Study highlights importance of dedicated exits for vulnerable populations in building evacuations
image: (a)different geometric relations can be calculated.The position directly below the drone is in nadir orientation. The vertical angle at which drone and sun are seen from the observed plot are the elevations of drone and sun respectively. The azimuth of the sun is the clockwise horizontal angle at which the sun is seen from the observed plot from north (b)The position of the plot can be described as planar distance between drone and plot in direction of the sun (b) or in sowing row direction (c) Another option to describe the positon of the plot relative to the drone is by viewing angles as is shown for angles relative to sun direction (d)but not shown for the sowing row direction. Elements in the principal optical planes in drone or sun direction are in bright blue, cardinal direction in dark blue. The dimensions of interest and related covariates are in orange. Small black angle marks and short parallel black lines indicate perpendicularity and parallelism respectively.
Credit: The authors
By applying a multi-view imaging and correction framework across 99 drone flights, the researchers showed that over 96% of observed variance in canopy temperature could be explained, allowing meaningful correlations with yield, plant height, canopy cover, and other phenotypic traits.
Canopy temperature is widely recognized as a sensitive proxy for plant water use, stomatal conductance, and stress responses. While drone-based thermography offers a fast and scalable way to capture CT data across hundreds of experimental plots, measurements are prone to confounding effects—ranging from solar radiation and wind, to thermal drift of sensors, soil heterogeneity, and image viewing geometry. Such distortions can obscure genotype-specific signals and lead to misleading conclusions in breeding trials. Based on these challenges, a robust method is urgently needed to distinguish true biological variation from environmental noise and technical artifacts.
A study (DOI: 10.1016/j.plaphe.2025.100046) published in Plant Phenomics on 30 April 2025 by Simon Treier’s team, Agroscope, transforms UAV-based canopy temperature data into reliable indicators of crop performance, enabling more accurate phenotyping for breeding and precision agriculture.
The team conducted two consecutive years of winter wheat variety testing at Agroscope research fields in Switzerland. They evaluated 30 European wheat varieties and additional Swiss genotypes under contrasting fertilizer and management regimens, amounting to nearly 500 experimental plots. Across 99 drone flights using an uncooled thermal infrared sensor, each plot was imaged from multiple angles. Unlike conventional orthomosaic-based approaches, the multi-view method provided several CT readings per plot and captured covariates such as trigger timing and camera geometry. These were then integrated into a two-stage statistical correction pipeline: first adjusting for temporal and spatial trends (e.g., sensor drift, field heterogeneity), and then for geometric effects (e.g., vignetting, bidirectional reflectance). Complementary field measurements—including yield, plant height, canopy cover, senescence, leaf rolling, and multispectral indices—were collected to validate CT estimates. Uncorrected CT measurements showed strong spatiotemporal distortions and often inconsistent correlations with phenotypic traits. After applying corrections, variance decreased dramatically—from as high as 8.4 K² in raw data to below 0.5 K² in corrected datasets. Corrected canopy temperatures correlated significantly and consistently with yield, plant height, and fractional canopy cover, while also reflecting stress indicators such as leaf rolling. For instance, in one trial, yield correlations with CT shifted from weak and inconsistent in raw data to strong and significant across all flights after correction. The study further demonstrated how treatment effects (e.g., fertilization) could mask genotype-specific responses, underscoring the value of separating treatment-driven variance from intrinsic plant traits. Overall, the approach enabled researchers to distinguish true physiological signals from noise, making CT a reliable phenotyping trait in large-scale breeding trials.
This methodology provides a foundation for more robust use of thermal phenotyping in agriculture. By establishing correction protocols that account for environmental dynamics, sensor limitations, and field geometry, the study opens new possibilities for breeding climate-resilient crops, monitoring water stress in precision agriculture, and advancing high-throughput plant phenotyping. As agriculture faces increasing pressures from climate change and resource constraints, integrating drone-based thermal imaging with advanced analytics could become a cornerstone for sustainable crop improvement and management strategies.
###
References
DOI
Original URL
https://doi.org/10.1016/j.plaphe.2025.100046
Funding information
This study was financed by Agroscope and the work of Simon Treier was in part supported by the two H2020 projects InnoVar and Invite.
About Plant Phenomics
Plant Phenomics is dedicated to publishing novel research that will advance all aspects of plant phenotyping from the cell to the plant population levels using innovative combinations of sensor systems and data analytics. Plant Phenomics aims also to connect phenomics to other science domains, such as genomics, genetics, physiology, molecular biology, bioinformatics, statistics, mathematics, and computer sciences. Plant Phenomics should thus contribute to advance plant sciences and agriculture/forestry/horticulture by addressing key scientific challenges in the area of plant phenomics.