VITOR GABRIEL PEREIRA JUNTA A INFLUÊNCIA DA ESTRUTURA DE HABITAT SOBRE AS COMUNIDADES DE INVERTEBRADOS SUBTERRÂNEOS DA REGIÃO DE SANTANA, BAHIA. LAVRAS-MG 2023 VITOR GABRIEL PEREIRA JUNTA A INFLUÊNCIA DA ESTRUTURA DE HABITAT SOBRE AS COMUNIDADES DE INVERTEBRADOS SUBTERRÂNEOS DA REGIÃO DE SANTANA, BAHIA. Dissertação apresentada a Universidade Federal de Lavras, como parte das exigências do Programa de Pós-Graduação em Ecologia Aplicada, área de concentração Ecologia e Conservação de Recursos em Paisagens Fragmentadas e Agrossistemas, para a obtenção do título de Mestre. Prof. Dr. Rodrigo Lopes Ferreira Orientador LAVRAS-MG 2023 Ficha catalográfica elaborada pelo Sistema de Geração de Ficha Catalográfica da Biblioteca Universitária da UFLA, com dados informados pelo(a) próprio(a) autor(a). Junta, Vitor Gabriel Pereira. A Influência da Estrutura de Habitat sobre as Comunidades de Invertebrados Subterrâneos da Região de Santana, Bahia. / Vitor Gabriel Pereira Junta. - 2023. 84 p. : il. Orientador(a): Rodrigo Lopes Ferreira. Dissertação (mestrado acadêmico) - Universidade Federal de Lavras, 2023. Bibliografia. 1. Comunidades. 2. Caverna. 3. Habitat. I. Ferreira, Rodrigo Lopes. II. Título. O conteúdo desta obra é de responsabilidade do(a) autor(a) e de seu orientador(a). VITOR GABRIEL PEREIRA JUNTA A INFLUÊNCIA DA ESTRUTURA DE HABITAT SOBRE AS COMUNIDADES DE INVERTEBRADOS SUBTERRÂNEOS DA REGIÃO DE SANTANA, BAHIA. THE HABITAT STRUCTURE INFLUENCE OVER SUBTERRANEAN INVERTEBRATE COMMUNITIES OF SANTANA REGION, BAHIA Dissertação apresentada a Universidade Federal de Lavras, como parte das exigências do Programa de Pós-Graduação em Ecologia Aplicada, área de concentração Ecologia e Conservação de Recursos em Paisagens Fragmentadas e Agrossistemas, para a obtenção do título de Mestre. APROVADO em 25 de abril de 2023. Prof. Dr. Rodrigo Lopes Ferreira – UFLA Prof. Dr. Paulo dos Santos Pompeu - UFLA Dr. Marcus Paulo Alves de Oliveira – BioEspeleo Consultoria Ambiental LTDA Prof. Dr. Rodrigo Lopes Ferreira Orientador LAVRAS-MG 2023 AGRADECIMENTOS Agradeço à Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) pela bolsa concedida. Agradeço por fazer parte do projeto que conta com financiamento do Termo Aditivo de Compromisso de Compensação Espeleológica nº 1/2018, firmado entre o Instituto Chico Mendes de Conservação e Biodiversidade (ICMBio-Vale S.A.) relativo a execução da compensação pelos impactos negativos irreversíveis as cavidades naturais subterrâneas com grau de relevância alto, através do projeto: “Dispersão versus confinamento: análise composicional e de estrutura de habitat como subsídio à compreensão de mecanismos responsáveis pela identidade faunística subterrânea” À Universidade Federal de Lavras, ao programa de pós-graduação em Ecologia Aplicada, assim como aos professores que de alguma forma ajudaram nessa trajetória, especialmente ao professor Lucas Del Bianco pelos grandes esclarecimentos nas análises estatísticas. Aos meus orientadores Rodrigo Lopes Ferreira e Marconi Souza-Silva, não só por terem me orientado durante o mestrado, mas por todos os ensinamentos desde o início da graduação. Agradeço muito pelas oportunidades e portas abertas que essas duas pessoas me proporcionaram. Obrigado por me apresentarem e fazerem eu me apaixonar pelo mundo das cavernas. À equipe do Centro de Estudos em Biologia Subterrânea, em especial: Alicia Helena, Gabrielle Pacheco, Júlio César Vaz, Guilherme Prado, Rafael Cardoso, Paulo Venâncio, Thaís Pelegrini, Lais Furtado, Felipe Carvajal e Priscila Souza pelo apoio nas campanhas de coleta e Gabriel Vaz pelo apoio nas triagens do material coletado. À Evânio Santos, que nos recebeu excelentemente bem em Santana-BA, foi nosso guia durante boa parte das expedições e se tornou um amigo. À toda população de Santana e região, que foram extremamente receptivos conosco, seja tendo paciência com os pedidos gigantescos na padaria, sendo nos levando ou apontando direções de cavernas. Aos especialistas que ajudaram na identificação: Giovanna Cardoso, Guilherme Prado, Leopoldo Benardi. À toda minha família e amigos que de alguma forma ajudaram a construir essa pessoa que sou hoje, que me incentivaram, me energizaram para continuar e me ouviram desabafar quando necessário. À cada pessoa que conheci através da Comunidade Zen Budista Daissen em todos os seus âmbitos, que há mais de três anos vem sendo um grande refúgio. Aos meus pais Rui e Ivanir, que nunca mediram esforços para me apoiar e fazer com que eu chegue o mais longe que eu puder. Obrigado por serem essa rocha sólida onde posso me apoiar para buscar meus objetivos. À Alicia Helena, minha namorada, que caminhou junto comigo nessa etapa, me apoiando, dando força e me acolhendo quando preciso. Que possamos caminhar mais e mais juntos. Muito obrigado a todos! Que todos os seres possam se beneficiar. RESUMO O presente trabalho visa avaliar a influência da heterogeneidade de habitat, através de atributos paisagísticos, físicos, tróficos e microclimáticos sobre composição e riqueza de invertebrados cavernícolas em distintas escalas amostrais, assim podendo contribuir com uma melhor conservação dos ambientes subterrâneos. O trabalho é composto de dois artigos redigidos conforme as normas do periódico Biodiversity and Conservation. O primeiro artigo teve como objetivo entender quais fatores do habitat possuem influência sobre comunidade de invertebrados em 24 cavernas da região de Santana-BA. A distância da entrada e as distâncias geográficas entre cavernas mostraram importantes para composição e riqueza das comunidades, juntamente com a heterogeneidade de habitat, representada por diferentes substratos usados como abrigos e recursos tróficos. O segundo capítulo traz um olhar mais específico para a quinta maior caverna do Brasil, a Gruta do Padre. Aqui, além de tentar entender quais fatores do habitat influenciam as comunidades de invertebrados, a caverna é definida como um novo Hotspot de Biodiversidade Subterrânea, com 25 espécies estritamente subterrâneas. Dentro da Gruta do Padre, os fatores do habitat que mais tiveram relação com a fauna foram a distância da entrada e os diferentes níveis de altitude da caverna. A presença de recursos tróficos e diferentes substratos inorgânicos também foi relevante para as comunidades. Palavras-chave: Cavernas. Invertebrados. Habitat. Comunidade. ABSTRACT This work aims to evaluate the influence of habitat heterogeneity, through landscape, physical, trophic, and microclimatic attributes on the composition and richness of cave invertebrates at different sample scales, thus being able to contribute to better conservation of subterranean environments. The work consists of two articles written according to the rules of the journal Biodiversity and Conservation. The first article aimed to understand which habitat factors influence the invertebrate community in 24 caves in the region of Santana-BA. Distance from the entrance and geographic distances between caves were important for community composition and richness, along with habitat heterogeneity, represented by different substrates used as shelters and trophic resources. The second chapter takes a more specific look at the fifth largest cave in Brazil, Gruta do Padre cave. Here, in addition to trying to understand which habitat factors influence invertebrate communities, the cave is defined as a new Hotspot of Subterranean Biodiversity, with 25 strictly subterranean species. Within the Gruta do Padre cave, the habitat factors that were most closely related to the fauna were the distance from the entrance and the different altitude levels of the cave. The presence of trophic resources and different inorganic substrates was also relevant for the communities. Keywords: Caves. Invertebrates. Habitat. Community. SUMÁRIO PRIMEIRA PARTE ................................................................................................................. 9 INTRODUÇÃO ........................................................................................................................ 9 REFERÊNCIAS ..................................................................................................................... 12 SEGUNDA PARTE-ARTIGOS ............................................................................................ 16 ARTIGO 1: ARE THERE CHOICES IN THE DARKNESS? HABITAT SELECTION AND CONSERVATION OF CAVE INVERTEBRATES IN A BRAZILIAN SEMI-ARID REGION ................... 16 ARTIGO 2: HABITAT SELECTION OF CAVE INVERTEBRATES IN A NEW SOUTH AMERICAN HOTSPOT OF SUBTERRANEAN BIODIVERSITY .......................................................................... 51 9 PRIMEIRA PARTE INTRODUÇÃO Os padrões de distribuição das espécies têm sido a anos foco de trabalhos visando entender fatores estruturantes de comunidades (Dunson & Travis, 1991; Kolasa & Pickett, 1991; Cushman & McGarigal, 2004; Steinitz et al., 2006; Talley, 2007; Pacheco et al., 2020). Tanto fatores bióticos, quanto abióticos foram levados em consideração em trabalhos anteriores, buscando entender esses padrões de distribuição. Dentre os agentes bióticos, podem ser destacadas as interações intra e interespecíficas, como a predação e a competição (Dunson & Travis, 1991). A respeito dos fatores abióticos, pode-se salientar a temperatura, umidade relativa do ar, salinidade, quantidade de abrigos, tipo de substratos, entre outros (Dunson & Travis, 1991; Pacheco et al., 2020). A heterogeneidade de habitat também vem sendo um fator importante estudado por diversos autores na busca de respostas aos padrões de distribuição das espécies em uma determinada área (Amarasekare & Nisbet, 2001; Cornell, 2010; Yang et al., 2015; Stein et al., 2015; Vargas-Mena et al., 2020; Souza-Silva et al., 2021). Assim, diferentes usos, por parte das espécies, de diferentes microhabitats são decretórios para a coexistência de múltiplas populações (MacArthur & Levins, 1967; Tilman, 1982; Chesson, 2000a; Mehrabi et al., 2014; Souza-Silva et al., 2021). A ideia da influência da heterogeneidade de habitat sobre as características das comunidades pode ser aplicada a diferentes ambientes, inclusive aos ambientes subterrâneos, onde essa heterogeneidade pode afetar as características das comunidades cavernícolas. (Pacheco et al., 2020; Souza-Silva et al., 2021) Os ambientes subterrâneos são caracterizados pela ausência permanente de luz, temperaturas estáveis durante todo ano (normalmente próxima à média anual do ambiente externo) e umidade relativa do ar tendendo à saturação e oligotrofia (Howarth, 1983). Dessa maneira, grande parte das fontes de energia para a rede trófica são alóctones (provenientes do ambiente externo), podendo ser carreadas para o interior das cavernas através da ação da água, animais ou por raízes de plantas (Howarth, 1983; Polis et al., 1997; Ferreira, 1998; Souza-Silva, 2003; Simon et al., 2007; Culver & Pipan, 2009). Organismos autótrofos quimiossintetizantes também podem ser de grande importância nutricional para os animais que compõem a fauna cavernícola (Howarth, 1983, Sarbu et al., 2018). Em consequência a essas limitações impostas pelos ambientes cavernícolas, as espécies animais colonizadoras desses habitats são limitadas e possuem usualmente pré-adaptações que 10 possibilitam sua existência nesses locais. Os organismos habitantes das cavernas usualmente são classificados por um sistema contendo três categorias, proposto por Schinner-Racovitza e modificado por Sket (2008). Os (a) trogloxenos são organismos que passam parte de seu ciclo de vida dentro das cavidades, mas ainda precisam do meio externo para completar todo seu ciclo de vida; os animais (b) troglófilos são os que conseguem manter populações viáveis tanto fora, quanto dentro das cavernas; e os (c) troglóbios são as espécies que possuem suas populações restritas ao ambiente cavernícola durante todo