MAELIA is a socio-agro-ecological modeling platform developed by INRAE since 2011. MAELIA combines agent-based and spatially explicit approaches to simulate interactions between biophysical processes (crops, soils, livestock, hydrology) and farmers’ decision-making at multiple scales: field, farm and region. MAELIA is an open-source integrated assessment and modelling tool implemented in the agent-based modelling platform GAMA. It combines agent-based decision rules with mechanistic but parsimonious process models, in a spatially explicit GIS framework with daily time steps. This design allows dynamic interactions between agricultural operations, weather events and biophysical processes, and the emergence of systemic properties from plot to landscape scale. MAELIA enables the integrated assessment of environmental, economic and social impacts of changes in agricultural activities at the watershed and field scale, running simulations over multiple years under various climate change and policy scenarios. It has been mobilised in over twenty research projects and supports scenario-based prospective approaches at territorial scale, including food system planning, agri-environmental policy evaluation, and bioeconomy strategies.

Training objectives:

• Describe the challenges and relevance of territorial approaches to guide agroecological and bioeconomic transitions at the territorial scale.
• Use the core modules of a spatially explicit modeling platform for agricultural territories.
• Analyze the impacts of transition scenarios using a multi-criteria approach.

test new resource

Feedback on teaching experience

...

The Scaling Scan is a user-friendly tool to explore what is required to scale an innovation in a specific context, the implications this has for project management and collaborations the potential trade-offs on the environment as well as the social dynamics.

The Ladder of Life is a focus group tool conducted with poor women and men. It is meant to explore their understandings and interpretations of the different wellbeing groups and the poverty trend in their community, and the key factors and processes seen to shape these dynamics

The FARMSIM model is one of the most comprehensive integrated farm models and is used to predict the probable impacts of different farming interventions on nutritional and financial well-being at the household level.

It is a GIS framework that simulates the bio-physical processes and activities along livestock supply chains under a life cycle assessment approach. The aim of GLEAM is to quantify production and use of natural resources in the livestock sector and to identify environmental impacts of livestock in order to contribute to the assessment of adaptation and mitigation scenarios to move towards a more sustainable livestock sector.

FarmDESIGN is a static and exploratory whole-farm model that quantifies farm productive, economic and environmental performance on annual basis.

The Global Yield Gap Atlas (GYGA) is an agronomic database platform with local and global relevance to support your decision on yield optimisation. The database offers high-quality and locally relevant data on yield gap and potential, water productivity, nutrient requirement, and other agronomic data such as weather, soil, and crop management system. GYGA now has coverage of up to 13 food crops across 70 countries and six continents with the possibility of expansion.

RHoMIS was designed to improve the process of gathering information from farming households in the rural developing world. Household surveys are very widely carried out, but the data is rarely comparable, and so the opportunities for learning between individual projects are limited. For these reasons, the tool balances standardisation with flexibility.

CLEANED is a rapid environmental impact assessment tool that allows users to explore multiple impacts of developing livestock value chains using forecasting methods. It models the foreseen impact of changes in the livestock production systems and value chains along several pathways on land use, productivity, economics, water impacts, greenhouse gas emissions, and soil health in order to support decision making with multiple objectives.

This document presents integrated modeling methods for assessing the response of agricultural systems to scenarios combining innovation and policy. The approach considers the farm system as a key level with its interactions with dow-level (the fields) and up -levels (the landcape and markets). Farm diversity in the landscape is modelled with typologies. Farm and field behaviours and performances are simulated respectivelly with farm models and crop models and a set of indicators. The document also stress the importance of participatory modeling, involving local stakeholders in the design of scenarios and selection of assessment indicators. Finally, concrete case studies illustrate how these methodological frameworks help predict the adoption of new practices and the impact of policies while measuring socio-economic and environmental trade-offs.

Feedback on teaching experience

This course was given during 10 years (2008 to 2018) to Master Students at l'institut Agro Montpellier. I will be happy to collaborate with colleagues who want to develop similar course (in english or french).

Manual to develop farm household typologies using R based on Principal Component Analysis (PCA) and Hierarchical Clustering (HC)

This book originates from the observation that codesign processes in Living Labs often favour local knowledge over academic knowledge and are rarely informed by systematically collected data. While local knowledge is invaluable, relying exclusively on it can inadvertently limit the creativity and problem-solving potential of the codesign process. In practice, this can lead to innovations that replicate existing practices rather than exploring novel solutions to key bottlenecks in the systems under study.
The book presents a set of reproducible, data-driven workflows designed to enhance the codesign process within Living Labs. These workflows integrate KoBoToolbox, an open-access digital platform for participatory data collection, with R, a powerful open-source environment for statistical computing and visualization. Using Mbire District, Zimbabwe, as a case study, the book provides clear, step-by-step guidance for researchers and practitioners on how data can be used to enrich codesign processes. It was compiled using the bookdown package, enabling the seamless integration of R code, analytical results, and narrative into a single, fully reproducible document.
The book includes a collection of XLSForms (inputs to KoBoToolbox) and well-documented R scripts developed for the specific context of Mbire District during the implementation of the CGIAR initiative ‘Transformational Agroecology across Food, Land, and Water Systems’. You can download the PDF version of the book, or use the web version.

Ce memento est destiné aux étudiants en Sciences Agronomiques pour leur permettre d’acquérir et d’utiliser les concepts de base nécessaires à la compréhension des enseignements (cours, TD, rapports) d’Agronomie Systémique (AS).
Comme toute discipline scientifique l’AS peut se définir par ses concepts, ses méthodes et outils et les différentes postures avec lesquelles on les met en œuvre sur des objets et à des échelles spécifiques. Le document est organisé autour de ces différentes facettes de la discipline. La démarche systémique utilisée est basée sur l’analyse des relations structure-fonctions-flux dans les systèmes complexes dynamiques décrite dans un autre document d’enseignement (Wery, 2017 ; Démarche de conceptualisation d’un problème en système en Agriculture).

Les concepts et méthodes présentés peuvent s’appliquer à toutes les formes d’agriculture (intensive en intrants de synthèse, intensive en travail, agroécologique, biologique, de conservation, agroforesterie…), dans tous les secteurs de production (grandes cultures, maraichage, arboriculture, viticulture, fourrages…) et dans toutes les régions (tempérée, méditerranéenne, tropicale…).

Les mots entre guillemets sont la traduction anglaise des termes.

A well-calibrated and evaluated GROPGRO module of the Decision Support System for Agro-technological Transfer (DSSAT) was used to simulate productivity of soybean in northern Nigeria under climate change. Both historical (1990–2019) and projected climate scenarios from 5 general circulation models (GCMs) under two representative
concentration pathways (RCP 4.5 and RCP 8.5) in the mid-century (2040–2069) and end of the century (2070–2099) periods were used. Depending on climate scenario, the minimum temperature is expected to rise by 1.7–4.4oC at Kano in the Sudan savanna (SS) agroecological zone (AEZ) and 1.4–4.0oC at Zaria in the northern Guinea savanna
(NGS) AEZ, while maximum temperatures are projected to increase by 1.7–4.1oC in the SS and 1.3–3.6oC in the NGS. Seasonal average rainfall will increase by 4.8–14.5% in the SS and decrease by 2.6–3.8% in the NGS, relative to the baseline climate. The model predicted delaying trends for days to flowering and maturity for both varieties
in all climate scenarios in the two AEZs.