This INNO-VEG project aimed to (i) evaluate the suitability of using crop sensing data to carry out measurements in field experiments and (ii) to define and implement a new approach for delivering cost effective research in the field vegetable and potato sectors. A key advantage of using crop sensing data to assess treatments is the ability to upscale from small plot to field scale experiments, as crop sensing data can be relatively easily collected from larger field areas using drones or tractor mounted sensors.
Crop sensing is simply the process of using sensors to collect information about a growing crop. Differences in reflectance at specific wavelengths can be expressed as a Vegetation Index (VI)– these can be calculated in many ways, but the most well known is the Normalised Difference Vegetation Index (NDVI). Since reflectance from the crop is determined by the size and vigour of its canopy, VIs have been shown to correlate well with crop characteristics such as above ground biomass and crop vigour. The INNO-VEG project focussed on seven VIs: NDVI, MCARI2, MTCI, CIgreen, CIRedEdge, NDRE and REIP.
In 2019 a programme of 46 small plot field experiments was carried out across the UK, France, Belgium, and the Netherlands to develop an overarching ‘Protocol’ for integrating crop sensing data into field research methodologies. In these experiments, the results from traditional field measurements (i.e. hand harvest or experimental harvest machine assessments of yield and crop quality) were correlated against VI data to evaluate the suitability of crop sensing data to assess treatment differences in field experiments. These experiments covered several horticultural crop groups including potatoes, brassicas, alliums, leafy salads, carrots, vining peas, and cucurbits to ensure the ‘Protocol’ has broad relevance for field vegetable and potato research. There was a good relationship between VI data and crop yields for most crops tested, although the strength of the correlations varied between crops and depending on the timing of the measurements relative to the crop growth stage.
In 2020, field validation experiments were set up to test the ‘Protocol’ in larger field-scale experiments and to develop this into a ‘Framework for farmer-led research’. Drone mounted sensors were used to collect crop reflectance data. This data was processed to calculate spatial VI data, which was statistically analyzed using the ADAS Agronomics process to estimate average treatment effect(s). In 2021, the Framework was tested in 20 farmer led field scale experiments. In these experiments, the farmers took a greater role in planning and setting up the experiments.
Both the ‘Protocol’ and ‘Framework’ are available to download from the ‘Resources’ section of the INNO-VEG project website (https://inno-veg.org/en/Resource). The ‘Protocol’ is aimed at researchers, agronomists and farmers who want to use crop sensing technology to assess their crops and aims to support them to make best use of the technology. The ‘Framework’ provides farmers with the information they require to set up and run field scale experiments including experimental design, application of treatments and sourcing crop sensing data.
Co-authors: Dowers, J., Roques, S., Williams, J.R., Ampe, E., Van Oers, C. and Cohan, J.P.
Acknowledgements: This project has received funding from the Interreg 2 Seas programme 2014-2020 co-funded by the European Regional Development Fund under subsidy contract No 2S05-032.