Abstract
Due to climate change, industrial tomato crops have been prone to negative impacts at global scale on yield and phenological timing, creating a snowball effect that could affect food and economic security of countries and regions with significant agricultural production.
The case study “Food Security” implemented within the Horizon Europe project EO4EU and coordinated by MEEO with the support of CMCC, aims at demonstrating a methodology to identify the most impactful climate and spatial factors using Earth Observation data and Machine Learning (ML), to obtain useful information for the implementation of mitigation actions.
The area of interest is represented by four of the most productive Italian regions, with a focus on the Piacenza province. Climate indicators were identified to define optimal and stress conditions of tomato and they were correlated with agERA5 reanalysis climate data. The adverse climate conditions to the crop yield were identified also by studying and comparing the curve of NDVI values of tomato fields over three years (2021-2023) with a reference curve derived
from aggregated satellite data. Initial results show that industrial tomato so far has not been affected by climate change on all fronts and it may expand its geographical range in the future. Further analyses will confirm or refute these preliminary results.
The described methodology is important for highlighting critical climate conditions of other crops. Better and more sustainable land management and climate change mitigation practices may also result through these new ML algorithms for yield projections with the integration of climate simulations at very high (km scale) resolutions.