Herbicide management is at the heart of farmers' concerns. There is a strong need for references in a context where the number of active ingredients is decreasing, resistance is developing and climate change is having an impact on weed dynamics and intervention times.

Following on from the XpeGE 0 project, which tested systems without phytosanitary products, Herbi1&NoPhy is based on less risk-taking for farmers and supply chains: the use of herbicides is a low priority, well after agronomic levers - but not excluded - and the use of other phytosanitary products is circumvented by technical choices.

The evaluation of this new project should make it possible to measure the technical and economic consequences of very drastically reducing the use of phytosanitary products, with a view to adapting systems to climate change, based on the notion of diversity (species, varieties, tillage, sowing periods, etc.).

The aim of this experimentation and transfer project is to consolidate, build up and disseminate references on field crop and mixed farming systems, with a marked break in the use of phytosanitary products adapted to the context of climate change. It is based on a strong technical partnership between research, development, teaching and farmers.

The experimental sites (2 platforms and 10 observatories) will test annual cropping systems that represent a major break with the past, using all available agronomic, genetic, mechanical and biocontrol levers, with minimal use of herbicides and the use of other products only when absolutely necessary.

The aim of the project is to create benchmarks for low-use phytosanitary practices. Estimating the technico-economic impact, the consequences in terms of equipment investment or working time, and assessing the acceptability of these systems to farmers through controlled observatories are the results expected from the project to help farmers adopt these systems.

The tools currently used to calculate nitrogen fertilization for winter rapeseed are based on the projected nitrogen balance method. However, this method is regularly called into question due to the difficulty of estimating the yield target and the nitrogen supply by the soil at the time the balance is opened. For this reason, a method has recently been developed for integral management of nitrogen fertilization in straw cereals, based on remote sensing monitoring of the crop's nitrogen nutrition status in spring. This method uses an indicator that provides direct, real-time information on the crop's nitrogen requirements: the nitrogen nutrition index (NNI). The main challenge of the project is to develop this method for rapeseed.

• Definition of a threshold trajectory for the nitrogen nutrition index during the period of interest (from winter emergence to flowering), which is not detrimental to yield.
• Creation of operational decision rules for dynamic management of nitrogen fertilization (maintaining the nutrition index above the threshold trajectory at all times during the period of interest), based on existing ecophysiological knowledge, expertise and the use of a crop model.
• Development of reliable tools for indirect estimation of the nitrogen nutrition index (remote sensing) to ensure the operationality of decision rules.

The method that will be developed in this project aims to offer a real time method for managing nitrogen fertilization, enabling the doses of nitrogen applied to be better adjusted to the crop's needs than the forecast balance method. The results of this project will have an innovative scope, as there is no tool using this method for winter rapeseed. The project will encourage the creation of projects in partnership with private companies in the sectors of agricultural cooperation and trade, remote sensing and the marketing of digital farm management solutions and crop management aids.

The project could make it possible to propose several levers:

• Improving the competitiveness of the industry: in broiler production, feed is the main expense item. However, with the rising cost of raw materials, customized feed could reduce the total quantity of nutrients distributed to the animals, thus significantly reducing feed costs.
• Protecting the environment by reducing nitrogen and phosphorus discharges, and taking into account the environmental impact of raw materials: adjusting nutritional intakes to strictly meet animal needs, and using highly digestible PRPMs, will enable proteins to be better valorized, thus reducing nitrogen and phosphorus excretions (pollutants) and limiting losses to the environment.
• Improving protein and nitrogen self-sufficiency: European poultry farming, and French poultry farming in particular, is dependent on soya supplies, most of which come from Brazil. However, by adjusting the protein content of feed on a daily basis to strictly meet the animals' needs, we can avoid situations where the protein intake (and therefore the use of soya) is too high in relation to the animals' actual requirements.

The overall aim of the project is to propose an innovative feeding strategy for broiler chickens to improve nutrient utilization efficiency. This "tailor-made" feeding strategy is based on the use of mathematical models to adjust ration composition on a daily basis. The operational objectives are :

• Develop a model for predicting nutritional requirements that can be adapted to production objectives for two types of broiler production (standard and label);
• Build robust models to predict the performance (growth and feed intake) of a batch of broilers, based on the previous days' performance, in order to adapt rations as closely as possible to the animals' needs;
• Provide a tool for multi-objective formulation of customized feeding strategies, adapted to logistical and operational constraints in the field;
• Provide a technical, economic and environmental assessment of the customized feeding strategies developed on the experimental station.

• Development of prediction models and feed formulation tools adapted to customized feeding.
• Distribution of a decision support tool for day-to-day adjustment of the ration to be distributed according to the performance of the current batch.