Dung beetles are among the most cost-effective of all animal taxa for assessing biodiversity patterns, but relatively little is known about the dung beetle fauna of Suriname. I sampled dung beetles using baited pitfall traps and flight intercept traps in the Kwamalasamutu Region of southern Suriname. I collected 4,554 individuals represented by 94 species. Species composition and abundance varied quite strongly among sites. Dung beetle diversity correlated positively with large mammal species richness, and was highest at the most isolated site (Kutari), suggesting a possible cascading influence of hunting on dung beetles. Small-scale habitat disturbance also caused local dung beetle extinctions.
The dung beetle fauna of the Kwamala region is very rich relative to other lowland forests of Suriname and the Guianas, and contains a mix of range restricted endemics, Guiana Shield endemics, and Amazonian species. I estimate that about 10–15% of the dung beetle species collected here are undescribed. While most species were coprophagous, 26 species were never attracted to dung; 4 of these were attracted exclusively to carrion or dead invertebrates and the other 22 were only captured in flight intercept traps. The abundance of several large-bodied dung beetle species in the region is indicative of the intact wilderness that remains. These species support healthy ecosystems through seed dispersal, parasite regulation and other processes. Maintaining continuous primary forest and regulating hunting (such as through hunting-restricted reserves) in the region will be essential for conserving dung beetle communities and the ecological processes they sustain.
Dung beetles (Coleoptera: Scarabaeidae: Scarabaeinae) are an ecologically important group of insects. By burying dung as a food and nesting resource, dung beetles contribute to several ecological processes and ecosystem services that include: reduction of parasite infections of mammals, including people; secondary dispersal of seeds and increased plant recruitment; recycling of nutrients into the soil; and decomposition of dung as well as carrion, fruit and fungus (Nichols et al. 2008). Dung beetles are among the most cost-effective of all animal taxa for assessing and monitoring biodiversity (Gardner et al. 2008a), and consequently are frequently used as a model group for understanding general biodiversity trends (Spector 2006). Dung beetles show high habitat specificity and respond rapidly to environmental change. Since dung beetles primarily depend on dung from large mammals, they are excellent indicators of mammal biomass and hunting intensity. Dung beetle community structure and abundance can be rapidly measured using standardized transects of baited traps, facilitating quantitative comparisons among sites and studies (Larsen and Forsyth 2005).
I sampled dung beetles at all three sites (Kutari, Sipaliwini, and Werehpai) using standardized pitfall trap transects. Ten traps baited with human dung were placed 150 m apart along a linear transect at each site (see Larsen and Forsyth 2005 for more details). Traps consisted of 16 oz plastic cups buried in the ground and filled with water with a small amount of liquid detergent. A bait wrapped in nylon tulle was suspended above the cup from a stick and covered with a large leaf. At each site, traps were collected every 24 hours for four days, and were re-baited after two days. I set three flight intercept traps at each site to passively collect dung beetle species that are not attracted to dung. I also placed additional pitfall traps whenever possible with other types of baits that included rotting fungus, carrion, dead millipedes, and injured millipedes. All traps were collected daily. I opportunistically collected dung beetles that I encountered in the forest, usually perching on leaves during both day and night.
From August 19–24, 2010, I collected dung beetles at the Kutari site (N 02° 10′ 31″, W 056° 47′ 14″) in primary forest characterized by small hills and several swampy areas. From August 27 – September 4, 2010, I collected dung beetles at the Sipaliwini site (N 02° 17′ 24″, W 056° 36′ 26″) in primary forest with small hills and relatively dry, hard soils with high bedrock. From September 2–7, 2010, I collected dung beetles at the Werehpai site (N 02° 21′ 47″, W 056° 41′ 52″) in primary forest as well as in bamboo (1 dung trap) and secondary forest (1 dung trap). Beetles were identified and counted as they were collected in the field and voucher specimens were stored in ethanol for further study and museum collections. Beetle specimens are deposited at the National Museum of Natural History at the Smithsonian Institution in Washington, DC, USA and at the National Zoological Collection of Suriname in Paramaribo.
