4. DISCUSSION
The conversion of caves into touristic attractions has major impacts on subterranean ecosystems (Fernandez-Cortes et al., 2011; Mulec, 2014; Pulido-Bosh et al., 1997), and causes large-scale indirect anthropogenic disturbances, such as land use change (Jiménez-Sánchez et al., 2008) and climate change (Mammola et al., 2019). Yet, studies show inconsistent responses of the microbial communities to human disturbance, pointing out either a positive (Mammola et al., 2017; Marques et al., 2016), negative (Alonso et al., 2019; Shapiro & Pringle, 2010) or variable effect (Bercea et al., 2019; Mulec & Oarga-Mulec, 2012). These controversial outcomes are likely attributed to differences in the examined groups and to the environmental variations across different caves (Pfendler et al., 2018, 2019). Examining human-induced changes in subterranean microbial communities across different caves is often hampered by the great differences in local species composition, which is strictly influenced by geographic and geological features such as cave size, morphology, and water dynamics (Saiz-Jimenez, 2012). To overcome these limitations, we simultaneously analysed the response of the three main microbial groups inhabiting caves, namely Fungi, Bacteria and Archaea, combining the data collected across multiple show caves. By relating both species richness and community composition in the tourist parts to those observed in control areas, we could depict the human-induced effects on microbial communities, without confounding factors due to local environmental features. By adopting a metacommunity framework, we disentangled the effect of visitors on subterranean microbial communities from the habitat filter and the dispersal limitation for the first time.
Our results pointed out an effect of visitors on the diversity and composition of microbial communities in show cave sediments, but with different outcomes depending on the examined microbial group. When separately analysing the response of the three groups, our results pointed out that human disturbance does not affect the species richness of the prokaryotic component, as demonstrated by the absence of a significant response of the models fitted on the ASV-ratios. However, we clearly showed that visitors influence the composition of microbial communities of Bacteria and Archaea by causing a turnover of species as demonstrated by the decline of β-replacement at increasing distance from the tourist path. These results are in accordance with the outcomes of a meta-analysis testing the response of microbial communities to disturbance, which showed only a weak response in terms of species richness but the effects were evident on the community composition (Jurburg et al., 2021). The observed species turnover on the tourist path could be ascribed to the replacement of resident species by propagules of microorganisms vehiculated by visitors’ shoes and clothes, which are spread around while visiting the cave (Mulec, 2014; Saiz-Jimenez, 2012; Zhelyazkova et al., 2020), similarly to what observed by other authors (Alonso et al. 2019; Dong et al., 2020; Ikner et al. 2007).
The response of Fungi showed a U-shaped response to the human disturbance. The ASV-ratio in samples close to the path was similar to those found in control sites, but decreased significantly at intermediate distance. This outcome suggests that visitors increase fungal species richness either directly —by vehiculating propagules of novel species (Jurado et al., 2021; Mammola et al., 2017; Taylor et al., 2013)— or indirectly —by introducing high amounts of organic matter, beyond the lampenflora biomass proliferating in the visited paths due to the lighting system, that favour more competitive species (Jurado et al., 2010; Marquez et al., 2016; Addesso et al., 2020). However, in sampling sites located deeper in the cave, the pattern is reversed, with increasing ASV-ratio at increasing distance from the tourist path. We can hypothesize that fungal communities at the cave entrance are composed of different species, which can cope well with the new conditions induced by visitors with respect to the assemblages in the deepest parts. On the contrary, fungal communities in the control area are composed of resident species that are expected to be less resilient to human disturbance.
Variation of fungal communities at different cave depths is supported by the observed decline in species replacement at increasing distance from the cave entrance. Based on these results, the composition of species assemblages of Fungi is driven by dispersal, as demonstrated by their higher turnover at the entrance of the cave with respect to the deepest parts. This is in agreement with previous studies demonstrating that Fungi from the external environment can be vehiculated in caves by air circulation (Docampo et al., 2011; Ogórek et al., 2014, 2016).
Environmental filtering plays a significant role in determining the assemblages of Fungi and, to a lesser extent, of Archaea. Both groups showed a decline of their ASV-ratio at increasing dissimilarity of sediment composition from the control areas, together with an increase in species replacement for Fungi. Given that the distance from the tourist path and from the cave entrance did not influence sediment composition, this response can be specifically ascribed to the filter exerted by the environment. By examining the contribution of each physical and chemical parameter to the overall variation of sediment composition, we could identify Nitrogen concentration and, to a lesser extent, pH and the % of clay as the most important variables. We can therefore hypothesize that these parameters filter species based on their ecological requirements. This is in agreement with literature data, which pointed out a key role of substrate as a driver of species richness and composition of cave assemblages of both Fungi (Cailhol et al., 2020; Kukla et al., 2018) and Archaea (Marques et al., 2017).
Overall, by adopting a metacommunity framework, we could provide new perspectives on the dynamics and patterns of microbial communities under human disturbance. Specifically, our results highlighted that anthropogenic pressure affects all microbial communities but with different effects. Similarly, the three examined groups show differential responses to environmental filtering and dispersal pointing out that a proper understanding of the underlying selective mechanisms require a comprehensive and multi-taxonomic approach.
Funding: this work was realized within the framework of the PRIN SHOWCAVE “A multidisciplinary research project to study, classify and mitigate the environmental impact in tourist caves” - code 2017HTXT2R (PI: Marco Isaia), funded by the Italian Ministry of Education, University and Research. The SHOWCAVE project is endorsed by AGTI (Associazione Grotte Turistiche Italiane) and the SSI (Società Speleologica Italiana). The grant of EP is co-financed by the PON “Research and Innovation” Programme (Axis IV “Education and Research for recovery” – Action IV.6 “Research contracts on Green themes”).
Conflicts of interest: the authors declare no conflicts of interest.
Acknowledgments: we are grateful to Benedetta Baroni for her help during the field work. We also thank Cooperativa Alto Corsaglia (Grotta di Bossea), Mondolè touristic office (Grotta del Caudano), Vittorio Verole and Mario Verole (Grotta del Vento), and to the president, Dr. Francescantonio D’Orilia, and the scientific director, Prof. Mariana Amato, of MIdA Foundation (Grotte di Pertosa-Auletta) for endorsing our research and for their logistic support during the sampling activities.