Within the Swiss living lab, On Farm Experiments (OFE) are spread on the “Plateau” region which host 57% of the Swiss’s arable lands located between 600 to 800m asl., more especially on Fribourg and Vaud cantons. Among the participating farmers, two main topics of interest have emerged: efficient crop nutrition and sustainable weed management, both considered in close interaction with the cross-cutting issue of crop establishment. In relation to these topics, farmers defined their own objectives during a co-construction workshop held in February 2025, with clearly defined targets to be achieved by the end of the project in 2027. During the 2025 field season, OFEs were implemented progressively to test a range of techniques tailored to the specific conditions of each farm.
“Nature, to be commanded, must be obeyed” said Francis Bacon
Behind this quote F. Bacon emphasise the fact that we must consider the main physical and biotic laws of nature and the effects of their interactions on the agroecosystems. And it is precisely with what the farmers have to deal with every day. In order to approximate the combine effect of soil, climate and cropping system on nitrogen efficiency and weed harmfulness we decided with the Swiss farmers to use two models:
- Syst'N: This tool developed by “RMT Bouclage” between 2007 and 2019 in France aims to estimate the average losses of nitrogen in all their forms at cropping-system scale. It has been tested in the Swiss context (with local soil and meteorological data and real farming practices) on 4 farms of the SIMONE’s project. Now, we compare the outputs with our filed monitoring to see if the modelling is consistent or not.
- COMBHERPIC: This model assesses the effect of innovative cropping systems reducing herbicide use on weed harmfulness indicators (e.g. weed biomass, weed-related yield loss, seed replenishment in the soil). 3 farms of the SIMONE’s project are simulating their cropping system with this model. We will compare its outputs to field observation that we performed on the Swiss OFEs in 2025 and beyond.
For then on, these models after their validation in the Swiss context will be used to help farmers: For Syst’N to visualise the nitrogen dynamics in their fields in order to better understand if this resource is available or not for the crop in a sufficient quantity at the right time.
Syst'N output of the nitrogen dynamic throughout the cropping succession
Syst’N will help to figure out where are the main leaks on average and adapt the strategy to have an efficient crop rotation and practices that allow storage and release of nitrogen according to the precise needs of the crop, as presented in the graphs above. In Switzerland we have two kinds of opposite context: farms without cattle that don’t have any nitrogen resources on their farms like manure or slurry. Their strategy is an efficient use of nitrogen inputs mainly under mineral forms. And the seconds are farms with cattle or pigs where nitrogen is not a limiting resource. They want to apply it at the right time to avoid leaching and anticipate the release of compounds directly accessible to crops according to the weather. During the last field season, we performed trials on sugar beets and maize to understand the nitrogen release of organic deposit.
Regarding sustainable weed management, the temptation to simply substitute chemical weeding for mechanical weeding is strong… But in the field, reality is quite different. Farmers needs to think about the short and the long-term effect of their practices at the same time: the combination of strategy and tactic! By using Combherpic, we want to visualize the long-term effect of cropping systems on weed development and harmfulness and choose those that minimise the use of herbicides. The selected cropping systems will then be tested in OFEs to identify the conditions for their implementation and their concrete results from 2026 onwards.
Swiss farmers are actively engaged in the agroecological transition and are supported in this effort through direct payments aimed at encouraging practices such as reduced soil tillage, lower input use, and mechanical weed control. However, they must prioritise the issues they face and identify the key limiting factors associated with each strategy under consideration. In this context, modelling tools can provide valuable support by helping farmers assess trade-offs and make informed, evidence-based decisions.
More info?
Please contact Sébastien Galland, Technical‑Scientific Collaborator in Arable Crops Weed Research at Agroscope.
sebastien.galland@agroscope.admin.ch/