Independent variable | IRR1
| Standard error | 95 % CI2
|
p-value |
|---|
Near-ground temperature | 1.009 | 0.002 | 1.005–1.012 | <0.001 |
daylight length | 1.006 | 0.000 | 1.006 | <0.001 |
relative humidity (RH) | 1.004 | 0.001 | 1.003–1.006 | <0.001 |
year | | | | <0.001 |
2002 | 0.821 | 0.023 | 0.777–0.868 | <0.001 |
2003 | 0.539 | 0.018 | 0.506–0.574 | <0.001 |
2004 | 0.949 | 0.027 | 0.898–1.003 | 0.062 |
2005 | 1.276 | 0.033 | 1.214–1.342 | <0.001 |
2006 | 0.924 | 0.027 | 0.874–0.979 | 0.007 |
period | | | | <0.001 |
summer | 0.767 | 0.023 | 0.723–0.814 | <0.001 |
autumn | 0.785 | 0.044 | 0.703–0.876 | <0.001 |
- Parameter estimates were determined from the Poisson regression model. Temperature and daylight length are continuous variables whereas year and seasonal period are categorical variables. The reference year is taken as 2001 and the reference season as spring. IRRtemperature =1.009 indicates that if the temperature increases by 1 °C, the incidence increases 1.009 times (by 0.9 %). IRRdaylight =1.0062 indicates that for a change of 1 min, the incidence increases 1.0062 times (by 0.62 %). IRRRH = 1.004 indicates that for a change of 1 % RH, the incidence increases 1.004 times (by 0.4 %). For year, IRR values less or greater than 1 indicate that the number of nymphs in the given year is lower or higher than in 2001, the reference year. For period, IRR values are less than 1 indicating that numbers of questing nymphs are 23 % (summer) and 44 % (autumn) less than during the spring period. All the independent variables were analysed simultaneously in one model so that their effects are adjusted for each other. All IRR values are significant (p = 0.007 or <0.001) except for 2004, which was not significantly different from 2001
-
IRR incidence rate ratio; 95 % CI 95 % confidence interval