To estimate total species richness at each site and assess sampling completeness, I compared the observed number of species with the expected number of species on the basis of randomized species accumulation curves computed in EstimateS (version 7, R. K. Colwell, http://purl.oclc.org/estimates) (Colwell and Coddington 1994). I used an abundance-based coverage estimator (ACE) because it accounts for species abundance as well as incidence, providing more detailed estimates. I also used EstimateS to calculate similarity among sites, using the Morisita-Horn similarity index which incorporates species abundance as well as incidence.
RESULTS AND DISCUSSION
I sampled a total of 94 species and 4,554 individuals of dung beetles during the RAP (Table 1, Appendix A). Species richness was similar at all sites. Among dung traps, for which sampling effort was identical at all sites, species richness was highest at Sipaliwini (49 species), followed by Werehpai (47 species) and Kutari (44 species) (Table 1, Fig. 1). Species accumulation curves for dung-baited pitfall traps (based on abundance-based coverage estimator) indicated that I sampled an estimated 88% of all coprophagous species occurring in the area. However, sampling completeness was lowest at Werehpai where I sampled only 72% of the dung-feeding species likely to occur at the site (Table 1, Fig. 1). Consequently, species richness estimators predict that Werehpai supports the highest number of coprophagous species (65 species), followed by Kutari (60 species) and Sipaliwini (57 species) (Table 1).
These differences between observed and predicted species richness are probably explained by strong differences in abundance among sites. As with observed species richness, abundance was highest at Sipaliwini and lowest at Kutari; Sipaliwini supported almost three times as many individuals as Kutari (Table 1, Fig. 1). Low abundance at Kutari may have been influenced by the large areas of swamp and flooded forest at the site, conditions which negatively affect many dung beetle species whose larvae develop in the soil. However, diversity, measured by the Shannon index, showed the opposite pattern to observed species richness. Diversity was highest at Kutari and lowest at Sipaliwini, due to greater evenness of species' abundance distributions at Kutari (Table 1).
Out of 94 species sampled during this RAP survey, only 68 were attracted to dung. Considering all trap types and capture methods, Kutari supported the greatest number of dung beetle species (70 species) (Appendix A, B). Four species were attracted only to carrion or to dead invertebrates (Appendix B). 22 species were sampled only in flight intercept traps (Appendix B), and many of these species are poorly represented in collections because they are difficult to sample and in some cases, their diet is unknown. Some of these species show unusual specializations, such as millipede predation or colonization of leaf-cutter ant nests (see interesting species discussion below).
Species composition and community structure varied strongly among sites (Table 2). Sipaliwini and Werehpai were relatively similar in terms of community structure, showing a high Morisita-Horn index. Kutari was very distinct from both Sipaliwini and Werehpai, and contained many species not present at the other sites. Some of the most abundant species at a particular site were rare or completely absent from other sites (Appendix A). For example, I caught 211 individuals of Ateuchus simplex at Sipaliwini, and none at Kutari, despite the relative close proximity of both sites.
Dung beetle species richness was strongly reduced by habitat disturbance. Second growth forest supported only 70% of the total species richness found in primary forest, while bamboo supported only 40% of primary forest species richness (Fig. 2). Only one species, Uroxys gorgon, occurred in bamboo or secondary forest but did not occur in primary forest. Uroxys gorgon is known to be phoretic in sloth fur, and sloths are often hyper-abundant in secondary forest. The absence of other disturbance-adapted species in the Kwamala region was somewhat surprising, given the high number of ‘weedy' species found in other parts of South America. Their absence might be explained by the extraordinarily low proportion of disturbed habitats occurring in southern Suriname.
