Influence of Ammonia Emissions on Aerosol Formation in Northern and Central Europe

Abstract

High concentrations of particles pose a threat to human health and the environment. In this study the influence of ammonia (NH₃) emissions on aerosol concentration in central Europe is investigated. Depending on crop growth, temperature and local legislation individual temporal profiles for fertilizer and manure application are calculated for each model grid cell of the SMOKE-EU emission model. The emission data was used as input for the CMAQ chemical transport model. Comparisons to EMEP observations indicate that the new ammonia emission module leads to a better agreement of modeled and observed concentrations. The model was used then to assess the influence of emission changes. It was found that a reduction of ammonia emissions by 50 % lead to a 24 % reduction of total PM_2.5 concentrations in the model domain during winter, mainly driven by reduced formation of ammonium nitrate.

Questions and Answers

Questioner Name: Anthony Dore

Q: Your model has a different temporal ammonia emission profile for each grid square. How well can you capture different regional farming practices and national legislation on the timing of application of fertilizer and manure?

A: Regional farming practices are only taken into account through the crop growth model and its dependency on the meteorological conditions. The results of the model study show that the application of manure and fertilizer takes place earlier in the year if the regional climate is warmer, which is e.g. visible if comparing southern France and Denmark. Additionally the first and last day of fertilizer application and manure application per country if mentioned in the legislation have been adopted to capture the legislations. In countries with a federal system the closed periods, adopted from a study of the European commission on the variation of manure N efficiency throughout Europe (EC 2009), differ from region to region. The closed periods used in this study represent a simplified version of reality as they don’t differentiate between regional legislations (EC 2009).

Questioner Name: Wouter Lefebvre

Q: Can you specify the decrease in meat consumption in the RCAP scenario?

A: To calculate the decrease in NH₃ arising from the consumption change, the actual consumption of ruminant, pork and poultry meat (including eggs) per country has been compared with the consumption in case of a healthy diet. The data of meat consumption per capita has been adopted from the statistic division of the food and agriculture organization (FAOSTAT 2014) for all countries of the European Union. The per capita consumption of ruminant meat consists of food commodities of cattle, sheep and goat meat in 2008. The difference of the actual consumed amount and the recommended amount has been calculated for the three categories, so that the percentage of reduction per species and country could be applied on the gridded livestock inventories of the Food and Agriculture Organization (FAO) (FAO 2007).

E.g. for Poland the percentage of ruminants per grid cell has been decreased by 25 % the percentage of pork by 93 % and the percentage of poultry by 43 %. For the Netherlands the decrease in ruminants is higher (79 %), while pork decreases by 89 % and poultry by 60 % per grid cell. Other sectors as manure application have not been affected by this decrease in animal density. As the recommendation assumes no change in the intake of milk products, only cattle and no change in the amount of dairy cows has been considered in the calculations. A change in the export of animal products has not been considered either, assuming that other regions in the world with an increasing consumption will be able to meet their own market demands in the future (FAOSTAT 2014; FAO 2007).

Questioner Name: Sebnem Aksoyoglu

Q: We did a similar study using the TNO emission inventory. The modelled and measured NH₃ temporal profile were different. Measurements had higher peaks in summer than in spring. Did you have a similar case in your domain?

A: Our emission model is based on a statistical approach. Thus, we cannot reproduce single emission peaks. However, the main fertilization peak which is assumed at the beginning of Mach in the ‘standard temporal profiles’ occurs much later in our emission model. For the central European domain shown in this work it is, on average, at the beginning of May (Fig. 5.1). In colder regions it can be as late as June.

Questioner Name: Limei Ran

Q: How many crops are modeled in the crop growth model?

A: The crop growth parameters used are based on the four crop types: spring, barley, winter wheat, sugar beets, and grass crops.

Q: What is the main cause of the peak in NH₃ emission in the fall?

A: The main cause for the peak in NH₃ emissions in the fall is the dumping of surplus manure on fields due. This effect is assumed to be a reaction on a European law forcing the farmers to have enough capacity of manure storage in relation to the amounts of animals for the whole winter.

Questioner Name: P.J.H. Builtjes

Q: How big is the largest difference between the EDGAR and the EMEP files?

A: Total European ammonia emissions are quite similar in both emission inventories. However, EDGAR has a better spatial resolution and distinguishes between emissions from agricultural activity and animal husbandry. On the other hand, we found a few unexplained features in the EDGAR inventory. For example, there were extremely high emissions from agriculture in Luxembourg which we deemed unrealistically high. Thus, we used the national totals from EMEP to adjust the EDGAR inventory.