Scale-dependent effects of climate on two copepod species, Calanus glacialis and Pseudocalanus minutus, in an Arctic-boreal sea

Transcription

1 The following supplement accompanies the article Scale-dependent effects of climate on two copepod species, Calanus glacialis and Pseudocalanus minutus, in an Arctic-boreal sea Jonas Persson 1,3, Leif Chr. Stige 1, Nils Chr. Stenseth 1, *, Nikolay Usov 2, Daria Martynova 2 1 Centre for Ecological and Evolutionary Synthesis, Department of Biology, University of Oslo, Postboks 1066, Blindern, 0316 Oslo, Norway 2 White Sea Biological Station, Zoological Institute, Russian Academy of Sciences, Universitetskaya nab., 1, , St. Petersburg, Russia 3 Present address: Norwegian Institute for Water Research (NIWA), Gaustadalléen 21, 0349 Oslo, Norway *Corresponding author. n.c.stenseth@bio.uio.no Marine Ecology Progress Series 468: (2012) Supplement. Residual diagnostics plots (Figs. S1 to S5) and model predictions for the biomasses of the different stages of Calanus glacialis and Pseudocalanus minutus at different temperatures and months (Table S1)

2 Fig. S1. Quantile quantile plot from each model. The dotted line shows the 1:1 relationship Calanus glacialis Pseudocalanus minutus Residual quantiles Standard normal quantiles

3 Fig. S2. Autocorrelation functions of residuals from each model. The height of the bars indicates the autocorrelation function at different lags (months). Blue dashed lines: statistical significance = Significant positive autocorrelations at lag 12 mo suggest that the models fail to account for effects of unknown factor(s) that vary on a multi-year time-scale Autocorrelation function Lag (mo)

4 Fig. S3. Monthly residuals from each model Calanus glacialis Pseudocalanus minutus Model residual Year

5 Fig. S4. Autocorrelation functions of annual means of residuals from each model. The height of the bars indicates the autocorrelation function at different lags (years). Blue dashed lines: statistical significance = Significant positive autocorrelations suggest that the models fail to account for effects of unknown factor(s) that vary on a multi-year time-scale Calanus glacialis Pseudocalanus minutus Autocorrelation function Lag (yr)

6 Fig. S5. Annual means of the residuals from each model Calanus glacialis Pseudocalanus minutus Model residual Year

7 Table S1. Calanus glacialis, Pseudocalanus minutus. The 10, 50 and 90% percentiles of ambient temperatures and predicted biomasses at these temperatures for each stage and the total biomasses of C. glacialis and P. minutus. Biomass prediction means are accompanied by 95% confidence intervals (lcl: lower confidence limit; ucl: upper confidence limit). Difference is defined as the proportional difference in predicted biomass between cold and warm temperatures; the results on a log scale and on a normal scale are given. The largest absolute difference for each stage and scale is in bold Temperature ( C) Biomass (mg dry wt m 2 ) Difference Normal Stage Month 10% 50% 90% Lcl 10% Mean 10% Ucl 10% Lcl 50% Mean 50% Ucl 50% Lcl 90% Mean 90% Ucl 90% Log scale scale (%) Calanus glacialis Nauplius larvae

8

9 Adult males Adult females

10 Pseudocalanus minutus Nauplius larvae

11

12 Adult males Adult females

13