seu ciclo de vida. Os animais classificados como troglóbios comumente podem apresentar adaptações em resposta às pressões seletivas presentes nos ambientes subterrâneos, chamadas troglomorfismos. Essas adaptações podem ser morfológicas, como redução de estruturas oculares, perda da pigmentação melânica e alongamento de apêndices; fisiológicas, como diminuição da taxa metabólica e estratégia de vida K; ou comportamentais (Romero & Green, 2005). Além disso, são raros e endêmicos (em sua maioria) devido a uma longa história evolutiva em ambientes tão estáveis como as cavernas, e acabam se tornando organismos sensíveis a mudanças no sistema, tais quais pequenas variações de temperatura e umidade (Mammola et al., 2019; Culver & Pipan, 2009). A fauna cavernícola brasileira passou a ser estudada a partir de 1980, especialmente nos estados de São Paulo, Goiás, Minas Gerais, Bahia, Paraná, Mato Grosso e Ceará (Dessen et al., 1980; Pinto-da-Rocha, 1995). Entretanto, muitos estudos foram feitos dentro de Unidades de Conservação e por isso encontram-se fragmentados e escassos em determinadas regiões do Brasil (Ferreira, 2005; Ferreira et al. 2010; Gnaspi e Trajano, 1994; Pinto-da-Rocha, 1995; Souza-Silva, 2008; Trajano, 2000; Zepon & Bichuette, 2017). Diante do exposto, entender os aspectos ecológicos da fauna subterrânea é de suma importância, já que, apenas os levantamentos taxonômicos podem não ser eficientes para a conservação da biodiversidade cavernícola (Trajano et al. 2010). Logo, compreender como os padrões de riqueza e composição são influenciados pelos fatores que atuam na manutenção e estruturação dessas comunidades no tempo e espaço é invariavelmente necessário para a conservação das espécies cavernícolas (Legendre et al. 2005; Jost et al., 2010). A distância das entradas, disponibilidades de recursos alimentares e heterogeneidade de habitats mostram-se como fatores que exercem forte influência sobre as características das comunidades de invertebrados em cavernas (Prous, 2005; Oliveira, 2014; Pellegrini et al., 2016; Gomes, 2017; Zepon & Bichuette, 2017), já que esses animais tendem a buscar por 11 microhabitats preferenciais dentro dos ambientes subterrâneos (Culver & Pipan, 2009; Mammola et al., 2016; Souza-Silva & Ferreira, 2009; Souza-Silva et al. 2021). Tais microhabitats podem compreender componentes orgânicos como detritos vegetais, guano e outros tipos de matéria orgânica, assim como componentes físicos, como frestas, espaços sob rochas, corpos d’água, entre outros (Ferreira et al., 2007; Culver & Pipan, 2009; Souza-Silva et al., 2011; Simões et al., 2015; Gomes, 2017). Dessa forma, a heterogeneidade de habitat pode agir sobre a estrutura das comunidades de invertebrados cavernícolas por meio da disponibilização de diferentes habitats para a fauna, agindo então sobre a distribuição e riqueza das espécies (Zagmajster et al., 2018). Logo, este estudo visa analisar a estrutura de habitat como subsídio à compreensão dos mecanismos responsáveis pela composição e riqueza de comunidades subterrâneas na região de Santana, estado da Bahia. Para tal, foram utilizados parâmetros paisagísticos, físicos, tróficos e microclimáticos em diferentes escalas amostrais. A dissertação é composta por dois artigos redigidos nas normas do periódico Biodiversity and Conservation. O primeiro artigo visa elucidar quais características do habitat influenciam a riqueza e composição de comunidades de invertebrados cavernícolas em 24 cavernas da região de Santana-BA. O segundo artigo foca as análises em uma importante caverna da região, a Gruta do Padre. Na quinta maior caverna do mundo, além de entender quais fatores do habitat influenciam nas comunidades de invertebrados subterrâneos, o objetivo foi também defini-la como um novo Hotspot de Biodiversidade Subterrânea, graças às 25 espécies troglóbias encontradas. 12 REFERÊNCIAS AMARASEKARE, P.; NISBET, R. Spatial heterogeneity, source-sink dynamics and the local coexistence of competing species. American Naturalist, v. 158, p. 572–584, 2001. https://doi.org/10.1086/323586 CHESSON, P. General theory of competitive coexistence in spatially varying environments. Theoretical Population Biology, v. 58, p. 211–237, 2000a. https://doi.org/10.1006/tpbi.2000.1486 CORNELL, H. W. Niche overlapping. In: HASTINGS, A.; GROSS, L. J. (eds.). Encyclopedia of theoretical ecology. Oakland: University of California Press, 2010. CULVER, D. C.; PIPAN, T. The biology caves and other subterranean habitats. Oxford: Oxford University Press, 2009. CUSHMAN, S.; MCGARIGAL, K. Patterns in the species–environment relationship depend on both scale and choice of response variables. Oikos, v. 105, p. 117-124, 2004. https://doi.org/10.1111/j.0030-1299.2004.12524.x DESSEN, E. M. B.; ESTON, V. R.; SILVA, M. S. Levantamento preliminar da fauna de cavernas de algumas regiões do Brasil. Ciência e Cultura, v. 32, p. 714-725, 1980. DUNSON, W. A.; TRAVIS, J. The role of abiotic factors in community organization. The American Naturalist, v. 138, p. 1067-1091, 1991. https://doi.org/10.1086/285270 FERREIRA, R. L. A medida da complexidade ecológica e suas aplicações na conservação e manejo de ecossistemas subterrâneos. Tese de doutorado, Universidade Federal de Minas Gerais, 2004. FERREIRA, R. L. A vida subterrânea nos campos ferruginosos. O Carste, v. 3, n. 17, p. 106-115, 2005. FERREIRA, R. L.; PROUS, X.; BERNARDI, L. F. O.; SOUZA-SILVA, M. Fauna subterrânea do estado do Rio Grande do Norte: Caracterização e impactos. Revista Brasileira de Espeleologia, v. 1, p. 25-51, 2010. FERREIRA, R. L.; MARTINS, R. P.; PROUS, X. Structure of bat guano communities in a dry Brazilian cave. Tropical Zoology, v. 20, n. 1, p. 55-74, 2007. FERREIRA, R. L.; MARTINS, R. P. Diversity and Distribution of Spiders Associated with Bat Guano Piles in Morrinho Cave (Bahia State, Brazil). Diversity and Distributions, v. 4, n. 5-6, p. 235-241, 1998. GOMES, A. M. Uma luz na escuridão: desvendando os processos estruturadores da fauna cavernícola de Lagoa Santa, Minas Gerais. Tese de doutorado, Universidade Federal de Minas Gerais, 2010. HOWARTH, F.G., 1983. Ecology of cave arthropods. Annual Review of Entomology, v. 28, p. 365–389. JOST, L., DEVRIES, P., WALLA, T., GREENEY, H., CHAO, A., & RICOTTA, C., 2010. Partitioning diversity for conservation analyses. Diversity and Distributions, 16 (1), 65-76. 13 KOLASA, J., PICKETT, S.T.A. (Eds.), 1991. Ecological heterogeneity. Ecological studies. Springer, New York. https://doi.org/10.1007/978-1-4612-3062-5 LEGENDRE, P., BORCARD, D. & PERES-NETO, P. R., 2005. Partitioning the Spatial Variation of Community Composition Data. Ecological Monographs, 75, 435–450. MACARTHUR R. H., LEVINS R., 1967. The limiting similarity, convergence, and divergence of coexisting species. Am Nat 101, 377–385. https://doi.org/10.1086/282505 MAMMOLA S., GIACHINO P. M., PIANO E., JONES A., BARBERIS M., BADINO G. & ISAIA M., 2016. Ecology and sampling techniques of an understudied subterranean habitat: the Milieu Souterrain Superficiel (MSS). The Science of Nature, 103: 88. https://doi.org/10.1007/ s00114-572 016-1413-9 MAMMOLA, S.; PIANO, E; CARDOSO, P; VERNON, P.; DOMÍNGUEZ-VILLAR, D; CULVER, D. C.; PIPAN, T; ISAIA, M., 2019. Climate change going deep: The effects of global climatic alterations on cave ecosystems. The Anthropocene Review. 6 (1–2), 98– 116. https://doi.org/10.1177%2F2053019619851594 MEHRABI Z., SLADE E. M., SOLIS A., MANN D. J., 2014. The Importance of microhabitat for biodiversity sampling. PLoS ONE 9 (12) e114015. https://doi.org/10.1371/journal.pone.0114015 OLIVEIRA M. P. A., 2014. Os métodos de coleta utilizados em cavernas são eficientes para a amostragem da fauna subterrânea? Dissertação de Mestrado, Universidade Federal de Lavras, Lavras. PACHECO, G. S. M., SOUZA-SILVA, M., CANO, E., FERREIRA, L. F., 2020. The role of microhabitats in structuring cave invertebrate communities in Guatemala. International Journal of Speleology, 49 (2), 161–169. https://doi.org/10.5038/1827-806X.49.2.2333 PELLEGRINI T. G. & FERREIRA, R. L., 2016. Are inner cave communities more stable than entrance communities in Lapa Nova show cave? Subterranean Biology, 20: 15. https://doi.org/10.3897/subtbiol.20.9334 PINTO-DA-ROCHA, 1995. Sinopse da fauna cavernícola do Brasil (1907 - 1994). Papéis Avulsos de Zoologia, 39 (6): 61-163. POLIS, G. A.; ANDERSON, W. B.; HOLT, R. D., 1997. Towards an integration of landscape and food web ecology: the dynamics of spatially subsidized food webs. Annual Review of Ecology and Systematics v. 28, n. 289-316. PROUS X., 2005. Entradas de cavernas: interfaces de biodiversidade entre ambientes externos e subterrâneos. Distribuição dos artrópodes da Lapa do Mosquito, Minas Gerais. Dissertação de mestrado apresentada no programa de Pós-graduação em Ecologia, Conservação e Manejo da Vida Silvestre. Universidade Federal de Minas Gerais. ROMERO A., GREEN M., 2005. The end of regressive evolution: examining and interpreting the evidence from cave fishes. Journal of Fish Biology 67:3-32. SARBU, S. M.; LASCU, C.; BRAD, T., 2018. Dobrogea: Movile Cave. Cave and Karst Systems of the World. 429–436. https://doi.org/10.1007/978-3-319-90747-5_48 14 SIMÕES, M. H., SOUZA-SILVA, M. & FERREIRA, R. L., 2015. Cave physical attributes influencing the structure of terrestrial invertebrate communities in Neotropics. Subterranean Biology, 16: 103-121. SIMON K. S., PIPAN T., CULVER D. C., 2007. A conceptual model of the flow and distribution of organic carbon in caves. Journal of Cave and Karst Studies 69 (2), 279-284. SOUZA-SILVA M., 2008. Ecologia e conservação das comunidades de invertebrados cavernícolas na Mata Atlântica Brasileira. Tese apresentada ao Programa de Pós-graduação em Ecologia,Conservação e Manejo de Vida Silvestre do Instituto de Ciências Biológicas da Universidade Federal de Minas Gerais, 211–224 SOUZA-SILVA, M. S., MARTINS, R. P., & FERREIRA, R. L., 2011. Cave lithology determining the structure of the invertebrate communities in the Brazilian Atlantic Rain Forest. Biodiversity and Conservation, 20(8), 1713-1729. SOUZA-SILVA, M., & FERREIRA, R. L., 2009. Caracterização ecológica de algumas cavernas do Parque Nacional de Ubajara (Ceará) com considerações sobre o turismo nestas cavidades. Revista de Biologia e Ciências da Terra, 9 (1). SOUZA-SILVA, M., 2003. Influência da disponibilidade e consumo de detritos na composição e estrutura de mesofauna cavernícola. Dissertação de Mestrado Em Ecologia. Universidade Federal de Minas Gerais, 80. SOUZA-SILVA, M., CERQUEIRA, R. F. V., PELLEGRINI, T. G., FERREIRA, R. L., 2021. Habitat selection of cave-restricted fauna in a new hotspot of subterranean biodiversity in Neotropics. Biodiversity and Conservation. https://doi.org/10.1007/s10531-021-02302-8 STEIN A., BECK J., MEYER C., WALDMANN E., WEIGELT P., KREFT H., 2015. Differential effects of environmental heterogeneity on global mammal species richness. Global Ecol Biogeogr 24 (9), 1072–1083. https://doi.org/10.1111/geb.12337 STEINITZ, O., HELLER, J., TSOAR, A., ROTEM, D., KADMON, R., 2006. Environment, dispersal and patterns of species similarity. Journal of Biogeography, 33, 1044-1054. https://doi.org/10.1111/j.1365-2699.2006.01473.x TALLEY, T.S., 2007. Which spatial heterogeneity framework? Consequences for conclusions about patchy population distributions. Ecology, 88, 1476-1489. https://doi.org/10.1890/06-0555 TILMAN D., 1982. Resource competition and community structure. Monographs in population biology. Princeton Univ. Press, Princeton. TRAJANO E, BICHUETTE M., 2010. Relevância de cavernas: porque estudos ambientais espeleobiológicos não funcionam. Espeleo-Tema. 21, 105–112. TRAJANO E., 2000. Cave faunas in the Atlantic tropical rain forest: composition, ecology and conservation. Biotropica 32: 882-893. VARGAS-MENA J. C., CORDERO-SCHMIDT E., RODRIGUEZ-HERRERA B., 2020. Inside or out? Cave size and landscape effects on cave-roosting bat assemblages in Brazilian Caatinga caves. J Mammal 101 (2), 464–475. https://doi.org/10.1093/jmammal/gyz206 15 YANG Z., LIU X., ZHOU M., AI D., WANG G., 2015. The effect of environmental heterogeneity on species richness depends on community position along the environmental gradient. Sci Rep 5 (1), 1–7. https://doi.org/10.1038/srep15723 ZAGMAJSTER M., MALARD F., EME D. & CULVER D. C., 2018. Subterranean Biodiversity. Patterns from Global to Regional Scales. In Cave Ecology. 