Ecological Archives E095-014-A1

Nicole L. Goebel, Christopher A. Edwards, Michael J. Follows, Jonathan P. Zehr. 2014. Modeled diversity effects on microbial ecosystem functions of primary production, nutrient uptake, and remineralization. Ecology 95:153–163. http://dx.doi.org/10.1890/13-0421.1

Appendix A. Additional diversity-effect interpretations and calculations.

The tripartite partitioned diversity effects of Fox (2005) can be interpreted as follows: A positive TICE, on average, occurs when phytoplankton types perform better than expected and without domination, inferring occupancy of different niches (that may partially overlap), facilitation, and/or negative interactions. DE is analogous to natural selection or competition; a positive (negative) DE value indicates dominance by a phytoplankton type with high (low) performance in monoculture, at the expense of other types. A large, positive DE indicates competitive exclusion among phytoplankton types that occupy similar niches in monoculture. TDCE quantifies the difference in performance of phytoplankton types in the assemblage from that expected based on monoculture performance. TDCE is positive when the high performing monocultures perform relatively high in the assemblage (high modeled Relative Yield), but without detriment to other phytoplankton types. Negative TDCE occurs when a phytoplankton type with low monoculture performance increases performance in polyculture but without compromising others. A TDCE that is larger in magnitude than a DE indicates that competition played a relatively minor role. Calculations of the diversity effects of Fox (2005) are compared to diversity effects of Loreau and Hector (2001) in Table A1.A–B.

Dmax (Loreau 1998) and log-response ratios (Hedges et al. 1999) for the maximum and net effects (LRmax and LRnet) are described here. Dmax quantifies the effect of species interactions on the performance of the assemblage (Loreau 1998). It is calculated as the difference in the response variable Y between the total of all types in polyculture (Yp) and the maximum monoculture (), normalized by . A Dmax value greater than zero indicates transgressive overyielding, whereby the polyculture outperforms any of the constituent monocultures. Non-transgressive overyielding occurs when the polyculture outperforms the mean of the constituent monocultures, but not the best performing monoculture. Underyielding occurs when the polyculture does not perform as well as expected based on constituent monocultures. The maximum effect log response ratio is the natural log of the proportional difference in the response variable Y between the mean of all types in polyculture and , and if greater than zero indicates transgressive overyielding. The net log response ratio is the natural log of the proportional difference between and the mean of constituent monocultures eq3, and tests for the change in as richness increases when averaged across all types in the assemblage. These diversity metrics have been calculated for each model simulation (Fig. A1).

 

Table A1. (A) Diversity effect calculations and (B) results that compare productivity of phytoplankton types in monoculture to productivity of each type in the 10-phytoplankton type assemblage of the baseline experiment. In (A) phytoplankton types are listed from most to least productive. Monoculture (Mi) and polyculture (Ymi) yields are in units of mg C·m-3·d-1). Calculations in white shaded area after Loreau and Hector (2001). Calculations in gray shaded area after Fox (2005).

 TableA1

 

FigA1

Fig. A1. (A) Dmax and log-response ratios (B) LRmax and (C) LRnet vs. number of phytoplankton types based on spatially and temporally averaged, vertically integrated primary productivity. Ensembles containing one to five phytoplankton types consisted of 15 replicates. The remaining ensembles consisted of 10 replicates. Ensemble means (black line) reveal a concave-down structure. Box plots show the median (gray line), 25th and 75th percentiles (box edges), the range in the most extreme data points excluding outliers (whiskers), and the outliers (points beyond approximately 2.7 standard deviations from the mean, shown as stars) for each ensemble.


 

Literature Cited

Fox, J. W. 2005. Interpreting the 'selection effect' of biodiversity on ecosystem function. Ecology Letters 8(8):846–856.

Hedges, L. V., J. Gurevitch, and P. S. Curtis. 1999. The meta-analysis of response-ratios in experimental ecology Ecology 80:1150–1156.

Loreau, M. 1998. Separating sampling and other effects in biodiversity experiments. Oikos 82:600–602.

Loreau, M. and A. Hector. 2001. Partitioning selection and complementarity in biodiversity experiments. Nature 412(6842):72–76.


[Back to E095-014]