Nitrous oxide is a potent greenhouse gas, with a global warming potential 298 times greater than that of CO2. In agricultural soils, N2O emissions are influenced by a large number of environmental characteristics and crop management techniques that are not systematically reported in experiments. Random Forest (RF) is a machine learning method that can handle missing data and ranks input variables on the basis of their importance. We aimed to predict N2O emission on the basis of local information, to rank environmental and crop management variables according to their influence on N2O emission, and to compare the performances of RI: with several regression models. RF outperformed the regression models for predictive purposes, and this approach led to the identification of three important input variables: N fertilization, type of crop, and experiment duration. This method could be used in the future for prediction of N2O emissions from local information. (c) 2013 Elsevier Ltd. All rights reserved.

Insight in the phosphorus (P) flows and P balances in the food chain is largely unknown at county scale in China, being the most appropriate spatial unit for nutrient management advice. Here, we examined changes in P flows in the food chain in a typical agricultural county (Quzhou) during 1980–2017, using substance flow analyses. Our results show that external P inputs to the county by feed import and fertilizer were 7 times greater in 2017 than in 1980, resulting in a 7-fold increase in P losses to the environment in the last 3 decades, with the biggest source being animal production. Phosphorus use efficiency decreased from 51% to 30% in crop production (PUEc) and from 32% to 11% in the whole food chain (PUEf), but increased from 4% to 7% in animal production (PUEa). A strong reduction in P inputs and thus increase in PUE can be achieved by balanced P fertilization, which is appropriate for Quzhou considering a current average adequate soil P status. Fertilizer P use can be reduced from 7276 tons yr⁻¹ to 1765 tons yr⁻¹ to equal P removal by crops. This change would increase P use efficiency for crops from 30% to 86% but it has a negligible effect on P losses to landfills and water bodies. Increasing the recycling of manure P from the current 43%–95% would reduce fertilizer P use by 17% and reduce P losses by 47%. A combination of reduced fertilizer P use and increased recycling of manure P would save fertilizer P by 93%, reduce P accumulation by 100% and P loss by 49%. The results indicate that increasing manure-recycling and decreasing fertilizer-application are key to achieving sustainable P use in the food chain, which can be achieved through coupling crop-livestock systems and crop-based nutrient management.