Dung-beetles (Coleoptera: Scarabeidae) from the Zona Protectora Las Tablas, Talamanca, Costa Rica

Dung-beetle species are considered an important focal indicator group in tropical forests. During 2007, eighteen traps were set in two permanent biodiversity plots during one week surveys in Las Tablas Protected Zone within La Amistad Biosphere Reserve. The main purpose of this study was to determine the group composition and diversity as the basis for permanent monitoring and to measure potential impacts of land use change and climate change on mountain diversity. Pitfall traps were placed randomly within the plots with human feces as bait. A total of 26 species distributed in 13 genera and six tribes were collected. The composition of species includes eight endemic species, seven for Costa Rica (CR) and Panama and one only for CR. The diversity found in this study was typical of previous studies in the region. Further sampling is needed to obtain the total number of species for the area. This study is part of an on-going research project about climate change impacts and biodiversity monitoring in this important ecoregion of Mesoamerica.


Introduction
Dung-beetles are important indicator species for monitoring spatial and temporal patterns of biodiversity of tropical forests (Favila andHallfter 1997, Davis et al. 2001), where they provide important ecological benefits including being secondary dispersers, accelerating nutrient recycling rates, and increasing plant yields. They have been used extensively in biodiversity monitoring and other studies across the tropics (Mittal 1993, Andresen 1999, Perez-Ramos et al. 2007).
Dung-beetles' ecology and taxonomy has been extensively studied in Costa Rica, with numerous studies across the country and with an important collection held by the National Biodiversity Institute (Kohlmann et al. 2007). However, the species composition in mountain forests of the country is one of the least studied, including the Talamanca highlands (Kohlmann et al. 2007). Taxonomically, this group belongs to the Order Coleoptera, Family Scarabaeidae, it comprises about 5000 species worldwide including 12 tribes (Cambefort 1991). In Costa Rica 175 species have been reported, a high number compared with other biologically rich countries in the region (Kohlmann et al. 2007). Here we present one of the first species lists for dung-beetles of the Talamanca Eco-Region, focusing on Las Tablas Protected Zone (ZPLT).

Study Area
The Cordillera de Talamanca (Figure 1) is an eco-region located on southeastern Costa Rica, and western Panama covering a total of 667,825 ha (INBio 2007). The region is considered one of the most important protected corridors of forest in Mesoamerica and the most endemic species of eco-regions in Costa Rica (Gonzalez-Maya et al. 2008a). The eco-region is composed of protected areas, indigenous reserves, private reserves and private farms, and it includes the largest Protected Area in Costa Rica, La Amistad International Park, shared between Costa Rica and Panama. The area has been designated a Biosphere Reserve and World Heritage Site by UNESCO.
Las Tablas is a Protected Zone in southern Costa Rica covering approximately 19,062 ha. Protected Zones are passively managed areas with restrictions on land use, but are privately owned and maintained including many farms. The zone represents an important conservation area since it is part of the Buffer Zone of La Amistad International Park. It also represents the most important forest patch in a proposed corridor to link the Talamanca Mountain Range with the Osa Peninsula, composed of three important protected areas for the country (Cespedes et al. 2007). The present study was conducted in Las Alturas del Bosque Verde, a 11,600 ha private farm with an elevation range from 1000 to 2100 masl (Gonzalez-Maya et al. 2008b). The mean annual precipitation is around 3500 mm with a mean temperature of 27 °C (INBio 2007a). All permits were granted under "general collect" for INBio.

Dung-Beetle sampling
During 2007, two permanent sampling plots for biodiversity surveys were established on an elevation gradient from 800 to 1300 m. The plots were surveyed during two months (September-October) on each sampling plot. Eighteen baited pit-fall traps (Larsen and Forsyth 2005) were set and maintained regularly during the survey period. The length of the study period was determined by the amount of time needed to obtain the alpha diversity (Spector and Forsyth 1998). The traps were baited with human feces from the same individual (Howden and Nealis 1975), which were renewed every day for continued efficacy. The traps (plastic cups) were filled with water and detergent to maintain the collected individuals in the trap.
The sample from each trap was collected and stored separately in 70% alcohol, dried and deposited in INBio collection. The identification was made by comparing with material in the INBio collection, and all the specimens were deposited there and were catalogued for future reference. All the identification was undertaken by one of the authors with the support of INBio´s curator (Angel Solis).
Basic diversity analyses were made using McIntosh Dominance Index, Shannon Diversity Index, and Jaccard Similarity Index.