195-227. https://doi.org/10.1007/978-3-319-98852-8_9. ZEPON T., BICHUETTE M. E., 2017. Caracterização e análise dos estudos ecológicos sobre comunidades de invertebrados subterrâneos brasileiros. In: RASTEIRO, M. A.; TEIXEIRA 16 SEGUNDA PARTE-ARTIGOS Artigo 1: Are there choices in the darkness? Habitat selection and conservation of cave invertebrates in a Brazilian semi-arid region Artigo redigido conforme as normas do periódico Biodiversity and Conservation. 17 Are there choices in the darkness? Habitat selection and conservation of cave invertebrates in a Brazilian semi-arid region Vitor Gabriel Pereira Junta¹, Marconi Souza Silva¹, Rodrigo Lopes Ferreira¹* ¹Center of Studies in Subterranean Biology, Ecology and Conservation Department, Natural Sciences Institute, Federal University of Lavras, Lavras, MG, Brazil *Corresponding Author vitor.junta@outlook.com marconisilva@ufla.br drops@ufla.br Abstract Habitat characteristics are key factors for fauna distribution inside caves, as distinct species usually present different microhabitat requirements. Hence, this work aimed to understand which habitat traits influence the richness and composition of invertebrate communities in 24 caves located in southwestern Bahia state, in a semi-arid area. Landscape, physical, trophic, and microclimatic traits were used as predictors and were analyzed in both meso and microscale inside the caves. In total, 338 species from 37 orders and at least 93 families were found, with 41 of them considered troglobitic. The results showed that the distance from the nearest entrance and the geographic distances between caves were important for communities’ composition and richness, along with habitat heterogeneity, represented by different substrates used as shelters and trophic resources. Introduction Different environmental traits can lead to several responses from the associated biota, thus determining singular distribution patterns (Tews et al. 2004; Odum & Barret 2006). Habitat heterogeneity is one of the keys factors for the distribution patterns since a higher heterogeneity usually mailto:vitor.junta@outlook.com mailto:marconisilva@ufla.br mailto:drops@ufla.br 18 allows the coexistence of a higher number of species (Yang et al. 2015; Stein et al. 2015; Vargas-Mena et al. 2020; Pacheco et al. 2020; Souza-Silva et al. 2021). In subterranean environments, habitat heterogeneity is also determinant for species distribution, even considering that the subterranean realm many times presents unique habitat characteristics (Pacheco et al. 2020; Souza-Silva et al. 2021). From theories proposed at the beginning of the XX century to advances in sampling techniques and analysis achieved at the end of the same period, important mechanisms for the definition of subterranean biodiversity started to be unraveled (Mammola et al. 2016; Rabelo et al. 2020). Even though the subterranean environments are not limited to caves, they are amongst the most important subterranean habitats, and are the most studied subterranean habitats due to its dimensions (Juberthie et al. 1980; Mammola et al. 2016; Rabelo et al. 2020; Pacheco et al. 2020; Souza-Silva et al. 2021). These natural cavities are known for having peculiar characteristics, such the higher climatic stability when compared to the epigean surrounding environments, with stable temperature all over the year and high humidity (Howarth 1980, 1983). These habitats are also characterized by the absence of light and a tendency towards the oligotrophy (Culver & Pipan 2009). Among the mechanisms determining the subterranean biodiversity, the distance from the caves´ entrances stands out (Ficetola et al. 2018; Mammola 2019; Souza-Silva et al. 2021; Furtado et al. 2022). Since most cave trophic resources come from the external habitats and are transported to caves through their entrances, a higher distance from the entrance limits the organic supply availability for the communities (Tobin et al. 2013; Moseley 2008; Ficetola et al. 2018; Mammola 2019). Moreover, regardless of the general stability, caves present a gradient of conditions from near-to-entrance (usually more unstable) to deep zones (Moseley 2009; Tobin et al. 2013; Lunghi et al. 2014; Prous et al. 2015; Mammola & Isaia 2018; Lunghi & Manenti 2020; Souza-Silva et al. 2021). This zonation creates distinct microhabitats for cave species, which vary regarding substrate types as well as climatic and photic properties (Moseley 2008; Souza-Silva et al. 2011b; Du Preez et al. 2015; Lunghi et al. 2017; 19 Mammola & Isaia 2017; Mammola 2019; Lunghi & Manenti 2020; Mammola et al. 2020; Souza-Silva et al. 2021). The uniqueness of the environmental traits found in caves usually selects pre-adapted species, which are the only capable to colonize these habitats. It also creates singular evolutive pressures that may lead to restricted subterranean species, called troglobitic, which can maintain viable populations exclusively in subterranean environments (Racovitza 1907; Sket 2008). The evolution in subterranean habitats can produce, in the restricted species, adaptations. These adaptations can be morphological (named troglomorphisms), such as reduction in ocular structures, loss of melanic pigmentation, and appendage elongation. It also can be physiological, like metabolic rate reduction and K life strategy, and even behavioral (Romero & Green 2005). These animals are, in their majority, rare and endemic, being very sensitive to slight variations in their habitats, as in temperature and moisture content (Culver & Pipan 2009; Mammola et al. 2019). Troglobitic species descend from epigean ancestors that migrate to subterranean environments as became isolated, for example by past climatic events, such as retractions of rain forests. Accordingly, caves can function as a refuge for fauna in moments of climate change, besides being a showcase of the fauna from the past (Culver & Sket 2000; Sobral-Souza et al. 2015). The western Bahia state, located in the Brazilian northeast, is a semi-arid region that presents areas with high speleological potential. The region of the municipalities of Santana, Santa Maria da Vitória, and Canápolis, presents approximately 120 registered caves (CECAV 2022) and is home of important caves, such as the Gruta do Padre cave, with along two other caves (Gruta do Cipó Cave and Gruta da Bananeira Cave) represent the longest hydrological subterranean system from Brazil. Considering the high potential and the few studies accomplished in this area, the main objective of this work was to identify habitat variables determining the richness and composition of cave invertebrates. Using landscape, physical, trophic, and microclimatic traits of the caves, these hypotheses were tested: i) species richness will reduce with the increase of distance from the entrance; ii) geographically closer caves will present more similar communities, and iii) invertebrates will respond to different habitat 20 components and habitat heterogeneity on the cave floor. In addition, we discuss the importance of preserving this high-potential area for subterranean biodiversity. Material and Methods Study area The sample campaigns occurred between August 23 and September 02 of 2021, and May 23 and June 02 of 2022 in the municipalities of Santana, Santa Maria da Vitória e Canápolis, Bahia State, Brazil. This transition zone between Seasonal Dry Forests and Caatinga is classified by Köppen (1936) as Aw, with dry winters and rainy summers, and has a high potential for endemic species (Dinerstein 2017). The high potential found in this area may be due to many paleoclimatic changes caused by expansions and retractions of the Amazonian and Atlantic rains forests (Sobral-Souza et al. 2015). The area is located within the limits of the Corrente River basin and its affluents which are an important tributary of the São Francisco River. For security reasons, the samples were conducted only in the dry season, since many caves in this region present subterranean rivers with wide capitation basins, which make the rainy season highly dangerous for those intending to visit such caves (Fig. 1; Table 5). This area is inserted in the Bambuí Group, the largest carbonate formation in the country, with more than 145,000 Km², and more than 6,000 caves registered (Auler et al. 2019). It is important to highlight that this region represents a priority area for the Brazilian Speleological heritage conservation according to the map published by the Centro Nacional de Pesquisa e Conservação de Cavernas – CECAV. Field procedures Sampling design The composition and richness of cave invertebrates, as well as the habitat structure traits, were determined along 122 transects (mesoscale sampling - 10 × 3 m each) distributed on the floor of 24 caves, from the entrances to their deeper regions. Quadrats (micro-scale sampling - 1 m2) were placed in triplicates inside the limits of each transect (Fig. 2), totalizing 366 quadrats (Souza-Silva et al. 2021). Invertebrate sampling was executed by visual search along the transects and quadrats (Souza-Silva et 21 al. 2021). The sampling in the quadrats allowed the detection of small-size and low-mobility species, which could then be carefully searched in the remaining transect if discovered. The sampling was first conducted in the quadrats and later in the respective transect, always by three collectors, and was only finished when all the invertebrates had been sampled and/or accounted for. Since the several sampling regions along the cave presented a significant structural distinction, the time spent searching for invertebrates in each transect was variable. Moreover, direct intuitive search approaches were used in different cave locations to increase the discovery of troglobitic and stygobitic species (Wynne et al. 2019). Invertebrates were preserved in labeled vials containing 70% ethanol. In laboratory, the specimens were sorted with a Stemi 508 (ZEISS) stereomicroscope, identified until the lowest possible taxonomic level, and separated into morphotypes (Oliver & Beattie, 1996). Potential troglobitic species were identified by the presence of troglomorphic characteristics, such as pigmentation and eye reduction, and appendage elongation, among others (Culver & Pipan 2009). Furthermore, experts in several taxa were contacted to help to identify particular troglomorphic characteristics (specialists are acknowledged further on). The voucher specimens were deposited in the Collection of Subterranean Invertebrates of Lavras (ISLA), linked to the Center of Studies on Subterranean Biology (CEBS) of the Federal University of Lavras (UFLA). Environmental traits in different scales The measurement of the habitat structure traits in the transects was carried out according to the methodology used by Souza-Silva et al. (2021). In order to visually quantify the surface area occupied by various organic and inorganic substrates, each transect was divided into 10 parts (1 x 3 m) (Fig. XX). The area occupied by each substrate along the whole transect was then calculated by sum. The same researcher characterized all transects to reduce observer error. The humidity and temperature were measured using a digital thermo-hygrometer that was set up on the ground in the center of each transect. Photographs (4000 x 3000 pixels) of