Agriculture and Land Use
Development of the GreenREFORM sub-model for agricultural production and pollution of air and water is described below.
Agricultural production of animal and plant products accounts for over 20 pct. of total greenhouse gas emissions in Denmark. It is therefore imperative that GreenREFORM describes agricultural production at a relatively high level of detail.
Land use is another source of greenhouse gas emissions. LULUCF (Land Use, Land Use Change and Forestry), as it is often labeled, is responsible for approximately six pct. of Danish emissions. Agricultural emissions and emissions from land use are closely related, since agriculture uses approximately two thirds of total Danish land area. That being said, modelling the net emissions from Danish forests presents a number of separate challenges.
GreenREFORM will therefore have one model for agricultural production and emissions and another model for emissions related to land use.
The models are developed by Cecilie Jørgensen and Simon Christiansen from Aarhus University. DREAM’s model group assists the development with technical knowledge and in with the integration of the two models within the GreenREFORM model.
The agriculture sub-model is described in the following paper. Separate papers will be published during the fall of 2020, which describe the properties of the agriculture sub-model and the modelling of LULUCF respectively.
Link to about agriculture sub-model
The following gives a general introduction to the agriculture sub-model.
Agricultural Production Structure
The goal of the agricultural sub-model is to describe the choices made by agricultural producers and the associated greenhouse gas emissions and other environmental effects, such as water pollution from the release of nitrogen and phosphorus and air pollution from the release of ammonia.
There are three main challenges to modelling emissions from agriculture. The first challenge is that agricultural production is not sufficiently disaggregated in the Danish National Accounts to adequately model the environmental and climate effects. The second challenge is that agriculture has a different production structure than other sectors in the economy. The third challenge is that a significant share of emissions from agricultural production comes not from fuel consumption, but instead from the digestive process of livestock, the handling and storage of manure and the use of fertilizers in crop production.
The agricultural sub-model solves these three problems. With inspiration from the international literature as well as the existing Danish agro-economic model ESMERALDA, production functions have been constructed for animal and plant production, reflecting the actual production structure in the agricultural sector. Land use is for example included as an input in plant production. The production structure used allows for modelling of emissions from agriculture (energy-related and non-energy-related) in direct relation to the inputs, which give rise to emissions. Emissions from the use of fertilizers on fields are, for example, directly linked to the input of manure and artificial fertilizers.
Link to the documentation af ESMERALDA (in Danish)
Link to calculations made on the ESMARALDA model (in Danish)
Plant and animal production is divided into 12 different production subsectors, which combined correspond with the agricultural sector in the National Accounts. The production subsectors supply inputs to each other as well as agricultural goods to the rest of the economy.
Development of the agriculture sub-model has initially been focused on adequately modelling the emission of greenhouse gases from agriculture. The plan, however, is that the sub-model will also be able to describe a range of other environmental effects from agricultural production. The goal is for the model to explain the agricultural emission of:
- Greenhouse gases
- Particulate pollution
- Nitrogen (see also description below)
Technological Abatement Methods
The agricultural production function includes a number of methods by which producers can abate emissions. Agricultural producers can, for example, reduce their use of nitrogen fertilizers, which reduces the emission of greenhouse gases and nitro
There are also technologies that they can use, which are not implicitly included in the production function. Technical data about the abatement possibilities in the agricultural sector is therefore used to give a clearer picture of how the environmental and climate effect of agricultural production can be reduced.
The Danish Energy Agency’s Energy and Climate Outlook as well as Dubgaard and Ståhl’s abatement possibility catalogue is used for this purpose. These abatement methods are generally modelled in the same way as described in the paragraph about technological abatement possibilities. More detailed modelling will in some cases be possible.
Dubgaard and Ståhl’s abatement possibility catalogue (in Danish)
Modelling the Geographical Distribution of Water Pollution
The environmental impact of nitrogen and phosphorus emissions from agriculture depends on the geographical location of the emissions, due to the fact that wetlands are particularly susceptible to the emission of these pollutants. This presents a challenge, since the GreenREFORM model does not currently include a geographical dimension.
A method of implementing the geographical distribution of agricultural emissions, in a way that is consistent with the rest of the model, is therefore currently being developed.
Modelling Land Use
Land use is important for agricultural production, although land area is not exclusively used by the agricultural sector. Danish land area is also used for housing, industrial facilities, infrastructure, forests, grasslands, and wetlands. There are different emission-levels related to these land uses as well as changing from one land use to another.
The distribution of land area by use is assumed to be exogenous, which implies that it can be controlled through regulation and planning. The model describes emissions from land use as the result of this distribution as well as changes to the distribution over time.
The calculation of emissions from forests is complicated, due to the fact that trees differ in their absorption of CO2 from the atmosphere, while they also release CO2 during decomposition. Net absorption of CO2 depends on a number of factors, including the distribution of tree species as well as the age of the individual trees. To deal with this, a dynamic model is being developed, which uses biological information about the CO2-absorption from trees and relates this information to the stock of trees in Denmark. This allows for calculation of the yearly CO2-effect when, for example, land area is converted to forest.
A paper will be published during 2020, which describes the modelling of emissions from land use.