Djuradj Milošević1*, Milica Stojković1, Dubravka Čerba3, Ana Petrović2, Momir Paunović4

Djuradj Milošević1*, Milica Stojković1, Dubravka Čerba3, Ana Petrović2, Momir Paunović4, Vladica Simić2*

Different aggregation approaches in the chironomid community and the threshold of acceptable information loss

1 Department of Biology and Ecology, Faculty of Sciences and Mathematics, University of Nis, Visegradska 33, 18000 Nis, Serbia

2 Institute of Biology and Ecology, Faculty of Science, University of Kragujevac, Radoja Domanovica 12, 34000 Kragujevac, Serbia

3 Department of Biology, University of J. J. Strossmayer in Osijek, Cara Hadrijana 8A, HR-31000 Osijek, Croatia

4Institute for Biological Research “Siniša Stanković”, University of Belgrade, Bulevar despota Stefana 142, 11000 Belgrade, Serbia

*Correspondence to: Djuradj Milošević and Vladica Simić

E-mail: ;

Online resources

Appendix 1

Fig. S1Spearman’s correlation (ρ) values between the species-level matrix and each randomly aggregated matrix plotted against the corresponding lnϕ values. A dotted line indicates ϕlow sufficient to obtain results consistent with those obtained analyzing species-level data.

Appendix 2

Table S1Percentage of PERMANOVAs on random aggregated data consistent with those from species-level analyses, at gradient of aggregation levels (ϕ). Bold values represent the minimum number of surrogates (Gmin) and the corresponding lowest practicable aggregation level (ϕlow). Underlined values indicate the number of selected surrogates for BestAgg (GBestAgg) and the corresponding aggregation level for BestAgg (ϕBestAgg).

Table S2 Results of SIMPER analysis reporting the % contribution of taxa to assemblage dissimilarities between sites at different elevations (four categories: 1-up to 200 m.a.s.l., 2-from 200 to 500 m.a.s.l., 3-from 500 to 800 m.a.s.l. and 4-over 800 m.a.s.l.). Species with contributions higher than 3% in at least one pair-wise comparison are considered. The Average dissimilarities between assemblages are presented in brackets.

Table S3 Set of surrogates selected by the BestAgg procedure in the Southern Morava river system. Relevance is reported according to the SIMPER analysis (See Table S2). In the Easiness column, Y stands for easy identification and N for difficult identification (requires slide mounting). The Resemblance column contains aggregation criteria or NA if they did not apply. The number in brackets indicates the number of species included in the obtained surrogate or taxa. Priority was determined following the procedure described by Bevilacqua et al. (2013).