Quantifying Canopy-Scale Plant Water Use and Stress
Measuring fine-scale transpiration with UAV-based thermal and atmospheric data
Understanding how plants use water is critical for managing ecosystems in a changing climate. Yet measuring transpiration at the scale of individual plants has been a challenge, especially across large landscapes.
Our work focuses on directly estimating transpiration by capturing canopy-scale latent heat flux using a custom UAV platform equipped with thermal infrared, multispectral, and atmospheric sensors.
A new way to measure transpiration
Our method combines high-resolution thermal imagery with onboard atmospheric profiling to measure latent heat exchange.
We implemented two physics-based approaches:
Surface Energy Balance: This method calculates latent heat flux by closing the surface energy budget. We measure net radiation and estimate soil heat flux, then subtract sensible heat flux, which we derive from surface–air temperature differences and resistance to heat transfer.
Bowen Ratio: This approach uses vertical gradients in temperature and humidity to calculate the Bowen ratio, a measure of the partitioning between sensible and latent heat. From this ratio and the available energy, we can solve directly for transpiration without relying on resistance terms or empirical stability corrections.
Custom UAV system
Our UAV system includes:
MicaSense Altum camera (thermal + 5-band multispectral)
Dual pyrometers for incoming/outgoing shortwave radiation
TriSonica Mini 3D sonic anemometer for air temperature, humidity, wind, and pressure
Onboard GNSS and data logger for accurate georeferencing and synchronized data streams
All data required for latent heat calculation are acquired using the UAV, allowing us to map transpiration without relying on towers or ground meteorological stations.