Using thermal imagery and changes to stem radius to assess water stress in two coniferous tree species
With a warming climate and greater evaporative demand, many forest ecosystems are increasingly affected by water limitation as prolonged water deficits reduce tree-level growth and survival. Monitoring water deficit traditionally requires daily measurements of sap flow or radial flux from automated sensors mounted on individual trees. As an alternative approach, we evaluated the use of airborne thermal imagery from unmanned aerial vehicles as a rapid, scalable tool for assessing tree-level water stress. Plant water stress leads to higher leaf temperatures when soil moisture is low and evaporative demand is high. To detect this response, we modelled the difference between leaf and air temperature (∆T) as a function of local soil moisture vapour pressure deficit, and wind speed for two tree species, lodgepole pine (Pinus contorta var. latifolia) and white spruce (Picea glauca). We used those same weather and soil conditions to model dendrometer-based measurements of daily changes in internal tree water deficit (∆TWD). While canopy leaf temperature and daily change in tree water deficit showed little direct correlation with one another, both variables responded to soil moisture, vapour pressure deficit, and wind speed in a manner that reflects a common response to water stress as soil became progressively drier over the summer months. The two species showed similar responses to environmental conditions, with some differences related to species-specific strategies for drought avoidance. The application of thermal imagery to detect water stress in natural ecosystems can improve understanding how trees experience water stress across species and environmental conditions.