The species richness obtained in this study (23 spp) represents 18.5% of the estimated fauna of Scarabaeinae in the state of Chiapas (Sánchez-Hernández and Gómez 2018; Sánchez-Hernández et al. 2019). Although Scarabaeinae assemblages exhibit well-defined seasonal patterns (Arellano et al. 2008; Rodríguez-López et al. 2019), the results of this study indicate that a high sampling efficiency was obtained in each of the habitats and the richness recorded is comparable with the fauna reported with other works carried out in this same region, even when in them sampling was carried out during longer periods, different types of attractants were used or both seasonal periods were included (Arellano et al. 2008; Rodríguez-López et al. 2019; 2021; Sánchez-Hernández et al. 2021).
The study landscape is characterized by a fragmented structure caused by the action of land use change, formed by mosaics of agricultural and livestock use, secondary vegetation, and small remnants of primary deciduous forest vegetation (Pérez-Farrera and Espinoza 2010; Rocha-Loredo et al. 2010). The heterogeneous nature of the landscape allowed maintaining dung beetle assemblages mostly of generalist habits, and at the same time, gave room for some species that maintain abundant populations in other habitat types to become established (Rodríguez-López et al. 2019; Sánchez-Hernández et al. 2021); moreover, the broad dominance of a few species (five species accounted for > 86% of total abundance) is typical of fragmented landscapes where structural diversity of forest cover is poor compared to much denser forested areas (Navarrete and Halffter 2008; Sánchez-Hernández et al. 2018; 2022).
Our results show that the diversity and composition of dung beetle assemblages changed according to the characteristics of each habitat. As expected, high values of species diversity (qD) were found in the forest fragments, despite the degree of isolation or the limited extent to which they are subjected, so that these forest remnants can maintain populations of widely distributed species throughout the Central Depression of Chiapas. Despite the constant forest conversion and their isolation, these fragments possess exclusive species of Scarabaeinae, so the habitat loss and the disappearance of some species found in them may cause the alteration of the structure and composition of dung beetle assemblages in the landscape (Nichols et al. 2007; Rodríguez-García et al. 2021; Correa-Cuadros et al. 2022; López-Bedoya et al. 2022). This points to the importance of forest remnants in maintaining regional species populations, ensuring local diversity of dung beetle assemblages.
The NMDS and ANOSIM analyses showed the formation of two dung beetle assemblages, one of them formed by the species associated with the live fence and the forest, with similar dominant species and a high percentage of shared species (more than 80%), very different from what was observed in the cattle pasture. Live fences, despite their limited extent, are considered to be wooded elements that promote spatial heterogeneity within agricultural landscapes, thus representing an important component of forest cover by reducing the effects of isolation between forest patches (Harvey et al. 2003). Several studies have shown that live fences stand out as habitats of high diversity of different taxa in fragmented areas, such as birds (Lang et al. 2003; Ramírez-Albores 2010), bats (Estrada and Coates-Estrada 2001) and diurnal butterflies (Tobar and Ibrahim 2010).
Although live fences do not host a diversity similar to forest patches, the data obtained are in agreement with other studies suggesting that this type of habitats function as biological corridors for Scarabaeinae species, allowing them to move between forest remnants and, thus, reducing competition for resource availability mainly for those species that avoid open areas (Estrada et al. 1998; Díaz et al. 2010; Rangel-Acosta and Martínez-Hernández 2017). In this sense, live fences can help species with restricted movement capacity and little dispersal to successfully occupy other habitat types, such as areas of secondary vegetation (Sánchez-Hernández et al. 2018), patches of remnant vegetation (Rodríguez-López et al. 2019) and areas of primary vegetation (Sánchez-de-Jesús et al. 2016). This was observed with the species identified as forest and live fence indicators which, although they may be generalist in habit, rarely invade open areas such as pastures or livestock paddocks.
On the other hand, some species manage to maintain abundant populations as a result of their generalist capacity that allows them to invade different types of habitats in the landscape, regardless of the degree of disturbance they present (Estrada et al. 1998). Such is the case of D. amplicollis, which was the most abundant and dominant species in the three habitats. This species is considered endemic to Mexico but is widely distributed in the tropical forests of the country, mainly in the deciduous forests of the Mexican Pacific slope (Chamé-Vázquez et al. 2020). In this region of Chiapas, D. amplicollis has been found as a dominant species, both in areas of primary vegetation and in open or highly disturbed areas (Arellano et al. 2008; Navarrete and Halffter 2008; Arellano et al. 2013; Sánchez-de-Jesús et al. 2016; Sánchez-Hernández et al. 2018; 2022). In general terms, Dichotomius species are considered by Davis (1996) as fast burrowers, a particularly important characteristic in fragmented areas, since it allows the manure to be recycled quickly and not be exposed to the environment for prolonged periods, thus slowing the proliferation of pests, as well as allowing the soil to be oxygenated to a greater extent due to the body size of this group (Gregory et al. 2015).
Species composition in the cattle pasture indicates a clear difference in the beetle assemblages that prefer these habitats versus the assemblages that avoid open areas. Although the pasture areas presented the lowest species richness, they had high values of abundance, due to the predominance of heliophilous species (i.e., D. gazella, P. balthasari, and P. demon), which are favored by the continuity of large extensions of open areas, with the presence of greater volumes of cattle manure, which allows them to maintain much larger populations (Montes de Oca & Halffter 1998; Arellano et al. 2008; Edmonds and Zídek 2012; Arellano et al. 2013).
Among the species associated with cattle pasture, D. gazella stands out, a specialist and exotic coprophagous species, which is favored by anthropogenic activities (mainly cattle ranching), which has allowed it to rapidly establish in pastures in tropical and subtropical regions of Mexico and Central America, but without penetrating forest areas (Montes de Oca & Halffter 1998; Pablo-Cea et al. 2017; Noriega et al. 2020). Also noteworthy is the presence of two species of the genus Phanaeus in the cattle pasture, whose occurrence showed a high degree of fidelity to this habitat. These are two large borrowing species (P. demon and P. balthasari) that provide a valuable service in the removal of large masses of fecal matter in the grasslands. These beetle species (large paracoprids) can remove a greater volume of dung, as well as contribute more effectively to the biological control of flies and helminths of veterinary importance (Bang et al. 2005; Shahabuddin et al. 2008; Gregory et al. 2015).
The data obtained in this work concur that the presence of arboreal habitats, such as live fences, can be important elements for wildlife within fragmented landscapes, thus compensating for the loss of forest cover, since they help maintain the physical and functional connectivity of the landscape, acting as connectors between isolated patches of vegetation and as biological corridors for certain species to move through the agricultural matrix (Harvey et al. 2005). However, more information is needed to assess the role of different types of live fences (e.g., simple vs multi-strata) as quality habitats for the persistence of Scarabaeinae assemblages over the long term and at different scales, as the structure and configuration of the tree composition of these habitats be factors that also contribute to the diversity of species they can support (Garbach et al. 2010; Tobar and Ibrahim 2010).