Results and Discussion
A total of 26 species, representing 38% of the total number of species reported for this elevation range in the country (Kohlmann et al. 2007), and 566 specimens were collected during the entire study. All the specimens belonged to the Scarabeinae sub-family and a total of six tribes and 13 genera (Table1).
Of the total of 26 species, 8 were endemic (30.76 %), 7 shared between Costa Rica and Panama and one endemic to Costa Rica. This number represents the high endemism value of the area, and the importance for conservation and further research.
Given the level of sampling effort, the number of species as well as the total of 566 specimens is relatively high and comparable with other studies in the region conducted at lower elevations (Suatunce et al. 2004, Harvey et al. 2006).

----------------
The representativeness of the species collected was also high, according to Atta Database (INBio database including all the specimens collected and stored at the Institute´s collection; INBio 2008) which showed that the list presented here includes the 55.5 % of the previous reported species for the area and also adds 7 species (20.5 % of the total species for the area) not reported before (Table 2).
Among the species previously reported for the area and that are not found in this study, the richest genus was Onthophagus sp. (7 spp.), followed by Canthidium sp. (3 spp.), Deltochilum sp. and Coprophanaeus sp.
(1 sp. each). The absence of these species could have a number of causes, including the type of bait, as some species use different types of resources and not only dung, and the influence of wind, trap spacing, elevation among many other factors that influence the sampling and species composition obatined (Larsen and Forsyth 2005). Also, dung beetles have shown a strong relationship with seasonality and vegetation type, where rainfall regimes and forest structure can determine the composition of species Arellano 2002, Andresen 2005).
The newly reported species for the area include 4 genera; there are several differences among these species, varying from really common and widespread species according to literature (C. haroldi and C. centrali) to almost ecologically unknown species (D. pseudodidymus, O. inediapterus and O. nyctopus). The specimens also include one highly seasonal species (D. satanas). The most interesting case (O. orphnoides) is a species that is reported to feed on seeds from some species of Lauraceae (Persea sp.) and is not thought to feed on dung at all (Solis pers. comm.).
The distribution of abundances in the two plots showed a remarkable dominance of Dichotomius satanas (148 individuals), followed by Scatimus erinnyos (96 ind.), Ontherus pseudodidymus (78 ind.), and Corpophanaeus chiriquensis (77 ind.). There was a lower richness in the lower plot (800 m), and higher abundances in the higher one (1300 m; Figure 2). However, the McIntosh Dominance Index did not show significant differences between the two plots (D= 0.639 and 0.649 respectively). Shannon diversity indexes showed small differences among the two sites (H´=2.23 for first plot and H´=2.17 for the second plot), and the Jaccard similarity index showed a 73.1 % similarity between the two plots at different elevations.

Conclusions
Further sampling effort is needed in the area to obtain a comprehensive pool of potential species for the area, and to understand the elevation distribution of species. However, this first list represents an important advance in the knowledge of this important group of indicator species in the region. The value of this report resides in its importance for conservation planning and as an important input to research and conservation efforts in the area. It is also one of the first efforts to catalog and measure high biodiversity of the area, and it builds the foundation for long-term monitoring in middle elevations of this ecoregion.

Acknowledgments
This study is part of a long-term monitoring project concerning potential effects of climate change and forest monitoring in Talamanca held by Proyecto de Conservación de Aguas y Tierras, ProCAT Internacional. Special thanks to Angel Solis for his help, guidance and support to the identification of all the specimens. We thank Soledad Trujillo, Elizabeth Navarro, Gustavo Hernández and all the personnel from Cotón for their support and help in the field and the entire project. Also, we would like to thank specially to Addison Fisher and Fernando Castañeda for their support and work on conservation of the Talamanca region. Sarah Wyatt for their review and support and reviewers for their comments.