Periodic Reporting for period 1 - PHENET (Tools and methods for extended plant PHENotyping and EnviroTyping services of European Research Infrastructures)
Reporting period: 2023-01-01 to 2023-12-31
In PHENET, the European Research Infrastructures (RI) on plant phenotyping (EMPHASIS), ecosystems experimentation (AnaEE), long-term observation of interactions between nature and people (eLTER) and data science (ELIXIR) join their forces to co-develop, with a diversity of innovative companies, new tools and methods to help the identification of future-proofed agroecosystems in front of their burning challenges (climate change, agroecology transition, biodiversity loses…).
While RI are well equipped with highly instrumented sites, PHENET will design new services allowing wide access to enlarged sources of in-situ (on farm, in natura…) phenotypic and environmental data in support to the evaluation of agro and ecosystems thanks to big data strategies relying on (i) new IoT (Internet of Things) based multi-sensors devices and algorithms capturing a variety of phenotypic and environmental information (ii) unleashed access to high resolution Earth Observation data and (iv) new generation of modeling solutions leveraging on AI and digital twins. These developments are challenged by and implemented in a series of eight Use Cases covering a large range of agroecosystems and ecosystems in order to demonstrate the portability of solutions. A large effort is devoted to training RI staff and beyond through a collection of training material. Outreaching activities aim at enlarging the range of RI users. PHENET also ambitions to have impact on the development of innovative companies on phenotyping, envirotyping and precision agriculture as well as on the emergence of « future proofed » crop varieties and innovative practices fitted to climate change and agroecological transition.
While RI are well equipped with highly instrumented sites, PHENET will design new services allowing wide access to enlarged sources of in-situ (on farm, in natura…) phenotypic and environmental data in support to the evaluation of agro and ecosystems thanks to big data strategies relying on (i) new IoT (Internet of Things) based multi-sensors devices and algorithms capturing a variety of phenotypic and environmental information (ii) unleashed access to high resolution Earth Observation data and (iv) new generation of modeling solutions leveraging on AI and digital twins. These developments are challenged by and implemented in a series of eight Use Cases covering a large range of agroecosystems and ecosystems in order to demonstrate the portability of solutions. A large effort is devoted to training RI staff and beyond through a collection of training material. Outreaching activities aim at enlarging the range of RI users. PHENET also ambitions to have impact on the development of innovative companies on phenotyping, envirotyping and precision agriculture as well as on the emergence of « future proofed » crop varieties and innovative practices fitted to climate change and agroecological transition.
One of the objectives of PHENET is to co-develop with industry, new, environment friendly, low-cost, AI-based, automated and connected devices gathering sets of sensors able to capture a large diversity of traits (phenotyping), environmental variables (envirotyping) and processes in agroecosystems, in particular related to agroecological transition and climate change (e.g. soil carbon, biodiversity, plant biotic stress tolerance...). The 1st year of the project focused on establishing the specifications of these devices gathering scientists, engineers together with end users from UseCases. This was done through a series of meetings - including in persons - hand-on sessions. Engineers traveled from one place to another to challenge solutions. Regarding the « connected sticks » (IoT), two different solutions targeting distinct objectives were selected and first series of few tenths will be implemented in farmers’s field and in experimental stations from partners in 2024. Regarding the orchard phenomobile, several exchanges took place to agree on the traits to focus on.
Another objective of PHENET is to increase the use of satellite images which are still underused in support to phenotyping and envirotyping while their temporal and spatial resolutions rapidly improve. We designed connected stick that will be used in support to ground truthing. Moreover, an automated method was designed to produce maps of intra-field structural heterogeneity based on multi-year EO time series from Copernicus Sentinels satellite imagery. Once validated, this method will be used to analyse within farms variation of crop performances, intra-ecosystems variability, and help analyse year to year or field to field variation in innovative management practices such as inter-cropping for which huge variability still exists.
Several interactions between the Use Cases and the data management team of PHENET set the scene and develop appropriate tools to secure future data and analytic flows within the project in particular towards the modelling team that aims to test new modelling solutions (including hybrid modelling between AI and process based models and digital twins). As data are not generated yet from the project, the teams are proofing their methods against data from past projects.