each quadrat taken at the researcher's chest height with a Canon Powershot SX60HS camera at the closest possible angle to a 90° angle were used to calculate the percentage of each substrate in the quadrats. Posteriorly, Photographs were analyzed with the aid of 22 ImageJ 1.53K software. The distances were obtained by a laser tape measure or by the plot of each transect on the map. For the definition of the Micro Drainage Basins, the function “Channel Network and Drainage Basins” from the SAGA Next Generation plugin was used with the aid of a Digital Elevation Model (DEM) in the QGIS 3.22.11 software. The DEM was also used to extract the altitude information for each sector. The sectors with 600m or higher in elevation were classified as Recharge Zones and those under this altitude were classified as Discharge Zones. Data analysis Pre-analysis routine All the analyses were run in the R Studio 2022.07.02 Build 576 software. Prior to the analysis of invertebrate fauna composition and richness, the correlation between the variables was tested with the help of the CHART.CORRELATION function from the ‘PerformanceAnalytics’ package and the variables with correlation value > 0.70 were excluded from the models. The functions VIF and VIF.CCA from the ‘Car’ package was used to test the multicollinearity of variables, and the ones that present valor > 10 were discarded. A Shapiro-Wilk test was executed, to verify the normality of the data, using the SHAPIRO.TEST function from the package ‘Stats’. A Mantel test was performed to try the spatial autocorrelation between the samples, in mesoscale, for all the fauna types. Mesoscale abiotic features All the substrates in each sector were evaluated and classified into the following classes: guano - GU; feces - FZ; carcass - CRC; roots - RZ; litter - SER; vegetal debris - DTV (< 10mm); fine branch - GALF (11 - 30 mm); medium branch - GALM (31 - 50 mm); coarse branch - GALG (65 - 250 mm); trunk - TRO (> 250mm); termite mounts - TM; water streams - ST; water pond - WP; drip water - DP; phanerogams - FG; actinomycetes - ACT; another organic substrate - OTO; concrete floor - RC; rough rock - RR; large rock - XB (1000 - 4000mm); medium rock - MB (500 - 1000mm); small rock - SB (250 - 500mm); cobbles - CB (64 - 250mm); coarse gravel - CAG (16 - 64mm); fine gravel - CAF (2 - 16mm); sand - ARE (0.06 - 2mm); silt - SEF (≤ 0.05 mm); hardpan - HP; speleothems - ES; calcite rafts - JNS; 23 concreted calcite raft - JNC; stalactite - ESTC; stalagmite - EST; micro travertine - MTR; travertine - TRA; rough flowstone - ESR; flowstone - ESC; worm acorn - BM; retraction cracks - GRR; gastropod shell - COG. Based on those classes and using a Shannon-Weaver Index (Buttigieg & Ramette 2014), were determined the Substrate Diversity (all classes), the Shelter Diversity (WP, DP, XB, MB, SB, CB, CAG, CAF, ES, ESTC, EST, MTR, TRA, BM, GRR, COG) and the Trophic Resources Diversity (GU, FZ, CRC, RZ, SER, DTV, GALF, GALM, GALG, TRO, FG, ACT, OTO) for each sector. The classes were also used to generate, by sum, the Shelter Availability (WP, DP, XB, MB, SB, CB, CAG, CAF, ES, ESTC, EST, MTR, TRA, BM, GRR, COG) and the Trophic Resources Availability (GU, FZ, CRC, RZ, SER, DTV, GALF, GALM, GALG, TRO, FG, ACT, OTO) for each sector. For their use in analysis with individual substrate classes, a few categories were grouped in order to reduce variables. DTV, GALF, GALM, GALG, and TRO were grouped into DTV, while ES, JNS, JNC, EST, MTR, and TRA have grouped into the class ES. The abiotic attributes were then divided into Landscape features, such as Micro Drainage Basins, Water Zones, and Caves; Physical features, which comprise the Distance of each transect from the nearest entrance, Substrate Diversity, Shelter Diversity, and Shelter Availability; Trophic Resources grouped Trophic Resources Diversity and Trophic Resources Availability; the Microclimatic Variables considered were Temperature and Moisture. Micro-scale abiotic features For the micro-scale, the same substrate classes were evaluated, and the same groupings were made. The diversities and availabilities were calculated equally for the sectors. The Landscape, Physical and Trophic features were also formed similarly to the mesoscale. For the quadrats, the microclimatic variables were not measured. Habitat traits determining the communities’ richness and composition In order to understand the potential correlation between total, troglobitic, and non-troglobitic invertebrate richness with physical, trophic, and microclimatic traits, Generalized Linear Models (GLM) 24 and Generalized Linear Mixed Models (GLMM) were performed, with sectors and quadrats as sample units, using two models for each fauna group. The first model (Succinct Model) used Microclimatic Variables, Distance from the nearest entrance, Diversities (general substrate, shelter, and trophic resources), and Availabilities (shelter and trophic resources). The second model (Long Model) used Microclimatic Variables, Distance from the nearest entrance, Diversities (general substrate, shelter, and trophic resources), and each substrate class individually. For the micro-scale, the Microclimatic Variables were not used. For those models, the Poisson family was adopted because it better fitted the data. For the evaluation of model overdispersion, the function CHECK_OVERDISPERSION from the package ‘Performance’ was used. To obtain r² values of the GLMMs was used the function r.squaredGLMM from the ‘MuMIn’ package, while the function r.squaredLR from the ‘piecewiseSEM’ package was used to obtain r² values of the GLMs. To evaluate the possible correlation between total, troglobitic, and non-troglobitic invertebrate composition with landscape, physical, trophic, and microclimatic variables a Distance-Based Redundancy Analysis (dbRDA) was performed (Clarke et al. 2014), with sectors and quadrants as sample units, using two models for each fauna group. The first model (Succinct Model) used Landscape features, Microclimatic Variables, Cave, Distance from the nearest entrance, Diversities (general substrate, shelter, and trophic resources), and Availabilities (shelter and trophic resources). The second model (Long Model) used Landscape features, Microclimatic Variables, Cave, Distance from the nearest entrance, Diversities (general substrate, shelter, and trophic resources), and each substrate class individually. For the micro-scale, the Microclimatic Variables were not used. Results Richness and composition of cave invertebrates Considering all 24 caves (with samplings in the transects, quadrats, and other habitats) 2,754 specimens were found, totalizing 338 species from 37 orders and at least 93 families (Fig. 4). The most 25 expressive group was Araneae, with 66 species distributed in 17 families, totalizing 1,062 individuals. The spiders were followed by Diptera and Coleoptera, with 49 and 37 species respectively. Cave-restricted species The sampled caves presented at least 41 cave-restricted species belonging to 8 higher taxa and 14 families (Fig. 5; Table 1). The eight higher taxa were Crustacea (11), Arachnida (11), Hexapoda (10 spp.), Myriapoda (5), Mollusca (1), Nemertea (1), Annelida (1) and Osteicthyes (1) (Table 1). The richness of the observed obligate cave fauna was Gastropoda (1), Nemertea (1), Oligochaeta (1), Isopoda (10), Amphipoda (1), Polydesmida (4), Symplyla (1), Blattodea (1), Collembola (4), Coleoptera (2), Hemiptera (1), Orthoptera (1), Pseudoscorpiones (2), Araneae (4), Opiliones (3), Palpigradi (2), and Siluriformes (1). Is very important to highlight that from all 41 troglobitic species, 25 are found in the Gruta do Padre Cave, which represents more than 60% of the species. Furthermore, only 14% of the obligate cave species from this area are currently described. Such described species are Coarazuphium tessai (Godoy & Vanin 1990); Phaneromerium cavernicolum (Golovatch & Wytwer 2004), Spelaeogammarus santanensis (Koenemann & Holsinger 2000), Eusarcus cf. cavernicola (Hara & Pinto-da-Rocha 2010), Pectenoniscus santanensis (Cardoso et al. 2020) and Chaimowiczia tatus (Cardoso et al. 2021). Although Hara & Pinto-da-Rocha (2010) considered that E. cavernicola may represent an assembly of species that cannot be identified by external and genital characteristics, herein we considered the population of the Gruta do Padre cave (and other related caves) as a troglobitic species, due to the strong troglomorphic traits they present. Habitat traits determining the communities’ richness and composition On the mesoscale, the Mantel test revealed the existence of spatial autocorrelation for the general and non-troglobitic fauna (p=0.0004); therefore, the geographic distances between transects explain 20.43% and 20.17% respectively of the variation in the composition. For the troglobitic fauna, the geographic distances were not significant. 26 For the composition of communities, the Distance from the nearest entrance was significant for all models, both at the meso and micro-scale. The Cave itself was also an important variable, presenting significance except for the troglobitic fauna at the microscale. Shelter Availability was significant only for the general and troglobitic invertebrate fauna in the mesoscale. On the other hand, Trophic Availability, , was only related to the general and non-troglobitic fauna at the micro-scale. Trophic Diversity showed a significant correlation only with the non-troglobitic fauna on the micro-scale, for both long and succinct models. Concerning individualized substrates, guano (GU), vegetal debris (DTV), and course gravel (CAG) were important in some of the models. GU only was significant at the microscale for general and non-troglobitic fauna, while DTV was only important for the general fauna on the microscale. CAG was significant for both non-troglobitic fauna at the mesoscale, and troglobitic fauna at the micro-scale (Table 2). For community richness, the Distance from the nearest entrance did not show significance only for the troglobitic fauna in the long model of the mesoscale. The correlation between Distance from the nearest entrance and Richness was positive for the troglobitic fauna and negative for general and non- troglobitic. The Temperature always presented a negative correlation and was significant for the general and non-troglobitic fauna of the succinct model, and general fauna of the long model. The Trophic Availability was only important for the general fauna at the micro-scale, presenting a positive correlation (Table 3). For the individual substrates, GU had a negative correlation with the troglobitic fauna on the mesoscale, but showed a positive correlation with general and non-troglobitic fauna on the micro-scale, similar to DTV. Carcass (CRC), Phanerogams (FG), Sand (ARE), and Worm Acorn (BM) were also found to be positively significant, whereas Hardpan (HP) and Speleothems (ES) were negatively correlated, affecting general and non-troglobitic fauna at the mesoscale. On the mesoscale, Actinomycetes (ACT) were positively significant for the troglobitic fauna, while small rock (SB) and flowstone (ESC) were important for the general fauna, with negative and positive correlations, respectively. On the microscale, concrete floor (RC) showed a positive correlation with the non- 27 troglobitic fauna, while rough rock (RR) and retraction cracks (GRR) had negative significance for general and troglobitic fauna, respectively. Discussion Distance from the entrance and its relationship with communities’ traits For all habitat traits analyzed, the distance from the nearest entrance stood out as the most important. In terms of community richness, only the troglobitic fauna from the long model at the mesoscale did not respond to the distance. In all other models, the general and non-troglobitic fauna responded negatively to an increase in the distance from the nearest entrance, while the troglobitic fauna showed a positive response. For species composition, the Distance from the nearest entrance was important for all types of fauna, models, and scales, demonstrating its strong relationship with the fauna distribution. It is important to note that most of the general fauna is composed of non-troglobitic species, which may bias the results towards this group. The decrease observed in the richness of general and non-troglobitic fauna is the anticipated response for this trait, given that the distance from the cave entrance is a well- known limiting factor for species distribution. The restrictive influence of the distance from the entrance is linked to the reduction of trophic resources, as well as the habitat heterogeneity, which decreases from near-to-entrance to deep zones of the caves (Tobin et al. 2013; Moseley 2008; Ficetola et al. 2018; Mammola 2019; Souza-Silva et al. 2021). In this way, greater distances limit the amount of accessible energy for the communities, as the majority of cave trophic resources are transported from the entrances (Tobin et al. 2013; Moseley 2008; Ficetola et al. 2018; Mammola 2019; Souza-Silva et al. 2021, Furtado et al. 2022). Additionally, since zones closer to the entrances present less stable climatic conditions, a gradient of temperature, moisture, and sunlight is created. This gradient can also be observed on the cave floor, leading to the simplification and homogenization of substrates in deeper zones (Prous et al. 2004; Prous et al. 2015; Souza-Silva et al. 2021; Furtado et al. 2022). 