ELECTRONIC SUPPLEMENTAL INFORMATION
Heino J, Alahuhta J, Fattorini S (2015) Phylogenetic diversity of regional beetle faunas at high
latitudes: patterns, drivers and chance along ecological gradients. Biodiversity and Conservation.
Appendix S1. Alternative spatial modelling and variation partitioning.
As a substitute of latitude (see main text), we also used distance-based Moran eigenvector maps
(db-MEM) to model spatial structures among the provinces and to provide spatial variables for our
alternative modelling endeavors (Legendre & Legendre 2012). Of the methods belonging to the
family of db-MEMs, we used the traditional principal coordinates of neighbour matrix analyses
(PCNM) based on Euclidean distances among the centroids of the provinces (Borcard, Gillet &
Legendre 2011). We used the resulting PCNM eigenvectors showing positive spatial autocorrelation
as explanatory variables in further analyses aimed at explaining variation in phylogenetic diversity
and species richness across the 79 provinces. The first PCNM eigenvectors with large eigenvalues
describe broad-scale spatial structures, whereas the PCNM eigenvectors with small eigenvalues
describe fine-scale spatial variation (Legendre & Legendre 2012). The PCNM eigenvectors are
mutually orthogonal, linearly unrelated spatial variables and can thus be used to account for spatial
autocorrelation in diversity data (Legendre & Legendre 2012). Significant spatial patterns in biotas
related to such spatial variables may result from environmental autocorrelation, dispersal limitation
or historical effects on biodiversity (Legendre & Legendre 2012). PCNM analysis was conducted
using the R package PCNM (Legendre et al. 2012). We selected significant spatial variables in the
alternative linear regression models following the forward selection method with two stopping rules
(Blanchet et al. 2008) using the function “ordiR2step” in the R package vegan (Oksanen et al.
1
2013). We also ran variation partitioning between four sets of explanatory variables using the
“varpart” function in the R package vegan (Oksanen et al. 2013).
The db-MEM analysis produced 28 spatial variables showing positive spatial
autocorrelation. Forward selection with two stopping rules showed that the spatial models for
species richness, AvTD and VarTD included 11, 10 and 13 spatial variables, respectively. Spatial
variables related to both large and small scales were included in the spatial models. Alternative
variation partitioning with latitude substituted by the db-MEM spatial variables showed, not
surprisingly, that the pure spatial effects increased in importance, whereas those of the ecological
variables clearly decreased in importance (Table S1). Residuals of these models did not show
significant spatial autocorrelation.
References
Blanchet, F.G., Legendre, P. & Borcard, D. (2008) Forward selection of explanatory variables.
Ecology, 89, 2623–2632.
Borcard, D., Gillet, F. & Legendre, P. (2011) Numerical Ecology with R. Springer, New York.
Legendre, P., Daniel Borcard, D., Blanchet, F.G. & Dray, S. (2012) PCNM: MEM spatial
eigenfunction and principal coordinate analyses. R package version 2.1-2/r106.
http://R-Forge.R-project.org/projects/sedar/
Legendre, P. & Legendre, L. (2012) Numerical Ecology. Third Edition. Elsevier, Amsterdam.
Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O'Hara, R.B., Simpson, G.L.,
Solymos, P., Stevens, M.H.H. & Wagner, H. (2014). vegan: Community Ecology
Package. R package version 2.2-0. http://CRAN.R-project.org/package=vegan
2
Table S1. Alternative variation partitioning tables for species richness, average taxonomic distinctness (AvTD) and variation in taxonomic
distinctness (VarTD). For comparison, see Table 1 in the main paper.
