Table 2.
Ordinary least squares model results for canopy temperature (CT) repeatability (response variable) and the weather (explanatory) variables most significantly and strongly correlated with repeatability, namely Rso and VPD (positive associations always).
| Variables | P-values | R2 | ||||
|---|---|---|---|---|---|---|
| VPD | Rso | |||||
| Rso | **** | 0.815 | ||||
| 2016 ArduCrop CT | VPD | **** | 0.707 | |||
| VPD | Rso | **** | **** | 0.830 | ||
| Rso | **** | 0.990 | ||||
| 2016 Airborne CT | VPD | **** | 0.851 | |||
| VPD | Rso | * | **** | 0.992 | ||
| Rso | **** | 0.792 | ||||
| 2017 Airborne CT | VPD | **** | 0.666 | |||
| VPD | Rso | ns | **** | 0.793 | ||
| Rso | **** | 0.943 | ||||
| 2017 Airborne CT | VPD | **** | 0.874 | |||
| VPD | Rso | ns | **** | 0.943 |
Air temperature and Rs were omitted from the models due to their high correlation with VPD and Rso, respectively. Statistically significant P-values are denoted:
****
P < 0.0001;
***P < 0.001;
**P < 0.01;
*
P < 0.05; ns, not significant. Scatter plot associations between CT repeatability estimates and corresponding hourly weather data are shown in Figures S20–S23 for 2016 ArduCrop, 2016 airborne, 2017 ArduCrop and 2017 airborne, respectively.