Dung beetle diversity (measured by the Shannon index) was strongly positively correlated with species richness of large mammals (Fig. 3), with the highest beetle diversity and mammal richness occurring at Kutari. Kutari also appeared to support the most primate species of all sites (see Large Mammals Chapter), and primates provide one of the most important food sources for dung beetles. High dung beetle diversity at Kutari may have been influenced by higher mammal richness and by lower hunting intensity, although further data are needed. On the other hand, dung beetle species richness and abundance were not correlated with the large mammal community, although no robust analysis was possible due to the short sampling period for mammals and the small number of sites for both groups (N=3). Furthermore, dung beetle abundance and species richness may have been influenced by differences in habitat and soil conditions, as discussed above.
At least 23 dung beetle species sampled during this RAP survey are known to be distributed across the Amazon basin. Many of the Amazonian species were locally rare and sampled at Kutari (Appendix A), which was the southernmost site sampled during the RAP. Out of these 23 Amazonian species, 20 occurred at Kutari, and only 16 at Werehpai and 15 at Sipaliwini. The Kwamala area may straddle the northern range limit for these species.
For the few genera that have been revised and for which good distributional data exist, many of the remaining species are restricted to the northern Amazon region, the Guiana Shield, or show an even more restricted range, while several are data deficient (see also interesting species discussion below). For example, Coprophanaeus parvulus, Oxysternon festivum, and Eurysternus balachowskyi are endemic to the Guiana Shield and northern Amazon, while Oxysternon durantoni and Eurysternus cambeforti occur only in the extreme northeastern Guianas (Edmonds and Zidek 2004, Genier 2009, Edmonds and Zidek 2010).
Dung beetle species richness is high in the Kwamala region relative to other areas in northeastern South America and the Guianas (Table 3). Similar RAP surveys at Lely and Nassau in Suriname yielded only 35–48% of the species richness found around Kwamala. Other studies from Venezuela, French Guiana, and Brazil also showed lower species richness in lowland primary forest with comparable sampling effort. Further sampling around Kwamala may yield as many or more species than were found in French Guiana and at Jari, Brazil, where greater sampling effort was employed (Table 3).
I estimate that about 10–15% of the dung beetle species collected during this RAP (10 to 14 species) are undescribed. However, most of the genera collected here have never been revised, and determination of these undescribed species will require further comparisons with other museum collections. I sampled 26 species of Canthidium in the Kwamala area. Canthidium is a hyper-diverse yet very poorly known genus, and many of these species are almost certainly new to science. Ateuchus is also a poorly known yet diverse genus, and several Ateuchus species from the RAP are likely to be new. Canthon sp. 2 represents an undescribed species that is currently under study (see Appendix A).
Several large-bodied dung beetle species, such as Coprophanaeus lancifer (the largest Neotropical dung beetle species), Oxysternon festivum, and Dichotomius boreus, were sampled at all three sites. These species move long distances and require large, continuous areas of forest to persist. Their presence at the sites is indicative of the intact, contiguous landscape around Kwamala. These large dung beetle species are also the most ecologically important for burying seeds and controlling parasites.
Six species (Dendropaemon sp. 1, Deltorhinum guyanensis, and four Anomiopus species) were only sampled in flight intercept traps and their distinctive morphology, with strongly reduced tarsi and stout, compact bodies, suggest that they are myrmecophilous (associated with ant nests), as are several other dung beetle species. Based on a recent revision of the genus Anomiopus, this is the first record for all four of these species in Suriname (Canhedo 2006), although I collected A. parallelus and A. lacordairei on another RAP survey in Suriname (Larsen 2007; Appendix A). Both species were previously known only from French Guiana and northern Brazil. Deltorhinum guyanensis, endemic to the Guianas, was only described after this RAP survey was conducted (Genier 2010), and this is the first record of this species in Suriname.
Deltochilum valgum is a highly specialized predator of millipedes, and adults decapitate and feed on millipedes that are much larger than themselves. This unusual behavior was only discovered and described last year (Larsen et al. 2009). Canthidium cf. chrysis is a member of the escalerei species group which commonly feed on dead invertebrates. It was captured mostly with dead millipedes, but occasionally with carrion, and may be specialized to feed on millipedes. Canthidium sp. 20 (aff. chrysis), Canthon sp. 1 and Canthon sp. 2 were also most abundant at dead millipedes, but whether they specialize on millipedes or on dead invertebrates in general is not yet clear. Canthidium cf. gigas, which was represented by only one individual in a flight intercept trap, is a member of an unusual species group which may feed on fungus. This group includes by far the largest of all Canthidium species. Canthidium cf. minimum is an unusual species that may need to be transferred to a different genus (see Appendix A for this and other taxonomic notes).