Another objective of PHENET is to increase the use of satellite images which are still underused in support to phenotyping and envirotyping while their temporal and spatial resolutions rapidly improve. We designed connected stick that will be used in support to ground truthing. Moreover, an automated method was designed to produce maps of intra-field structural heterogeneity based on multi-year EO time series from Copernicus Sentinels satellite imagery. Once validated, this method will be used to analyse within farms variation of crop performances, intra-ecosystems variability, and help analyse year to year or field to field variation in innovative management practices such as inter-cropping for which huge variability still exists.
Several interactions between the Use Cases and the data management team of PHENET set the scene and develop appropriate tools to secure future data and analytic flows within the project in particular towards the modelling team that aims to test new modelling solutions (including hybrid modelling between AI and process based models and digital twins). As data are not generated yet from the project, the teams are proofing their methods against data from past projects.
As the project is just at its onset, first results are just arriving and impacts must then be considered as potentials. A few of them can be summarized here.
PHENET will contribute to the development of innovative companies, both directly with the involved partner companies and indirectly through companies interacting with PHENET partners through national and EU projects, initiatives, living labs etc. For instance, PHENET will improve services around treatments of satellite imaging with a higher spatial distribution, thereby considerably facilitating its access to phenotyping market, but also to the "precision agriculture" and "sustainable agriculture". More generally, PHENET will demonstrate that SME operating close to farmers / foresters decisions and innovation can develop as main actors of transition of agroecosystems e.g. towards carbon sequestration or precision farming/forestry applications. An innovation Hub (WP6) will connect the academic stakeholders and industry as well as practitioners
PHENET ensures a long-lasting relationship between industry and RI, which allows RIs to benefit from industrial standards, and to industry to access new knowledge and leverage on Use Cases for their developments. Specifically, European seed industry will strongly leverage on projects such as PHENET for the development of new phenotyping tools in support to the selection of future-proofed, agroecology adapted varieties. With PHENET, breeders will be able to phenotype more and gather a much larger data-set to enhance their genetic breeding predictions.
PHENET will contribute to a wider use of AI in research and enhance data-based research across Europe. PHENET will not only use AI as a plug-in technology but will develop AI solutions with various applications. AI will also be a lever for training as a new technology that must spread around RIs. We have already started to organise training around data management, information systems and semantic web so that every PhD and post-doc from PHENET will be trained to data science in support to his / her research topic. This will guarantee that the new generation of researcher is well equipped with data science skills.
PHENET will contribute to the development of innovative companies, both directly with the involved partner companies and indirectly through companies interacting with PHENET partners through national and EU projects, initiatives, living labs etc. For instance, PHENET will improve services around treatments of satellite imaging with a higher spatial distribution, thereby considerably facilitating its access to phenotyping market, but also to the "precision agriculture" and "sustainable agriculture". More generally, PHENET will demonstrate that SME operating close to farmers / foresters decisions and innovation can develop as main actors of transition of agroecosystems e.g. towards carbon sequestration or precision farming/forestry applications. An innovation Hub (WP6) will connect the academic stakeholders and industry as well as practitioners
PHENET ensures a long-lasting relationship between industry and RI, which allows RIs to benefit from industrial standards, and to industry to access new knowledge and leverage on Use Cases for their developments. Specifically, European seed industry will strongly leverage on projects such as PHENET for the development of new phenotyping tools in support to the selection of future-proofed, agroecology adapted varieties. With PHENET, breeders will be able to phenotype more and gather a much larger data-set to enhance their genetic breeding predictions.
PHENET will contribute to a wider use of AI in research and enhance data-based research across Europe. PHENET will not only use AI as a plug-in technology but will develop AI solutions with various applications. AI will also be a lever for training as a new technology that must spread around RIs. We have already started to organise training around data management, information systems and semantic web so that every PhD and post-doc from PHENET will be trained to data science in support to his / her research topic. This will guarantee that the new generation of researcher is well equipped with data science skills.