Ecological Archives E094-242-A2

Riikka Kaartinen, Bess Hardwick, Tomas Roslin. 2013. Using citizen scientists to measure an ecosystem service nationwide. Ecology 94:2645–2652.

Appendix B. Additional analyses of community structure vs. functional rates.

Several factors have been proposed to affect the rates of ecosystem services. These factors include species richness (Larsen et al. 2005, Slade et al. 2011), number of “functional groups” (Diaz et al. 2003, Larsen et al. 2005, Slade et al. 2007), biomass (O'Hea et al. 2010), and abundance (Larsen et al. 2005) of organisms contributing to the service in question. Different aspects of diversity may add a further dimension, as a community characterized by multiple species but dominated by a single one might perform differently from communities where the same number of species occur at more even densities (Cardinale et al. 2002, Kirwan et al. 2009).

In our experiment, five pats were used to sample dung beetles occurring in the area. These pats were removed after five days and all insects were extracted by floating each pat in a bucket of water (Koskela and Hanski 1977). All insects collected from the pats were sent to the authors for identification of dung beetle species. These data will allow us to test for some of the effects proposed above. However, the information will only be valid for a subset of taxa and treatments. As both dor beetles and earthworms tunnel under the pats, they are poorly quantified by the sampling methods employed (see main text). As a consequence, valid analyses can only be made on the impact of the Aphodius community on dung decomposition in treatment 4 (see Fig. 2). For this treatment, we calculated (1) the number of individuals, (2) the number of species, (3) the Shannon-Wiener diversity index (H') and (4) the equitability (J') of the dung beetle assemblage sampled from the five control pats. To estimate (5) overall biomass, we estimated the pooled biomass of the sample as åbns, where ns refers to the number of individuals of species s, and bs to the average dry mass of an individual of s. For bs, we used data from Roslin (2000). For three species for which masses were lacking (Aphodius foetens, A. lapponum, and A. rufus), we dried available individuals (for A. foetens, n = 5; for A. lapponum, n = 3; and for A. rufus, n = 28) in 60°C for 24 h, then weighed them and calculated averages across individuals.

To test for an impact of each individual factor (1)–(5) on remaining dung mass by the end of the experiment, we built a separate linear model of final pat mass (averaged across the two pats in treatment (4) for one factor at a time. To account for effects of precipitation (see main text), we started from a model including rain and then added the factor of interest. These models were fitted with PROC GLM in SAS for Windows (version 9.2, SAS Institute Inc., Cary, NC, USA).

Overall, species data were obtained from 70 farms. On three of the farms sampled, the dung pats examined contained no dung beetles. For the rest, we recorded a total of 13 Aphodius species from 370 pats examined. On average, each farm hosted 3.63 species (SE 0.22) of Aphodius beetles.

Beyond actual treatment effects (see main text), the final average mass of dung pats increased with the occurrence of rain (F1, 63 = 6.67, df = 1, P = 0.01). On top of this effect, we found limited imprints of community composition: pat mass decreased with an increasing biomass of Aphodius dung beetles (F2, 62 = 4.08, df = 1, P = 0.05), and with an increasing number of Aphodius individuals (F2, 62 = 4.69, df = 1, P = 0.03; these two factors were naturally strongly intercorrelated, r = 0.95, n = 65, P<0.0001). Nonetheless, the number of Aphodius species, the Shannon-Wiener diversity (H') of the dung beetle assemblage or its equitability (J') did not detectably affect the final weight of dung pats (Table B1).

Overall, these results show that a higher number of individuals and a higher biomass of Aphodius dung beetles are associated with higher decomposition rates. Due to experimental constraints limiting these analyses to Aphodius only (see above), they fall short of pinpointing the exact contribution of each factor to the net treatment effects observed, but add credence to former suggestions that an increase in the biomass of decomposers (O'Hea et al. 2010) and an increase in their abundance (Larsen et al. 2005) will affect decomposition rates. However, they add no support to past contentions that species numbers (Larsen et al. 2005, Slade et al. 2011) , diversity, or equitability (Gray 2000) would modify ecosystem functioning. We end by stressing that, as our data are clearly limited, these analyses should be seen as exploratory rather than conclusive.

Table B1. Generalized linear models examining the effects of Aphodius number of species, number of individuals, biomass, Shannon-Wiener index and equitability on average final pat mass per farm in treatment 4. Shown are type 3 F statistics from separate response-specific models, with the incidence of rainfall included as an offset variable (see main text). For all analysis, df = 2,62..

Explanatory variable



Number of species



Number of individuals






Shannon-Wiener index, H'



Equitability, J'



Literature Cited

Cardinale, B. J., M. A. Palmer, and S. L. Collins. 2002. Species diversity enhances ecosystem functioning through interspecific facilitation. Nature 415:426–429.

Diaz, S., A. J. Symstad, F. S. Chapin, D. A. Wardle, and L. F. Huenneke. 2003. Functional diversity revealed by removal experiments. Trends in Ecology & Evolution 18:140–146.

Gray, J. S. 2000. The measurement of marine species diversity, with an application to the benthic fauna of the Norwegian continental shelf. Journal of Experimental Marine Biology and Ecology 250:23–49.

Kirwan, L., J. Connolly, J. A. Finn, C. Brophy, A. Luscher, D. Nyfeler, and M. T. Sebastia. 2009. Diversity-interaction modeling: estimating contributions of species identities and interactions to ecosystem function. Ecology 90:2032–2038.

Koskela, H. and I. Hanski. 1977. Structure and succession in a beetle community inhabiting cow dung. Annales Zoologici Fennici 14:204–223.

Larsen, T. H., N. M. Williams, and C. Kremen. 2005. Extinction order and altered community structure rapidly disrupt ecosystem functioning. Ecology Letters 8:538–547.

O'Hea, N. M., L. Kirwan, and J. A. Finn. 2010. Experimental mixtures of dung fauna affect dung decomposition through complex effects of species interactions. Oikos 119:1081–1088.

Roslin, T. 2000. Dung beetle movements at two spatial scales. Oikos 91:323–335.

Slade, E. M., D. J. Mann, and O. T. Lewis. 2011. Biodiversity and ecosystem function of tropical forest dung beetles under contrasting logging regimes. Biological Conservation 144:166–174.

Slade, E. M., D. J. Mann, J. F. Villanueva, and O. T. Lewis. 2007. Experimental evidence for the effects of dung beetle functional group richness and composition on ecosystem function in a tropical forest. Journal of Animal Ecology 76:1094–1104.

[Back to E094-242]