Options to decrease N losses from our global food system

J.G. Conijn, J.J. Schroder, P.S. Bindraban

Research output: Contribution to conferenceConference paper


Food production causes losses of reactive nitrogen (N) to the detriment of the environment but the current
level of losses per unit food leaves room for improvement. Due to feedback mechanisms a comprehensive
analysis is needed and we developed a quantitative model of the whole food system to assess the effects of
improvement measures on the required amount of N fertilizer and resulting N losses as function of food
demand. For 2010 we calculate a total N loss from agricultural soils and ammonia volatilization of 172 Mt N
y-1 and an amount of 32 Mt N y-1 entering households in food items. This implies a N loss ratio of 5.4 kg N
lost per kg N purchased by households. Due to higher food demand and changed diet as projected for 2050,
the N loss ratio increases to almost 6.0 if equal N use efficiencies are used as in 2010 and the total N loss
amounts to 293 Mt N y-1. The effects of a number of improvement measures are explored, such as less
animal-based products in the human diet and reduced N loss from agricultural soils. Single measures can
reduce this ratio to as low as 3.8 but when all measures are combined, the ratio drops to 2.0 with a total N
loss of 84 Mt N y-1 without affecting the projected food demand for 2050. Our results clearly illustrate that
the effectiveness of measures cannot be realistically estimated without taking the whole system into account
and that the N loss ratio is a better indicator to estimate environmental impacts of N use than N use
Original languageEnglish
Number of pages4
Publication statusPublished - 2016
Event7th International Nitrogen Initiative 2016 - Melbourne, Australia
Duration: 4 Dec 20168 Dec 2016


Conference7th International Nitrogen Initiative 2016
Abbreviated titleINI2016

Fingerprint Dive into the research topics of 'Options to decrease N losses from our global food system'. Together they form a unique fingerprint.

Cite this