28 On the other hand, the restricted subterranean fauna richness shows a contrary response to that of the general and non-troglobitic fauna. The higher climatic stability found in the deeper zones of the caves favors the existence of troglobitic fauna due to their adaptations. These adaptations include reduced metabolic rates and cuticle thinning, which increase the risk of desiccation and may limit the distribution of these species to areas with minor temperature and moisture variations, often found in deeper areas (Tobin et al. 2013; Lunghi et al. 2014 and 2017; Kozel et al. 2019; Souza-Silva et al. 2021). This preference of troglobitic organisms for stabler habitats can, on the other hand, be associated with a cave characteristic initially thought to limit the communities, the oligotrophy. However, as demonstrated by Hüppop (2005), the K-strategy life history adopted by these animals, combined with their reduced metabolic rates, enables troglobitic fauna to survive for extended periods without food. Thus, restricted fauna is commonly more prevalent in areas with fewer trophic resources, allowing them to avoid non-troglobitic competitors (Sket 1999; Deharveng & Bedos 2000; Souza-Silva et al. 2021). Is important to emphasize that an increase in organic matter in these oligotrophic zones can be detrimental to troglobites, as it may attract more competitive and energetically needed non-troglobites (Sket 1999; Souza-Silva et al. 2021). Similarity between caves thus geographic position The Mantel test performed revealed that geographic distances between transects account for 20.43% of the variation in the composition of the general fauna and20.17% on the non-troglobitic fauna communities. This indicates that sampling units that are closer to each other exhibit a greater similarity in fauna than those that are farther apart. As demonstrated in previous studies, spatial autocorrelation can imply that cave communities can affect the composition of adjacent subterranean habitats (Christman et al. 2015; Jaffé et al. 2016; Jaffé et al. 2018). The findings of this study support this notion, but only for the general and non-troglobitic fauna, as the troglobitic composition did not exhibit any spatial autocorrelation. This lack of influence of geographic distance on the restricted fauna may suggest that, unlike in other studies, this region does not have a highly interconnected subterranean environment, or at least the 29 troglobitic species are not utilizing these connections (Ferreira 2005; Souza-Silva et al. 2011; Auler et al. 2014; Christman et al. 2005; Jaffé et al. 2016; Jaffé et al. 2018). On the other hand, the spatial autocorrelation observed in the composition of general and non- troglobitic fauna suggests that, despite the subterranean connections between caves being unused, non- restricted species may still disperse in epigean environments (Ribera et al. 2019). These surface movements by animals raise concerns for the overall conservation of the area, as degraded regions can obstruct the dispersal of fauna and negatively impact the subterranean diversity. Is important to note that spatial autocorrelation, and thus the potential for epigean dispersal movements of fauna, can account for approximately 20% of the variation in composition. However, a higher explanatory value is attributed to the grouping of other analyzed variables, such as physical, trophic, and climatic. Invertebrates’ response to habitat heterogeneity The trophic and physical attributes play a highly significant role in defining cave invertebrate communities in the semi-arid regions of Brazil.. Since caves are naturally oligotrophic environments, they depend greatly on the surface, with the majority of the energy of the system originating from epigean zones (Tobin et al. 2013; Moseley 2008; Ficetola et al. 2018; Mammola 2019). The presence and diversity of trophic resources are not only important for the existence of cave invertebrate communities, but they also affect their composition and richness. This suggests that different types of organic matter may provide a higher number of specific niches, allowing for greater richness and creating differences in composition. These findings align with previous research on non- troglobitic fauna (Schneider et al. 2011; Souza-Silva et al. 2011; Ladle et al. 2012; Ferreira 2019; Pacheco et al. 2020a; Furtado et al. 2022). Among the different types of organic matter observed in the sampled caves, bat guano and vegetal debris were found to be the most significant. Bat guano serves as one of the primary energy sources for invertebrate fauna, particularly in caves that are permanently dry (Ferreira & Martins 1999; Souza-Silva et al. 2011). However, the guano production in caves may vary seasonally due to external vegetation 30 fluctuations and, as a result, food availability for bats (Faria 1996; Souza-Silva et al. 2011). The transport of guano into caves is crucial because the piles formed by bats can sustain entire invertebrate communities that vary depending on the age and composition of the deposit (Ferreira & Martins 1999; Ferreira 2019). Furthermore, guano deposits may be critical for establishing richer cave communities, as they attract colonizer species that initiate a complex process, resulting in more extensive trophic webs (Ferreira & Martins 1999; Ferreira 2019; Pacheco et al. 2020a). Due to the prevalence of deciduous species in the limestone vegetation of this region, a significant amount of leaves and small branches accumulate in the litter during the dry season (Crowther 1987; Brina 1998; Souza-Silva et al. 2011). In turn, this accumulated vegetal debris frequently finds its way into caves via flood pulses during the rainy season. These floods typically carry substantial volumes of water downstream, transporting organic matter and other substrates into caves (Minshall et al. 1983; Carrling 1987; Downes & Street 2005). Therefore, as shown by Souza-Silva et al. (2011), the dependence of cave communities on external vegetal material causes the trophic dynamics of subterranean habitats to be influenced by seasonal changes outside the cave environment. The positive correlation observed between cave invertebrate communities and organic matter is generally applicable to non-restricted fauna. However, when considering troglobitic species, this correlation may not hold. Restricted subterranean animals typically have reduced metabolic rates and K-strategy life history, enabling them to endure periods of starvation (Hüppop 2005; Souza-Silva et al. 2021). Consequently, due to the greater energetic demand and superior competitive ability of non- troglobitic species, troglobites tend to seek for locations with limited organic matter, thus avoiding competitors (Sket 1999; Deharveng & Bedos 2000; Souza-Silva et al. 2021). It is essential to emphasize that the distribution of trophic resources inside caves is not uniform, with the majority concentrated near the entrance, creating an energy gradient that decreases towards deeper areas (Tobin et al. 2013; Moseley 2008; Ficetola et al. 2018; Mammola 2019; Souza-Silva et al. 2021, Furtado et al. 2022). This gradient encompasses not only trophic resources but also physical and 31 microclimatic conditions (Prous et al. 2004; Prous et al. 2015; Souza-Silva et al. 2021; Furtado et al. 2022). The data obtained from this study demonstrate that the presence of different types of substrates can have a positive effect on the composition and richness of cave invertebrate communities. This supports the notion that greater substrate diversity and variation in climatic traits generate more microhabitats, enabling a greater number of species to occupy the habitats by creating new niches (Prous et al. 2004; Prous et al. 2015; Souza-Silva et al. 2021; Furtado et al. 2022). However, troglobitic species respond differently from non-troglobitic fauna, displaying distinct responses to different substrates. This contrasting response may be attributed to the high level of specialization of restricted organisms, which seek specific conditions within subterranean habitats (Pacheco et al. 2020a; Souza-Silva et al. 2021). Conservation of subterranean environments in Santana region The Santana region lies in a transition zone between the Caatinga and Seasonal Dry Forests. While some caves have entrances that are covered by native vegetation, it is evident that the original forests surrounding the caves have been replaced by pastures and other monocultures. Deforestation in the caves’surroundings can have a direct and indirect impact on the energetic dynamics of subterranean invertebrate communities. The loss of vegetation in the cave surroundings can directly reduce the amount of litter available to be transported to hypogean environments. This litter, derived from vegetation, is one of the most significant food sources for cave invertebrates in the region, and its depletion can disrupt the entire subterranean trophic web (Crowther 1987; Brina 1998; Souza-Silva et al. 2011). Indirectly, deforestation can limit the availability of bat food, which may decrease the input of guano to the caves. Bat guano is one of the most critical energy sources for invertebrates in permanently dry caves, and its absence can pose a threat to entire communities (Faria 1996; Ferreira & Martins 1999; Souza-Silva et al. 2011; Ferreira 2019). Despite the high speleological potential in the Santana region, many locals are unfamiliar with most of the caves, even the larger ones like the Gruta do Padre Cave. Fear or lack of opportunities may contribute to this lack of awareness, which is not unique to the region but extends to the Brazilian 32 population as a whole. It is estimated that there are almost 23 thousand caves registered in the country, but this represents only a small fraction of the real potential, which is around 300 thousand caves (CECAV 2022). Similarly, in the Santana region, unexplored limestone outcrops are easy to find. This study alone identified nine previously unregistered caves out of 25 sampled, highlighting significant gaps and unprospected areas. It is crucial to emphasize that the general lack of knowledge among the population, both nationally and in Santana, is also attributable to the research community as a whole, not just in Brazil but worldwide. Regrettably, many researchers overlook the local population when conducting their work, neglecting to communicate scientific findings in a more accessible manner, which is equally true for speleology. Clearly, legislation plays a crucial role in protecting the speleological heritage. However, despite regulations and management controls, some damage to cave ecosystems is still caused by locals, and effective monitoring is a challenging task in a vast country like Brazil. Sometimes, due to lack of awareness or for basic survival needs, locals can inadvertently cause harm to subterranean environments and their surroundings. Therefore, it is imperative to provide accessible information about the legislation and the significance of cave habitats for the local communities in Santana's region. This could foster a sense of ownership and stewardship among locals, and they could become new and powerful allies in the conservation of the unique and fragile cave ecosystems of the Brazilian semi-arid region. Acknowledgments The authors would like to thank the Centro Nacional de Pesquisa e Conservação de Cavernas - CECAV and Instituto Brasileiro de Desenvolvimento e Sustentabilidade - IABS (n°. 