Table 2. Linear models of the drivers of multi-trait FDses for dragonfly (dr) and damselfly (da) assemblages for both analyses: across study regions (Main) and within the phenologically higher resolution dataset of Great Britain (Phenological). Coefficients are standardised. Assemblages of Great Britain within the main analysis were undersampled (see methods). sd shows, for the main analysis, parameter variation across 100 models using different assemblages subset.
Analysis   Model Predictor Coef ± sd t ± sd p
Main dr Multi-trait FDses ~ Temp*Study-region+ Richness +Day F7,388=14.54 ± 1.67, R2=0.19 ± 0.02 p<0.001 Richness 0.14 ± 0.01 3.46 ± 0.36 <0.05*
    Day 0.12 ± 0.01 3.04 ± 0.37 <0.05*
    Temp 1.50 ± 0.01 6.17 ± 0.11 <0.05*
    DatasetCat -0.14 ± 0.01 -0.55 ± 0.05 >0.05
    DatasetGB 0.67 ± 0.13 3.36 ± 0.68 <0.05*
    Temp*DatasetCat -1.75 ± 0.01 -5.90 ± 0.11 <0.05*
    Temp*DatasetGB -0.50± 0.10 -1.65 ± 0.34 >0.05
             
  da Multi-trait Fdses ~Day*Study-region F5,264=13.61 ± 2.42, R2=0.19 ± 0.03, p<0.001 Day -0.35 ± 0.01 -3.67 ± 0.11 <0.05*
    DatasetCat 0.28 ± 0.02 1.94 ± 0.14 <0.05*
    DatasetGB 0.34± 0.07 2.84 ± 0.62 <0.05*
    Day*DatasetCat 0.70 ± 0.01 5.55 ± 0.16 <0.05*
    Day*DatasetGB -0.14 ± 0.09 -1.03 ± 0.68 >0.05
             
Phenological dr Multi-trait FDses ~Temp +Richness F2,1553=116.47, R2=0.13, p<0.001 Richness 0.23±0.02 12.60 <0.05*
      Temp 0.11±0.02 5.88 <0.05*