High-throughput drone-based remote sensing reliably tracks phenology in thousands of conifer seedlings

Abstract

Phenology is an important indicator of environmental variation and climate change impacts on tree responses. In conifers, monitoring phenology of photosynthesis through remote sensing has been unreliable, because needle foliage varies little throughout the year. This is challenging for modelling ecosystem carbon uptake and monitoring phenology for enhanced breeding (genomic selection) and forest health. Here, we demonstrate that drone-based carotenoid-sensitive spectral indices, such as the Chl/carotenoid index (CCI), can be used to track phenology in conifers by taking advantage of the close relationship between seasonally changing carotenoid levels and the variation of photosynthetic activity. Physiological ground measurements, including photosynthetic pigments and maximum quantum yield of Chl fluorescence, indicated that CCI tracked the variation of photosynthetic activity better than other vegetation indices for 30 white spruce seedlings measured over 1 yr. A machine-learning approach, using CCI derived from drone-based multispectral imagery, was used to model phenology of photosynthesis for the entire pedigree population (6000 seedlings). This high-throughput drone-based phenotyping approach is suitable for studying climate change impacts and environmental variation on the physiological status of thousands of field-grown conifers at unprecedented speed and scale.

Keywords: Chl/carotenoid index (CCI); drone; evergreens; functional traits; high-throughput phenotyping; phenology; pigments; unmanned aerial vehicle (UAV).