The Kwamala area supports vast tracts of intact primary forest, which is important for many dung beetle species. Consequently, I found extremely high species richness of dung beetles in the area (94 species). To put this diversity into perspective, during a RAP survey at the Nassau and Lely plateaus in Suriname, I sampled only 24 species and 33 species at each site respectively (Table 3). I sampled extensively in lowland forest around Lago Guri in Bolivar, Venezuela, and found only 41 species (Larsen et al. 2008). On the other hand, small-scale habitat loss and disturbance around Kwamala led to local dung beetle extinctions, which would likely be exacerbated by more widespread habitat loss. Preventing mining operations and other drivers of deforestation from entering the area will be important for maintaining the high biodiversity of the Kwamala region.
In addition to high overall species richness, I found high Beta diversity at the sites across very small spatial scales, and Kutari supports a very distinct dung beetle community than the other sites. Consequently, it is important to protect the diversity of soils and habitats that occur in the Kwamala region even at small spatial scales. Plans for protected areas or reserves should incorporate this small-scale spatial heterogeneity.
Tropical ectotherms, such as dung beetles, are among the most sensitive organisms on Earth to climate change (Larsen et al. 2011). Climate warming is forcing many species to shift their distribution poleward or upslope, and these effects are strongest at the edge of species' ranges. Since the Kwamala area contains many Amazonian species near the edge of their range limit, it may present an excellent opportunity to monitor the response of populations and species' distributions to climate change.
High dung beetle diversity at Kutari, the most isolated site, was correlated with high mammal, including primate, species richness, and this may be explained by lower hunting pressures. The abundance and biomass of dung beetles in the Kwamala area overall was relatively high, and was higher than I observed at Nassau and Lely in other parts of Suriname. This suggests that in addition to the pristine state of the forest, populations of large birds and mammals are relatively stable. However, dung beetle abundance was lower than I expected based on surveys in other Neotropical primary forests where no hunting occurs. This is likely to reflect the relatively low abundance of spider monkeys, howler monkeys, and white-lipped peccaries, which are among the most important species for dung beetles but are also preferred for bushmeat. Reduced hunting on these key species would help to stabilize ecosystem dynamics not just for dung beetles, but for seed dispersal and other ecological processes as well. The establishment of hunting-restricted reserves such as the one at Iwana Samu is an excellent way to maintain sustainable populations of large mammals.
I would like to thank the Trio People of Kwamalasamutu, the game wardens, and the park guards, as well as Leeanne Alonso and Brian O'Shea for coordinating the RAP survey. Dana Price and Francois Feer provided valuable comparative data from French Guiana.
GenierF. 2009Le Genre Eurysternus Dalman, 1824 (Scarabaeidae: Scarabaeinae: Oniticellini) PensoftBulgariaGoogle Scholar
LarsenT. H. 2007Dung beetles of the Lely and Nassau plateaus, Eastern Suriname.99101 in L. E.Alonso J. H.Mol A rapid biological assessment of the Lely and Nassau plateaus, Suriname (with additional information on the Brownsberg Plateau).Conservation InternationalArlington, VA, USAGoogle Scholar
LarsenT. H. F.Escobar I.Armbrecht 2011Insects of the Tropical Andes: diversity patterns, processes and global change.228244 in S. K.Herzog R.Martinez P. M.Jorgensen H.Tiessen Climate Change and Biodiversity in the Tropical Andes.Inter- American Institute of Global Change Research (IAI) and Scientific Committee on Problems of the Environment (SCOPE)São José dos Campos and ParisGoogle Scholar