003cli\u003eBastolla, U. et al. The architecture of mutualistic networks minimizes competition and increases biodiversity. \u003cem\u003eNature\u003c/em\u003e\u003cstrong\u003e458\u003c/strong\u003e, 1018\u0026ndash;1020 (2009).\u003c/li\u003e\n\u003cli\u003eAdams, R. A., Bonaccorso, F. J. \u0026amp; Winkelmann, J. R. Revised distribution for \u003cem\u003eOtomops martiensseni\u003c/em\u003e (Chiroptera: Molossidae) in southern Africa. \u003cem\u003eGlob. Ecol. Conserv\u003c/em\u003e. \u003cstrong\u003e3\u003c/strong\u003e, 707\u0026ndash;714 (2015). \u003c/li\u003e\n\u003cli\u003eMagurran, A. E. \u003cem\u003eMeasuring Biological Diversity\u003c/em\u003e (Blackwell, Oxford, 2004).\u003c/li\u003e\n\u003cli\u003eWhittaker, R. H. Vegetation of the Siskiyou Mountains, Oregon and California. \u003cem\u003eEcol. Monogr\u003c/em\u003e. \u003cstrong\u003e30\u003c/strong\u003e, 279\u0026ndash;338 (1960).\u003c/li\u003e\n\u003cli\u003eTischler, W. \u003cem\u003eEinf\u0026uuml;hrung in die \u0026Ouml;kologie\u003c/em\u003e. 4th Ed. (Gustav Fischer, Stuttgart, 1993).\u003c/li\u003e\n\u003cli\u003eHanski, I. \u0026amp; Gyllenberg, M. Two general metapopulation models and the core-satellite species hypothesis. \u003cem\u003eAm. Nat\u003c/em\u003e. \u003cstrong\u003e142\u003c/strong\u003e, 17\u0026ndash;41 (1993).\u003c/li\u003e\n\u003cli\u003eAtmar, W. \u0026amp; Patterson, B. D. The measure of order and disorder in the distribution of species in fragmented habitat. \u003cem\u003eOecologia\u003c/em\u003e\u003cstrong\u003e96\u003c/strong\u003e, 373\u0026ndash;382 (1993). \u003c/li\u003e\n\u003cli\u003ePayrat\u0026oacute;-Borr\u0026agrave;s, C., Hern\u0026aacute;ndez, L. \u0026amp; Moreno, Y. Measuring nestedness: A comparative study of the performance of different metrics. \u003cem\u003eEcol. Evol\u003c/em\u003e. \u003cstrong\u003e10\u003c/strong\u003e, 11906\u0026ndash;11921 (2020). \u003c/li\u003e\n\u003cli\u003eAlmeida-Neto, M., Guimaraes, P., Guimaraes Jr, P. R., Loyola, R. D. \u0026amp; Ulrich, W. A consistent metric for nestedness analysis in ecological systems: Reconciling concept and measurement. \u003cem\u003eOikos\u003c/em\u003e\u003cstrong\u003e117\u003c/strong\u003e, 1227\u0026ndash;1239 (2008). \u003c/li\u003e\n\u003cli\u003eR Development Core Team. \u003cem\u003eR: A Language and Environment for Statistical Computing \u003c/em\u003e(R Foundation for Statistical Computing, Vienna, 2013). \u003c/li\u003e\n\u003cli\u003e\u0026Scaron;milauer, P. \u0026amp; Lep\u0026scaron;, J. \u003cem\u003eMultivariate Analysis of Ecological Data Using CANOCO 5\u003c/em\u003e (Cambridge University Press, 2014).\u003c/li\u003e\n\u003cli\u003eCrawley, M. J. \u003cem\u003eThe R Book\u003c/em\u003e (John Wiley \u0026amp; Sons, Chichester, 2007).\u003c/li\u003e\n\u003cli\u003ePinheiro, J., Bates, D., DebRoy, S., Sarkar, D. R. \u0026amp; R Core Team. \u003cem\u003eNlme: Linear and Nonlinear Mixed Effects Models\u003c/em\u003e. R package Version 3.1-152. https://CRAN.R-project.org/package=nlme (2021).\u003c/li\u003e\n\u003cli\u003eAldridge, H. D. J. N. \u0026amp; Rautenbach, I. L. Morphology, echolocation and resource partitioning in insectivorous bats. \u003cem\u003eJ. Anim. Ecol\u003c/em\u003e. \u003cstrong\u003e56\u003c/strong\u003e, 763\u0026ndash;778 (1987). \u003c/li\u003e\n\u003cli\u003eWilson, D. E. \u0026amp; Mittermeier, R. A. \u003cem\u003eHandbook of the Mammals of the World. Vol. 9. Bats\u003c/em\u003e (Lynx Edicions, Barcelona, 2019).