Geography
Variables
X1 = MEM
X2 = Area
X3 = Altitude mean
X4 = Altitude range
X1+X2
X1+X3
X1+X4
X2+X3
X2+X4
X3+X4
X1+X2+X3
X1+X2+X4
X1+X3+X4
X2+X3+X4
All
Pure fractions
X1 | X2+X3+X4
X2 | X1+X3+X4
X3 | X1+X2+X4
X4 | X1+X2+X3
Joint fractions
X1∩X2
X2∩X3
X1∩X3
X1∩X4
X2∩X4
X3∩X4
X1∩X2∩X3
X1∩X3∩X4
X2∩X3∩X4
X1∩X3∩X4
X1∩X2∩3∩X4
Residuals
Climatic variability
Richness
Adj. R2
0.779
0.310
0.387
0.413
0.778
0.789
0.798
0.479
0.491
0.406
0.794
0.809
0.796
0.486
0.806
AvTD
Adj. R2
0.610
0.065
0.084
0.132
0.618
0.605
0.625
0.096
0.133
0.152
0.618
0.653
0.639
0.153
0.666
VarTD
Adj. R2
0.807
0.412
0.414
0.414
0.810
0.805
0.808
0.567
0.557
0.416
0.810
0.817
0.806
0.562
0.814
0.320
0.010
-0.003
0.012
0.513
0.027
0.014
0.049
0.252
0.008
-0.002
0.004
0.069
0.000
-0.003
-0.005
-0.005
0.019
0.018
-0.001
-0.006
0.155
0.225
0.194
-0.026
0.000
0.007
0.008
-0.015
-0.014
0.026
-0.001
-0.005
0.025
0.058
0.334
0.138
0.000
0.007
-0.009
-0.004
0.002
0.010
-0.002
-0.001
0.148
0.263
0.186
Temperature
Variables
X1 = MEM
X2 = Area
X3 = TemAnnRange
X4 = PrecSeaCV
X1+X2
X1+X3
X1+X4
X2+X3
X2+X4
X3+X4
X1+X2+X3
X1+X2+X4
X1+X3+X4
X2+X3+X4
All
Pure fractions
X1 | X2+X3+X4
X2 | X1+X3+X4
X3 | X1+X2+X4
X4 | X1+X2+X3
Joint fractions
X1∩X2
X2∩X3
X1∩X3
X1∩X4
X2∩X4
X3∩X4
X1∩X2∩X3
X1∩X3∩X4
X2∩X3∩X4
X1∩X3∩X4
X1∩X2∩3∩X4
Residuals
3
Richness
Adj. R2
0.779
0.310
0.092
0.276
0.778
0.796
0.776
0.301
0.383
0.269
0.794
0.775
0.794
0.411
0.792
AvTD
Adj. R2
0.610
0.065
-0.012
0.024
0.618
0.634
0.606
0.104
0.056
0.054
0.636
0.615
0.646
0.137
0.648
VarTD
Adj. R2
0.807
0.412
0.108
0.192
0.810
0.810
0.810
0.406
0.423
0.185
0.811
0.813
0.809
0.432
0.811
0.381
-0.002
0.017
-0.002
0.512
0.002
0.033
0.012
0.379
0.002
-0.002
0.000
0.144
0.001
0.010
0.111
0.000
-0.001
0.068
-0.035
0.000
-0.035
0.135
0.208
0.081
0.007
0.047
0.020
0.000
-0.015
0.033
-0.058
-0.001
-0.027
0.001
0.352
0.245
0.001
0.010
0.027
0.000
0.003
0.052
-0.017
0.000
-0.018
0.130
0.189
Variables
X1 = MEM
X2 = Area
X3 = Mintem
X4 = MaxTem
X1+X2
X1+X3
X1+X4
X2+X3
X2+X4
X3+X4
X1+X2+X3
X1+X2+X4
X1+X3+X4
X2+X3+X4
All
Pure fractions
X1 | X2+X3+X4
X2 | X1+X3+X4
X3 | X1+X2+X4
X4 | X1+X2+X3
Joint fractions
X1∩X2
X2∩X3
X1∩X3
X1∩X4
X2∩X4
X3∩X4
X1∩X2∩X3
X1∩X3∩X4
X2∩X3∩X4
X1∩X3∩X4
X1∩X2∩3∩X4
Residuals
Richness
Adj. R2
0.779
0.310
0.392
0.528
0.778
0.778
0.830
0.400
0.593
0.750
0.776
0.841
0.828
0.769
0.846
AvTD
Adj. R2
0.610
0.065
0.045
0.421
0.618
0.614
0.676
0.055
0.414
0.421
0.616
0.705
0.671
0.430
0.706
VarTD
Adj. R2
0.807
0.412
0.380
0.512
0.810
0.811
0.804
0.444
0.641
0.728
0.812
0.808
0.810
0.724
0.809
0.077
0.019
0.005
0.071
0.276
0.035
0.002
0.090
0.085
-0.001
0.001
-0.003
0.001
-0.008
0.170
0.299
-0.021
-0.008
0.010
0.054
0.009
-0.079
0.247
0.154
-0.025
-0.006
0.015
0.286
-0.033
-0.004
0.033
-0.011
0.013
-0.024
0.060
0.294
-0.003
0.004
0.082
0.283
0.001
0.000
0.066
0.128
-0.001
-0.052
0.218
0.191
1
Fig. S1. Correlations between species richness, average taxonomic distinctness (AvTD), variation
2
in taxonomic distinctness (VarTD) and latitude. Upper diagonal shows Pearson correlations
3
between the variables.
0.594
-0.759
-0.794
AvTD
-0.570
-0.452
VarTD
0.707
71
200
Richness
400
600
4
1.0e+07
340
360
380
65
67
69
x
8.0e+06
Latitude
5
200
400
600
340
4
360
380