006/2021. TCCE ICMBio/Vale (01/2018) for the financial support; CNPq (National Council for Scientific and Technological Development) for the productivity scholarship provided to RLF (CNPq n. 302925/2022- 8); to the Vale Company and the team from the Center of Studies in Subterranean Biology (CEBS/UFLA) for the support in the field trips. VGPJ is grateful to Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for the master’s degree scholarship granted. We are also thankful 33 to Giovanna Cardoso, Guilherme Prado, and Leopoldo Bernardi for their help in the identification of some taxa. References Auler A, Piló L, Parker C, Senko J, Sasowsky I, Barton H. 2014. Hypogene cave patterns in iron ore caves: convergence of forms or processes? In: Klimchouk A, Sasowsky ID, Mylroie J, Engel SA, Engel AS, eds. Hypogene cave morphologies. Leesburg: Karst Water Institute, 15–19 Auler AS, Rubbioli E, Menin D, Brandi R (2019) Cavernas. Atlas do Brasil Subterrâneo. Editora IABS, Brasília. Brina A. E. 1998. Aspectos da dinâmica da vegetação associada a afloramentos calcários na APA carste de Lagoa Santa, MG. Dissertação de Mestrado em Ecologia Conservação e Manejo da Vida Silvestre, UFMG. 105pp Buttigieg PL, Ramette A (2014) A Guide to Statistical Analysis in Microbial Ecology: a community- focused, living review of multivariate data analyses. FEMS Microbiol Ecol 90:543–550. https://doi. org/10.1111/1574-6941.12437 Carling PA (1987) Bed stability in gravel streams, with reference to stream regulation and ecology. In: Richards, K. (ed.) River Channels: Environment and Process. Blackwell, Oxford, pp. 321–347. Centro Nacional de Pesquisa e Conservação de Cavernas – CECAV (2022) Cadastro Nacional de Informações Espeleológicas (CANIE) Christman MC, Culver DC, Madden MK, White D (2005) Patterns of endemism of the eastern North American cave fauna. Journal of Biogeography 32:1441–1452 DOI 10.1111/j.1365-2699.2005.01263.x. Clarke KR, Gorley RN, Somerfeld PJ, Warwick RM (2014) Change in marine communities: an approach to statistical analysis and interpretation, 3rd ed. PRIMER-E, Plymouth Crowther J. 1987. Ecological Observations in Tropical Karst Terrain, West Malaysia. II. Rainfall Interception, Litterfall and Nutrient Cycling. Journal of Biogeography, 2, (14):145-155. 34 Culver DC, Sket B (2000). Hotspots of subterranean biodiversity in caves and wells. Journal of Cave and Karst Studies. 62:11–17 Culver DC, Pipan T (2009) Biology of Caves and Other Subterranean Habitats, 1st ed. Oxford, New York Deharveng L, Bedos A (2000) The cave fauna of southeast Asia. In: Wilkens H, Culver DC, Humphries WF (eds) Origin, evolution and ecology—subterranean ecosystems. Elsevier, Amsterdam, pp 603–632. Dinerstein E (2017) An Ecoregion-Based Approach to Protecting Half the Terrestrial Realm. BioScience, [s. l.], 67, 6, 534–545. https://doi.org/10.1093/biosci/bix014 Downes, B.J. and Street, J.L. 2005. Regrowth or dispersal? Recovery of a freshwater red alga following disturbance at the patch scale. Austral Ecology 30, 526–536. Du Preez G, Forti P, Jacobs G, Jordaan A, Tiedt LR (2015) Hairy Stalagmites, a new biogenic root speleothem from Botswana. Int J Speleol 44(1):37–47. https://doi.org/10.5038/1827-806X.44.1.4 Ferreira RL, Martins RP (1999) Trophic structure and natural history of bat guano invertebrate communities with special reference to Brazilian caves. Tropical Zoology, 12:231-259. https://doi.org/10.1080/03946975.1999.10539391 Ferreira RL (2004). A medida da complexidade ecológica e suas aplicações na conservação e manejo de ecossistemas subterrâneos. Tese de doutorado, Universidade Federal de Minas Gerais, 158p. Ferreira RL (2005) A vida subterrânea nos campos ferruginosos. O Carste 3:106–115. Ficetola GF, Lunghi E, Canedoli C, Padoa-Schioppa E, Pennati R, Manenti R (2018) Diferences between microhabitat and broad-scale patterns of niche evolution in terrestrial salamanders. Sci Rep 8(1):1– 12. https://doi.org/10.1038/s41598-018-28796-x Ferreira RL (2019) Guano Communities. In: Encyclopedia of Caves (Third Edition). Academic Press. 474- 484. https://doi.org/10.1016/B978-0-12-814124-3.00057-1. Furtado L, Ferreira RL, Fernández JIR, Souza-Silva M (2022) Recreational caving impact of visitors in a high-altitude cave in Bolivian Andes: Main effects on microhabitat structure and faunal distribution. International Journal of Speleology, 51(2), 93-103. https://doi.org/10.5038/1827-806X.51.2.2418 https://doi.org/10.1093/biosci/bix014 https://doi.org/10.5038/1827-806X.44.1.4 https://doi.org/10.1038/s41598-018-28796-x 35 Howarth FG (1980) The zoogeography of specialized cave animals: a bioclimatic model. Evolution 394- 406. Howarth FG (1983) Ecology of cave arthropods. Annual Review of Entomology 28:365-389. https://doi.org/10.1111/j.2041-210X.2009.00001.x Hüppop K (2005) Adaptation to low food. In: Culver D, White WB (ed) Encyclopedia of caves. Ensevier, Amsterdam, pp 4–10. Jaffé R, Prous X, Zampaulo R, Giannini T, Imperatriz-Fonseca V, Maurity C, Oliveira G, Brandi I, Siqueira J (2016) Reconciling mining with the conservation of cave biodiversity: a quantitative baseline to help establish conservation priorities. PLOS ONE 11:e0168348 DOI 10.1371/journal.pone.0168348 Jaffé R, Prous X, Calux A, Gastauer M, Nicacio G, Zampaulo R, et al (2018) Conserving relics from ancient underground worlds: assessing the influence of cave and landscape features on obligate iron cave dwellers from the Eastern Amazon. Peer J 6:e4531; DOI 10.7717/peerj.4531 Juberthie C, Delay B, Bouillon M (1980) Sur l’existence d’un milieu souterrain superficiel en zone non calcaire. C. R. Acad. Sc. Paris. 1 (290) 49–52. Koppen W (1936) Das geographische system der klimat. Handbuch der klimatologie. 46. Kozel P, Pipan T, Mammola S, Culver DC, Novak T (2019) Distributional dynamics of a specialized subterranean community oppose the classical understanding of the preferred subterranean habitats. Invertebr Biol 138(3):e12254. https://doi.org/10.1111/ivb.12254 Ladle RJ, Firmino JV, Malhado AC, Rodriguez-Duran A (2012) Unexplored diversity and conservation potential of Neotropical hot caves. Conservation Biology, 26(6),978-982. Lunghi E, Manenti R, Ficetola GF (2014) Do cave features affect underground habitat exploitation by non- troglobite species? Acta Oecol 55:29–35. https://doi.org/10.1016/j.actao.2013.11.003 Lunghi E, Manenti R, Ficetola GF (2017) Cave features, seasonality, and subterranean distribution of non- obligate cave dwellers. PeerJ 5:e3169. https://doi.org/10.7717/peerj.3169 Lunghi E, Manenti R (2020) Cave communities: from the surface border to the deep darkness. Diversity 12(5):167. https://doi.org/10.3390/d12050167 https://doi.org/10.1111/j.2041-210X.2009.00001.x 36 Mammola S, Giachino PM, Piano E, Jones A, Barberis M, Badino G & Isaia M (2016) Ecology and sampling techniques of an understudied subterranean habitat: the Milieu Souterrain Superficiel (MSS). The Science of Nature, 103: 88. https://doi.org/10.1007/ s00114-572 016-1413-9 Mammola S, Isaia M (2017) Spiders in caves. Proc R Soc B. https://doi.org/10.1098/rspb.2017.0193 Mammola S, Isaia M (2018) Day-night and seasonal variations of a subterranean invertebrate community in the twilight zone. Subterr Biol 27:31–51. https://doi.org/10.3897/subtbiol.27.28909 Mammola S (2019) Finding answers in the dark: caves as models in ecology fifty years after Poulson and White. Ecography 42(7):1331–1351. https://doi.org/10.1111/ecog.03905 Mammola S, Piano E, Cardoso P, Vernon P, Domínguez-Villar D, Culver Dc, Pipan T, Isaia M (2019) Climate change going deep: The effects of global climatic alterations on cave ecosystems. The Anthropocene Review. 6 (1–2), 98–116. https://doi.org/10.1177%2F2053019619851594 Mammola S, Arnedo MA, Fišer C, Cardoso P et al (2020) Environmental filtering and convergent evolution determine the ecological specialization of subterranean spiders. Funct Ecol 35(5):1064–1077. https://doi.org/10.1111/1365-2435.13527 Minshall GW, Andrews DA, Manuel-Faler CY (1983) Application of island biogeographic theory to streams: macroinvertebrate recolonization of the Teton River, Idaho. In: Barnes, J.R. and Minshall, G.W. (eds) Stream Ecology: Application and Testing of General Ecological Theory. Plenum Press, New York, pp. 279–297. Moseley M (2008) Size matters: scalar phenomena and a proposal for an ecological definition of ‘cave’. Cave Karst Science 35(3):89–94 Moseley M (2009) Are all caves ecotones? Cave Karst Science 36(2):53–58 Odum EP, Barrett GW (2006) Fundamentals of Ecology. Brooks Cole, Thomson Learning. Inc., Bangalore, India. Oliver I, Beattie AJ (1996) Invertebrate morphospecies as surrogates for species: a case study. Conservation Biology, 10(1):99–109 https://doi.org/10.1111/ecog.03905 https://doi.org/10.1111/1365-2435.13527 37 Pacheco GSM, Souza-Silva M, Cano E, Ferreira LF (2020) The role of microhabitats in structuring cave invertebrate communities in Guatemala. International Journal of Speleology, 49 (2), 161–169. https://doi.org/10.5038/1827-806X.49.2.2333 Prous X, Ferreira RL, Martins RP (2004) Ecotone delimitation: Epigean–hypogean transition in cave ecosystems. Austral Ecol 29(4):374–382. https://doi.org/10.1111/j.1442-9993.2004.01373.x Prous X, Ferreira RL, Jacobi CM (2015) The entrance as a complex ecotone in a Neotropical cave. Int J Speleol 44:177–189. https://doi.org/10.5038/1827-806X.44.2.7 QGIS Development Team (2023) QGIS Geographic Information System. Open-Source Geospatial Foundation Project. http://qgis.osgeo.org Rabelo LM, Souza-Silva M, Ferreira RL (2020) Epigean and hypogean drivers of Neotropical subterranean communities. Journal of Biogeography. 1 https://doi.org/10.1111/jbi.14031 Racovitza EG (1907) Essai sur les problèmes biospéologiques. Archives de Zoologie Expérimentale et Générale, 4(6):371–488. Rasband WS (1997) Image, J., US National Institutes of Health. “ImageJ64,” US National Institutes of Health, Bethesda, MD R Core Team (2019) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. https://www.R-project.org/ Romero A, Green SM (2005) The end of regressive evolution: examining and interpreting the evidence from cave fishes. Journal of fish biology. 