MAGGnet : Greenhouse gas emissions from agriculture

Background (completed research project)

Agriculture accounts for 12 - 13 % of total greenhouse gas emissions in Switzerland. Nitrous oxide and methane account for the majority of emissions. Carbon also plays an important role. The carbon content of the soil may increase or decrease considerably as a result of changing land use and management. Arable land can usually be assumed to contain fairly low levels of carbon. However, knowledge about the degree to which the carbon content is changing in agriculture and the underlying mechanisms is incomplete. In many cases, the CO2 emissions associated with soil carbon are also site-specific, meaning that measurement results cannot be extrapolated directly to other sites. Tillage, crop rotation and the type and quantity of fertilisers used, can help to reduce emissions and increase the carbon content of the soil. However, these measures may have beneficial and adverse knock-on effects, for example on yields. Long-term trials play a significant role since carbon contents can vary over a period of decades. In addition to the directly measured effects of land use and management, the data generated by long-term trials can be used to improve process-oriented models that in turn allow scenarios to be calculated, for example.

Aim

The aim of the research project was to compile data from long-term trials in Switzerland about the dynamics of the country’s carbon sources and sinks and to make the information available to an international database. Along with data from long-term trials carried out in other countries, these results will undergo a meta-analysis by coordinator Mark Liebig as part of the Croplands Greenhouse Gas Network (MAGGnet). This will make it possible to identify the major factors controlling emissions and the intensity of their impact. Data about the yield of individual crops will subsequently be included to evaluate the efficiency of various measures in terms of avoiding greenhouse gas emissions.

Results

Series of measurements of soil carbon, yields and associated parameters, such as weather data and soil density, from four long-term trials performed in Switzerland were compiled. The trials were three trials on arable land in Zurich-Reckenholz, Tänikon, Burgrain and a trial on grassland in Oensingen. In Zurich and Tänikon, grassland was turned over before the trial started, in Oensingen arable land was left to become fallow. A significant reduction in the content of soil carbon was measured in most of the processes used, even when farmyard manure was used and artificial grassland was part of the crop rotation. These changes can be explained in part by the change in land use. The extensive grassland also lost soil carbon, while intensive utilisation of the grassland led to an increase in the humus level of the soil compared with the arable land previously on the site.

Implication for research

Carbon content of the soil is still changing even after 60 years. This underlines the importance of long-term trials. Most long-term trials were not explicitly designed to study the carbon content of the soil when they were set up, and as a result the measurement series are incomplete. A more serious problem is that routine sampling was often only restricted to the topsoil. This is insufficient to generate a complete carbon balance because the subsoil, which is a more stable long-term storage medium, is also relevant for the assessment of longer-term changes. Moreover, reference samples are not available from all samples and years, making it difficult to produce more detailed interpretations based on new measurements of old samples (e.g. isotopes).

Implication for practice

The extensification of grassland and no-till farming are not effective ways of increasing the carbon content of soil. The transformation of grassland to arable land reduces the amount of carbon in the soil substantially. Declining soil carbon content could be a widespread phenomenon when farmyard manure is applied regularly and crop rotation patterns that include artificial grassland are used. Measures designed to increase the carbon content therefore always need to be reviewed carefully.

A model-based inventory system for soil carbon is currently being established in Switzerland. The soil carbon models are being tested and optimised using data from the long-term trials, making the results significant for climate reporting too.

Original title

Quantifying Greenhouse Gas Mitigation Effectiveness through the GRA Croplands Greenhouse Gas Network (MAGGnet)

Project leaders

• PD Dr. Jens Leifeld, Agroscope, Zurich
• Dr. Mark Liebig, U.S. Department of Agriculture, USDA, USA (coordinator)
• Dr. Alan Franzluebbers, U.S. Department of Agricuture, USDA, USA
• Dr. René Dechow, Johann Heinrich von Thünen Institute, Germany
• Prof. Gervasio Piñeiro, University of Buenos Aires, Argentina
• Prof. Kristina Regina, MTT Agrifood Research, Finland
• Prof. Pier Paolo Roggero, University of Sassari, Italy
• Dr. Yasuhito Shirato & Dr. Ayaka Kishimoto, National Institute for Agro-Environmental Sciences, Japan