CANOP (Remotely sensed leaf biochemistry intra-individual variability in orchard tree CANOPies for agroecology)

Objective

The CANOP project targets (sub)centimeter scale studies for mapping the intra-individual variability of foliar biochemical traits in orchard trees using multi-modal optical remote sensing data. The objective is to improve the characterization of the health/nutritional status of each tree, for the evaluation of two agroecological practices for apricot and peach orchards: optimized management of inputs and selection of resilient varieties. The scientific challenges are as follows: (i) a better physical understanding of small-scale wave-matter interactions (particularly directional ones), based on the complementarity of spectroscopic, polarimetric and LiDAR 3-D data, (ii) the development of robust and transposable regression methods, between vegetation traits and metrics derived from remote sensing data, by combining physical modelling approaches and machine learning methods, and (iii) performing a multi-scale analysis to assess the contribution of intra-individual versus inter-individual variability mapping of these vegetation traits, using experiments carried out both in the laboratory and in the field, drone acquisitions and the use of satellite data, i.e. with spatial resolutions ranging from the (sub)centimeter to the (deca)meter scale. The results are expected to strengthen the role of optical remote sensing that is non-destructive, high-throughput and dynamic to assess vegetation condition on a large scale compared to laborious and local observations in the field and in laboratory, for future tree-based agricultural practices for precision farming.

Role of CAPTE/EMMAH

Within CANOP, EMMAH CAPTE (Sylvain Jay) brings its expertise in remote sensing of vegetation at centimetric scales for phenotyping (optical data acquisition and processing, radiative transfer and estimation methods at leaf and canopy scales). With Karine Adeline (PI, ONERA) , he co-supervise wthe PhD of Nathan Sikora, whose objective is to evaluate the complementarity of spectroscopic, polarized and LiDAR data to better estimate leaf biochemical traits at centimeter scale and capture variability in the functional state of individual trees.