67(1): 3–32. Schneider K, Christman MC, Fagan WF (2011) The influence of resource subsidies on cave invertebrates: results from an ecosystem-level manipulation experiment. Ecology, 92, 765-776. https://doi.org/10.1890/10-0157.1 Sket B (1999) The nature of biodiversity in hypogean waters and how it is endangered. Biodivers Conserv 8:1319–1338. https://doi.org/10.1023/A:1008916601121 Sket B (2008) Can we agree on an ecological classifcation of subterranean animals? J Nat Hist 42:1549– 1563. https://doi.org/10.1080/00222930801995762 https://doi.org/10.5038/1827-806X.49.2.2333 http://qgis.osgeo.org/ https://doi.org/10.1890/10-0157.1 38 Sobral-Souza T, Lima-Ribeiro MS, Solferini VN (2015) Biogeography of Neotropical Rainforests: past connections between Amazon and Atlantic Forest detected by ecological niche modeling. Evolutionary Ecology. 29(5):643–655. https://doi.org/10.1007/s10682-015-9780-9 Souza-Silva M, Parentoni RM, Ferreira RL (2011b) Trophic dynamics in a neotropical limestone cave. Subterranean Biology 9:127–138. https://doi.org/10.3897/subtbiol.9.2515 Souza-Silva M, Cerqueira, RFV, Pellegrini TG, Ferreira RL (2021) Habitat selection of cave-restricted fauna in a new hotspot of subterranean biodiversity in Neotropics. Biodiversity and Conservation, 1–28. https://doi.org/10.1007/s10531-021-02302-8 Stein A, Beck J, Meyer C, Waldmann E, Weigelt P, Kreft H (2015) Differential effects of environmental heterogeneity on global mammal species richness. Global Ecol Biogeogr 24 (9), 1072–1083. https://doi.org/10.1111/geb.12337 Tews J, Brose U, Grimm V, Tielbörger K, Wichmann MC, Schwager M, Jeltsch F (2004) Animal species diversity driven by habitat heterogeneity/diversity: the importance of keystone structures. Journal of biogeography, 31(1), 79-92. Tobin BW, Hutchins BT, Schwartz BF (2013) Spatial and temporal changes in invertebrate assemblage structure from the entrance to deep-cave zone of a temperate marble cave. International Journal of Speleology 42:203-214. https://doi.org/10.5038/1827-806X.42.3.4 Vargas-Mena JC, Cordero-Schmidt E, Rodriguez-Herrera B (2020) Inside or out? Cave size and landscape effects on cave-roosting bat assemblages in Brazilian Caatinga caves. J Mammal 101 (2), 464–475. https://doi.org/10.1093/jmammal/gyz206 Wynne JJ, Howarth FG, Sommer S, Dickson BG (2019) Fifty years of cave arthropod sampling: techniques and best practices. Int J Speleol 48(1):33–48. https://doi.org/10.5038/1827-806X.48.1.2231 Yang Z, Liu X, Zhou M, Ai D, Wang G (2015) The effect of environmental heterogeneity on species richness depends on community position along the environmental gradient. Sci Rep 5(1):1–7. https:// doi.org/10.1038/srep15723 https://doi.org/10.1007/s10682-015-9780-9 https://doi.org/10.1111/geb.12337 https://doi.org/10.5038/1827-806X.48.1.2231 39 40 Figure 1. Location map of the caves sampled in Santana region. Caves identification are in table 5. 41 Figure 2. Sampling method. A–sampling method scheme showing both meso and microscale; B–sampling method being apply in one of the caves. 42 Figure 3. 15 of the 41 troglobitic species found in Santana region. A– Pectenoniscus santanensis; B– Chaimowiczia tatus; C– Chaimowiczia tatus; D– Pseudochthonius sp1; E– Ideoroncidae sp1; F– Eukoenenia sp2; G– Ochyroceratidae sp1; H– Caponidae sp1; I– Ochyroceratidae sp2; J– Escadabiidae sp1; K– Endecous sp1; L– Blattidae sp1; M– Arrhopalitidae sp1; N– Phaneromerium cavernicolum; O– Coarazuphium tessai. 43 Figure 4. Main invertebrate groups richness of Santana region caves. Figure 5. Richness of troglobitic species groups by sample scale in Santana region caves. There are 41 restricted species in total. 44 Figure 6. Distance-based Redundancy Analysis (dbRDA) on the mesoscale, succinct (A, C, E) and long (B, D, F) models. A and B–general fauna; C and D–troglobitic fauna; E and F–non-troglobitic fauna. 45 46 Figure 7. Distance-based Redundancy Analysis (dbRDA) on the microscale, succinct (A, C, E) and long (B, D, F) models. A and B–general fauna; C and D–troglobitic fauna; E and F–non-troglobitic fauna. Taxons Species and Morphotypes Ca Qu Sec Amphipoda Spelaeogammarus santanensis + Aranae Ochyroceratidae sp1 + + + Ochyroceratidae sp2 + + 47 Table 1. Troglobitic species and the sample scale where they were found. Ochyroceratidae sp3 + + Prodidomidae sp2 + Caponidae sp1 + Blattodea Blattidae sp1 + Coleoptera Clivina sp1 + + Coarazuphium tessai + Entomobryomorpha Paronellidae sp2 + + + Gastropoda Gastropoda sp3 + + Oligochaeta Lumbricina sp3 + Hemiptera Kinnaridae sp1 + + + Isopoda Chaimowiczia tatus + Pectenoniscus santanensis + + + Platyartridae sp1 + + Styloniscidae sp1 + + + Styloniscidae sp2 + + Pectenoniscus sp3 + Philosciidae sp1 + Trichorhina sp2 + Calabozoidea sp1 + Xangoniscus sp1 + + Nemertea Nemertea sp1 + Opiliones Escadabiidae sp1 + + Escadabiidae sp2 + Eusarcus cavernicola + + + Orthoptera Endecous sp1 + + + Palpigradi Eukoenenia sp1 + + Eukoenenia sp2 + + Poduromorpha Poduromorpha sp1 + Pseudoscorpiones Pseudochthonius sp1 + + + Garypoidea sp1 + + + Polydesmida Phaneromerium sp1 + + + Phaneromerium sp2 + Phaneromerium sp3 + + + Phaneromerium sp4 + Symphyla Symphyla sp1 + Symphypleona Arrhopalitidae sp1 + + + Arrhopalitidae sp2 + Siluriformes Pimelodella sp1 + 48 Table 2. P-values for Distance-based Redundancy Analysis (dbRDA). Table 3. P-values and Estimate values for the GLM and GLMM on the mesoscale. Variables Mesoscale Microscale Succinct Model Long Model Succinct Model Long Model General T n-T General T n-T General T n-T General T n-T Cave* 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 Distance* 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005 Shelter Av. 0.045 0.050 Trophic Av. 0.005 0.005 Trophic Div. 0.035 0.005 GU 0.005 0.005 DTV 0.005 CAG 0.025 0.050 Explanation 41.14% 33.51% 45.74% 39.75% 33.51% 45.75% 24.31% 9.52% 36.86% 24.66% 10.29% 36.63% Mesoscale Varibles Succinct Model Long Model General T n-T General T n-T P Est. P Est. P Est. P Est. P Est. P Est. Distance* 1.61E-05 -0.273 0.003 0.265 1.79E-09 -0.553 1.73E-07 -0.350 3.09E-14 -0.671 Temperature 0.010 -0.156 0.007 -0.187 0.040 -0.130 Trophic Av. GU 0.026 -3.349 DTV CRC 0.020 0.070 0.033 0.065 FG 0.001 0.085 1.00E-04 0.095 ACT 0.041 1.895 RC RR SB 0.049 -0.096 ARE 1.028 0.111 0.021 0.134 HP 0.035 -0.107 0.038 -0.112 ES 0.015 -0.133 0.007 -0.156 ESC 0.022 0.113 BM 0.027 0.114 0.013 0.155 49 Table 4. P-values and Estimate values for the GLM and GLMM on the microsocale Microscale Varibles Succinct Model Long Model General T n-T General T n-T P Est. P Est. P Est. P Est. P Est. P Est. Distance* 8.44E-08 - 0.390 2.79E-08 0.507 7.64E-13 - 1.055 6.45E-09 - 0.390 1.40E-06 0.475 1.85E-14 - 1.091 Temperature Trophic Av. 0.001 0.120 GU 0.015 0.075 0.003 0.092 DTV 0.001 0.119 3.00E-04 0.142 CRC FG ACT RC 0.029 0.121 RR 0.041 - 0.116 SB ARE HP ES ESC BM GRR 0.047 - 0.471 R²/R²M 24.34% 10.55% 64.15% 28.82% 16.25% 48.00% R²C 39.40% 72.21% GRR R²/R²M 38.80% 15.74% 60.42% 56.05% 31.52% 79.38% R²C 65.73% 80.71% 60.86% 50 Table 5. The 24 sampled caves’ coordinates. Sirgas 200; UTM 23S. Map ID Cave Long. X Lat. Y 1 Gruta Cânion da Baixa Verde Cave 596844 8537362 2 Gruta do Padre Cave 601311 8538762 3 Gruta Labirinto do Toxodon Cave 608884 8538822 4 Gruta do Boqueirao Cave 597800 8537451 5 Gruta da Pedra Escrevida I Cave 612496 8532062 6 Gruta das Duas Cobras Cave 608351 8537716 7 Gruta do Tunel II Cave 609390 8535922 8 Gruta São Geraldo Cave 609156 8535758 9 Olho D'água do Cumbra Cave 601050 8529525 10 Racha Bovina Cave 615975 8531872 11 Gruta do Tunel I Cave 609348 8535882 12 Gruta Couve-Flor Cave 609111 8535870 13 Gruta do Geraldo Cruz Cave 600656 8528434 14 Fenda Obliqua Cave 600891 8528568 15 Gruta do Cedro Cave 601082 8538937 16 Gruta do Cedrão Cave 601040 8539058 17 Gruta do Cedrículo Cave 601044 8539067 18 Gruta do Pajeú Cave 598760 8526941 19 Gruta Cristal Cave 599497 8527612 20 Gruta do Salobro Cave 588687 8554258 21 Gruta da Grota Cave 587858 8561161 22 Gruta do Leão Cave 601365 8539161 23 Gruta Cinquentona Cave 605268 8538657 24 Gruta da Pedra Escrevidinha Cave 605264 8538648 51 Artigo 2: Habitat selection of cave invertebrates in a new South American hotspot of subterranean biodiversity Artigo redigido conforme as normas do periódico Biodiversity and Conservation. 52 Habitat selection of cave invertebrates in a new South American hotspot of subterranean biodiversity Vitor Gabriel Pereira Junta¹, Marconi Souza Silva¹, Rodrigo Lopes Ferreira¹* ¹Center of Studies in Subterranean Biology, Subterranean Biodiversity Sector, Ecology and Conservation Department, Natural Sciences Institute, Federal University of Lavras, Lavras, MG, Brazil *Corresponding Author vitor.junta@outlook.com marconisilva@ufla.br drops@ufla.br Abstract The Gruta do Padre Cave, the fifty largest cave in Brazil, is the home of 25 cave-restricted species, thus becoming the fourth hotspot of subterranean biodiversity in South America. To understand the fauna associated with this cave were performed composition and richness analysis to access how these cave communities respond to habitat characteristics, such as climatic variables, different substrates, presence of shelters, and food resources. The results demonstrate that the Distance from the nearest entrance and the Zone within the cave where the transects are placed are the main factors for fauna distribution. Also, habitat heterogeneity demonstrates a correlation with the fauna richness and composition, with shelter and food availability as decisive traits. The information accomplished herein demonstrates the great importance of the conservation of Gruta do Padre Cave as a new hotspot of subterranean biodiversity and highlights the urge to protect the cave in all its complexity, since this is a very heterogeneous cave with unique habitats within it. 53 Introduction Several studies have demonstrated that the cave linear development tends to be strongly related to the species richness in such environments (Simões et al. 2015; Souza-Silva et al. 2020; Souza-Silva et al. 2021; Rabelo et al. 2021). This relation is probably because bigger caves tend to present both greater areas and higher habitat heterogeneity. However, the large linear development can be limiting, especially in caves with few entrances. It is well known that caves rarely present primary production, excepting a few subterranean chemautotrophically based ecosystems (Sarbu & Kane 1996; Galassi et al. 2017) and entrance zones (Souza-Silva et al. 2011; Prous et al. 2015). Thus, the main source of energy in caves comes from external environments, corresponding to the organic matter transported by physical or biological agents (Souza-Silva et al. 2011; Ferreira 2019). Hence, in large caves with few entrances, the organic matter hardly reaches deep areas, limiting the available energy for the communities (Tobin et al. 2013; Moseley 2008; Ficetola et al. 2018; Mammola 2019). As well as the large extension of the caves, the presence of rivers also contributes to the maintenance of richer communities (Simões et al. 2015). Rivers and streams act as trophic resource carriers, especially in the case of allogenic drainages, taking energy from external habitats and bringing it to deeper cave zones. Additionally, microclimate traits, such as temperature and moisture, are usually linked to the presence of great water bodies (Souza-Silva et al. 2011b; Lobo et al. 2015; Simões et al. 2015; Souza-Silva et al. 2020). Furthermore, despite the general stability observed in caves, a gradient of climate conditions is usually observed from near-to-entrance zones (which are more variable) to deeper areas (Moseley 2009; Tobin et al. 2013; Lunghi et al. 2014; Prous et al. 2015; Mammola and Isaia 2018; Lunghi and Manenti 2020; Souza-Silva et al. 2021). This zonation generates distinct microhabitats for cave fauna, which varies not only in climatic and photic characteristics but also in the types of substrates they present (Moseley 2008; Souza-Silva et al. 2011b; Du Preez et al. 2015; Lunghi et al. 2017; Mammola and Isaia 2017; Mammola 2019; Lunghi and Manenti 2020; Mammola et al. 2020; Souza-Silva et al. 2021). 