Specifying organic fertilizer composition in process-based models: overview of available data and sensitivity analysis with Biome-BGCMuSo

Organic fertilizers are widely used around the world, because they support the circular economy, sustainable agriculture, and improved soil quality, as well as carbon sequestration. State-of-the-art, process-based models can simulate the environmental impacts of organic fertilizer use and can address issues like the effect of fertilizer amount and type on crop production, soil fertility, soil organic matter accumulation, nitrate leaching, and greenhouse gas emission. However, the lack of information on the proper attribute settings for fertilizer inputs in the models hampers their application. In this study, the main goal was to support the setting of organic fertilizer attributes for process-based model applications. A comprehensive data collection was performed to gather organic fertilizer attributes that are relevant for the carbon and nitrogen cycle-related simulations. Based on the literature search, representative values are presented that can be instantly used in the models as generalized settings for several farmyard manure and slurry types. We also addressed the question of how fertilizer attribute-setting-related uncertainties propagate to the simulation outputs. We used the Biome-BGCMuSo biogeochemical model for that purpose with a maize monoculture simulation. The results indicate that manure type specific attribute setting is crucial for the nitrogen balance related model variables. For soil nitrous oxide efflux, improper composition settings can severely distort the simulation results. Sensitivity analysis suggested that dry matter content and organic nitrogen content are the two most important manure attributes that modellers must properly adjust. For slurry, the dry matter and ammonium content must be constrained for proper simulation results. The study supports crop and biogeochemical model setup with ready-to-use pragmatic information.