The “C and N Models Inter-comparison and Improvement to assess management options for GHG mitigation in agrosystems worldwide” (CN-MIP) addresses theme 1, topic 1 of the FACCE-JPI 2013 call. Our project will coordinate international development, evaluation and inter-comparison of agricultural process-based models to reduce uncertainty in estimating greenhouse gas emissions from crops, grassland and livestock systems. The project will focus on improving the simulation of management options to enable evaluation of credible mitigation strategies adapted to diverse agrosystems under different climatic conditions. CN-MIP responds to the priority of the core theme 5 "Mitigation of Climate Change" of the FACCE-JPI strategic research agenda, to improve the greenhouse gas (GHG) inventory methods, particularly the "certified" modellingTIER3 modelling approach for quantifying emissions and the effects of mitigation options. The project also supports initiatives outlined in the Global Research Alliance (GRA) on Agricultural Greenhouse Gases, which aim to improve measurement methodology and modelling, as well as inventory of GHG emissions and C sequestration in soils. The consortium comprises eleven partners: INRA (France), University of Aberdeen (UK), Helmholt-Zentrum Postam (GER), University of Florence (IT), CRA-Consiglio per la Ricerca in Agricoltura (IT), University of Milan (It), University of Sassari (IT), New Zealand Institute for Plant and Food Research (NZ), Colorado State University (USA), Woods Hole Research Center (USA), Queensland University of Technology (AU). The proposing partners are experienced modelers and experimentalists, already involved in internationally funded projects on measuring and modelling of greenhouse gas emissions,soil carbon sequestration, and reactive nitrogen, for a variety of agricultural conditions (annual crops, grasslands, tree crops) under temperate, Mediterranean and tropical conditions (GRA CN, Livestock and Cropland groups, AgMIP, MACSUR, Reactive N RCN, NANORP, etc.). This network will provide connections and sharing of models, modelling protocols and datasets, but also the necessary interactions with stakeholders. The project will be undertaken from January 2014 to December 2016, in 4 work packages (i) Definition of model data requirements, selection of process-based CN models (i.e. DNDC, DNDC mobile, DSSAT, Roth C, DayCent, PaSim, STICS, APSIM, EPIC, CN-SIM), selection of appropriate databases; (ii) development of common protocols for modelling and model inter-comparison; (iii) identification and testing of mitigation options, improvement of models for coverage, predictive capability and reliability; (iv) dissemination and training. Deliverables will be guidelines for the selection of database and the simulation of mitigation options, evaluation of uncalibrated and calibrated model performances for an array of GHG emission outputs, improved model tools, peer-reviewed research papers, communication and reports to policy makers and stakeholders, and training sessions for students and scientists.