54 Cave-restricted species (troglobitic species) usually present adaptations to the cave environments, which can be morphological, physiological, reproductive, or even behavioral (Romero & Green, 2005). Furthermore, those species are usually rare and endemic because of their long evolutive history in such stable environments, which makes them sensitive to changes in their ecosystems, such as slight variations of temperature and moisture (Culver & Pipan 2009; Mammola et al. 2019). The troglobitic species are so unique that caves with many cave-restricted species are classified as Hotspots of Subterranean Biodiversity (HSB). According to Culver and Sket (2000), a cave or a cave- system must have 20 or more cave-restricted species to be considered a hotspot. However, Culver et al. (2021) raised the cutoff to 25 species, arguing that the global list was too lengthy. It is important to consider, nevertheless, that 25 species is another arbitrary cutoff (as the originally purposed number of 20 species), that probably results from an analysis preferably considering temperate regions. Thus, it is important to mention that caves located in tropical areas will rarely present up to 25 cave-restricted species. Hence, we herein opted by keeping the original concept of HSB (Culver & Sket, 2000). Considering the original definition of Culver and Sket (2000), there are currently three HSB in Brazil: the Toca do Gonçalo cave (northeastern Brazil) with 22 troglobitic species; the Areias Cave System (southeastern Brazil) with 28 troglobitic species and the Água Clara Cave System (northeastern Brazil), with 30 troglobitic species (Souza-Silva & Ferreira 2016; Souza-Silva et al. 2021). Of those three HSB, only one (Areias Cave System) is protected within the limits of a State Conservation Unit (Parque Estadual Turístico do Alto Ribeira). The two remaining are in unprotected areas and are currently exposed to several anthropogenic threats (Souza-Silva & Ferreira 2016; Souza-Silva et al. 2021). Unfortunately, Brazilian speleological legislation does not protect these important areas. From 1990 onwards, Brazilian caves were fully protected, but in 2008, a decree determined that Brazilian caves should be classified according to their relevance degree, allowing some caves to be destroyed for mineral resources exploitation. Only those caves classified as presenting maximum cultural, geological, and/or biological value should be preserved (Decree nº 6.640). However, a new decree (Decree nº 10.935) from 2022, started to allow the destruction of even those caves with maximum relevance. 55 The Gruta do Padre cave is one of the largest known caves in Brazil, with 16,400 meters of mapped galleries, representing the fifth longest cave in the country. Additionally, it is part of the most extensive subterranean hydrological system in Brazil (Rubbioli et al. 2019). Considering the uniqueness and high biological relevance of this Cave, the main goal of this study was to identify the variables determining the spatial distribution of invertebrates along with presenting the fourth hotspot of subterranean biodiversity in the Neotropical region. Furthermore, as this cave presents areas with highly distinct conditions (upper dry galleries and lower stream conduits), variables describing the physical, trophic, and microclimatic attributes of the cave were used to test three hypotheses: i) upper and lower areas within the cave will present distinct communities regarding the species composition; ii) species richness will be reduced in areas far from the cave entrances; and iii) invertebrates will respond to different habitat components and habitat heterogeneity on the cave floor in the distinct cave compartments (upper and lower areas). In addition, we discuss the impacts over this cave and argue about the importance of preserving this unique new South American hotspot of subterranean biodiversity. Material and Methods Study area The Gruta do Padre cave is located at the Santana municipality (Fig. 1), in southwestern Bahia state, and is inserted in the Bambuí Group, the largest carbonate region in Brazil, with 146,378 Km² of area and approximately 6,302 registered caves. The Santana municipality is placed in a transition zone between the Caatinga and Seasonal Dry Forests and has a high potential for endemic species (Dinerstein et al., 2017). This high potential probably resulted from several paleoclimatic changes in the Brazilian semi-arid caused by the expansions and retreats of Atlantic and Amazonian humid forests (Sobral- Souza; Lima-Ribeiro; Solferini, 2015). The climate in the area is the Aw (Köppen, 1936), with dry winters and rainy summers. Due to the strong tropical rains that occur in the region during the summer, safe access to the cave is only possible in dry periods (March to October). The Padre Cave comprises the bigger portion of the longest subterranean hydrologic system in the country, formed by a long subterranean stretch of the Santo Antônio River (Auler et al., 2019). This 56 river flows through four caves before reaching the Corrente River. Firstly, the Santo Antônio River becomes underground in a small nameless cave, which siphons after the entrance. In a second moment, the river reappears in the Cipó Cave, with approximately 2.76 Km of extension, sinks again, and reappears in the Padre Cave. Into this cave, it flows through 6.2 Km of large conducts until sinks. Finally, the river reappears in the Bananeira Cave, the twenty-fifth longest cave in Brazil, with its 6.55 Km of waterlogged conducts, and then flows into Corrente River (Auler et al., 2019). The Padre Cave has two entrances, located around 1.7 Km far from each other. The upstream entrance is also locally known as Lapa do Cedro Cave or Lapa D’água Cave (Fig. 2–B) and presents rupestrian paintings and archeological engravings indicating its past use by native populations. The downstream entrance, known as Padre Cave or Santo Antônio Cave, was used as a peregrination point for religious in the 20th century. Although both entrances have a narrow area of dry forest protecting them (Fig. 2–A), pastures and monocultures are dominant in the surrounding landscape (Auler et al., 2019). The cave can be dived into three distinct levels. The lower level (Fig. 2–D), where the Santo Antônio river flows, is the longest, with very deep areas and a high ceiling, reaching 40 meters. The second level is located around 45 meters above the river level and comprises giant galleries (Fig. 2–C), some of which present almost 50 meters in width. These galleries are composed of different kinds of substrates, such as downed blocks, sandy areas, and speleothems. The third and last level is smaller and can be found 53 meters above the Santo Antônio river (Auler et al., 2019). Field proceduresSampling design The richness and composition of cave invertebrates, as well as the habitat structure traits, were determined along 53 transects (meso-scale sampling - 10 × 3 m each) distributed on the caves’ floor (Fig. 3), from the entrances to the deeper regions of the cave, encompassing both the lower level (river conduit) and the upper level (upper dry galleries). Quadrats (micro-scale sampling - 1 m2) were inserted in triplicates within the limits of each transect (Fig. 2), totalizing 159 quadrats (Souza-Silva et al. 2021). Invertebrate sampling was done by visual search along the transects and quadrats (Souza-Silva et al. 57 2021). The sampling in the quadrats allowed the detection of low mobility and small-size species, which could then be thoroughly searched in the remaining transect if detected. The sampling was first performed in the quadrats and later in the respective transect, always by three collectors, and was only completed when all the invertebrates had been sampled and/or accounted for. The search time varied among each sampling unit since the different sampling areas presented a considerable structural distinction along the cave. Additionally, to maximize the detection of troglobitic and stygobitic species, direct intuitive search techniques were also applied in other cave areas (Wynne et al. 2019). Invertebrates were preserved in properly labeled jars containing 70% ethanol. In laboratory, the specimens were sorted with a Stemi 508 (ZEISS) stereomicroscope, identified until the lowest possible taxonomic level, and separated into morphotypes (Oliver & Beattie, 1996). Potential troglobitic species were identified by the presence of troglomorphic traits, such as eyes and pigmentation reduction, appendage elongation, among others (Culver & Pipan, 2009). Furthermore, specialists in different taxa were also consulted to assist in the detection of specific troglomorphic traits (specialists are acknowledged further on). The voucher specimens were deposited in the Collection of Subterranean Invertebrates of Lavras (ISLA), linked to the Center of Studies on Subterranean Biology (CEBS) of the Federal University of Lavras (UFLA). Environmental traits in different scales The survey of the habitat structure parameters in the transects was performed according to the methodology used by Souza-Silva et al. (2021). Each transect was subdivided into 10 sections (1 × 3 m) (Fig. 3), in which the surface area occupied by distinct organic and inorganic substrates was visually quantified. Then, a sum was made to obtain the area occupied by each substrate throughout the entire transect. To minimize observer error, all transects were characterized by the same researcher. A digital thermo-hygrometer positioned at the ground at the center of each transect was used to measure the temperature and humidity. The proportion of each substrate in the quadrats was determined through photographs (4000 × 3000 pixels) taken of each quadrat as close as possible to a 90o angle with a Canon Powershot SX60HS camera, at the researcher’s chest high. Photographs were posteriorly analyzed with the aid of ImageJ 1.53K software. Each sector’s position in the cave was plotted on its map and then the 58 distances to the nearest entrance were obtained. The coordinates of each sector in the cave were obtained by the plot in the map with the aid of QGis 3.22 software. Data analysis Pre-analysis routine All the analyses were performed in the R Studio 2022.07.02 Build 576 software. Before running the analysis for invertebrate fauna composition and richness, the correlation between the variables was tested with the aid of the CHART.CORRELATION function from the ‘PerformanceAnalytics’ package. Variables with correlation values higher than 0.70 were excluded from the models. Variables correlated with more than one other variable and with less specificality were favorite to exclusion. Using the functions VIF and VIF.CCA from the ‘Car’ package, the multicollinearity of variables was tested, and those presenting values higher than 10 were excluded. Firstly the variable with highest value was excluded, then the test was repeated, and the process continue until no higher than 10 value was found. A Shapiro-Wilk test was executed, to verify the normality of the data, using the SHAPIRO.TEST function from the package ‘Stats’. A Mantel test was performed to try the spatial autocorrelation between the samples, in both meso and microscales. Mesoscale abiotic features All the substrates in each sector were evaluated and classified into the following classes: vegetal debris - DTV (< 10mm); fine branch - GALF (11 - 30 mm); medium branch - GALM (31 - 50 mm); coarse branch - GALG (65 - 250 mm); trunk - TRO (> 250mm); water pond - WP; drip water - DP; coleoptiles - COL; smooth rock - RL; concrete floor - RC; large rock - XB (1000 - 4000mm); medium rock - MB (500 - 1000mm); small rock - SB (250 - 500mm); cobbles - CB (64 - 250mm); coarse gravel - CAG (16 - 64mm); fine gravel - CAF (2 - 16mm); sand - ARE (0.06 - 2mm); silt - SEF (≤ 0.05 mm); hardpan - HP; speleothems - ES; calcite rafts - JNS; concreted calcite raft - JNC; stalagmite - EST; micro travertine - M