Table 1. Output of GLMM models for effect of EVI with varying time lags for total bee abundance.
Displayed are effect sizes, errors, and p-values for land use types (agricultural and urban) and EVI value. The last two columns denote whether or not there was a significant interaction (and in which direction) between each land use type and day. All significant explanatory variables with p-values <0.05 are in bold.
| Total bee abundance | AGR effect | AGR error | AGR p-value | URBAN effect | URBAN error | URBAN p-value | EVI effect | EVI error | EVI p-value | AGR × EVI | URB × EVI |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Closest collecting period | 0.03703 | 0.21186 | 0.86126 | 0.18141 | 0.24058 | 0.45081 | 0.10862 | 0.01371 | <0.001 | – | – |
| Closest collecting period—16 days | 1.09037 | 0.21596 | <0.001 | 0.6076 | 0.24396 | 0.0128 | 0.40918 | 0.01211 | <0.001 | – | – |
| Closest collecting period—32 days | 0.61977 | 0.20883 | 0.003 | 0.56755 | 0.23373 | 0.0152 | 0.44136 | 0.00947 | <0.001 | – | – |
| Closest collecting period—48 days | 0.38849 | 0.20316 | 0.0558 | 0.105908 | 0.22515 | 0.6381 | 0.368123 | 0.00807 | <0.001 | – | – |