\u003c/li\u003e\n\u003cli\u003eWaghiiwimbom, M. D., Eric-Moise, B. F., Jules, A. P., Aim\u0026eacute;, T. K. J. \u0026amp; Tamesse, J. L. Diversity and community structure of bats (Chiroptera) in the centre region of Cameroon. \u003cem\u003eAfr. J. Ecol\u003c/em\u003e. \u003cstrong\u003e58\u003c/strong\u003e, 211\u0026ndash;226 (2020). \u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table 1 and 3","content":"\u003cp\u003eTable 1 and 3 are available in the Supplementary Files section.\u003c/p\u003e\n"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"savanna, bats, community structure, spatial variation, nestedness, Kruger NP, South Africa","lastPublishedDoi":"10.21203/rs.3.rs-5295291/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5295291/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe savanna habitats often harbour abundant and species-rich bat communities. Whether they represent mere \u003cem\u003ead hoc\u003c/em\u003e assemblages of incidentally co-occurring forms or distinct entities integrated by locally specific adaptations and balanced resource partitionings is largely unknown, as are the natural drivers shaping community variation at different spatial scales. An extensive dataset (130,888 acoustic bat records, 31 OTUs) was collected in 60 plots across Kruger National Park (KNP), South Africa; the plots were located (i) at perennial rivers, (ii) at seasonal rivers, and (iii) on dry crests away from any water source. Besides the effect of water availability, distance to campsites, and microgeographic variation on bat community richness and structure, we revealed (i) extensive homogeneity in community structure at local, subregional, and regional scales contrasting to a mosaic between-plot variation, (ii) absence of robust effects of environmental biotic and abiotic predictors on the distribution of individual OTUs and community variation, (iii) nearly identical pattern of habitat preferences in all community members approaching the centroid of KNP habitat variation, and (iv) an exceptionally high degree of community nestedness. These results suggest that the bat community of the KNP savanna biome represents a single entity consistently integrated with a network of coexistence relations that probably arose locally during long savanna history.\u003c/p\u003e","manuscriptTitle":"Bat communities of savanna biome in the Kruger National Park, South Africa","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-11-15 16:31:35","doi":"10.21203/rs.3.rs-5295291/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-04-16T05:26:04+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-05T04:34:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"29600577811453312727030573997747596926","date":"2025-03-28T03:59:40+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-01-25T17:42:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"205615751423643789640553108889697650096","date":"2025-01-02T17:33:14+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"84023993667601978258081422304979529876","date":"2025-01-01T13:16:16+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-12-03T19:23:47+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-12-02T16:51:59+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-11-07T08:34:58+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-11-04T06:56:51+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2024-10-19T15:32:43+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e53fb759-d390-45ac-abe3-ef678c1f3b07","owner":[],"postedDate":"November 15th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-11T17:54:17+00:00","versionOfRecord":[],"versionCreatedAt":"2024-11-15 16:31:35","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5295291","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5295291","identity":"rs-5295291","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}