Nitrogen oxide and dioxide emissions (NO NO2=NOx) by international shipping account for approximately 9-15% of NOx global emission totals. As NOx participates in the catalytic production of tropospheric ozone, global chemistry climate models cannot neglect ship induced NOx in their forecast of future atmospheric development. Both the knowledge of the NOx emission flux as function of time and place and an understanding of the chemical processes transforming NOx in the atmosphere are needed. Thiswork addresses both aspects.The first part of this thesis analyses the method used by global models to include NOx ship emissions. For this purpose, the dispersion and chemical conversion of emissions in the near-field of a single ship are studied with two different modelling approaches. While both techniques use the same photochemical box model to solve the chemical equations, the dilution of the exhaust into the background air is different. One approach uses a Gaussian plume model and accounts for the expansion phase of a plume. The other one instantaneously disperses the emissions over a large gridbox, a technique commonly used by large scale models. The ozone change is overestimated by the global-model approach by up to a factor of three. One possibility to account for these sub-grid processes in global models is the use of effective emissions, i.e. actual emissions are changed and emissions of additional compounds like ozone are introduced in a way that they take sub-grid processes into account. It is shown for this case that the method is able to account for the neglect of sub-grid processes in global models for different emission times and emission strengths.In the second part of the thesis, an inventory of NOx emission from international shipping has been evaluated by comparing NO2 tropospheric columns derived from the satellite instruments GOME (January 1996 to June 2003), SCIAMACHY (January 2003 to February 2008), and GOME-2 (March 2007 to February 2008) to NO2 columns calculated with the atmospheric chemistry general circulation model ECHAM5/MESSy1 (January 2000 to October 2005). The data set from SCIAMACHY yields the first monthly analysis of shipinduced NO2 enhancements in the Indian Ocean. For both data and model consistently the tropospheric excess method was used to obtain mean NO2 columns over the shipping lane from India to Indonesia. In general, the model simulates the differences between the regions affected by ship pollution and ship free regions reasonably well. Therefore, it is concluded, that the NOx ship emission inventory used inthis study is a good approximation of NOx ship emissions inthe Indian Ocean for the years 2002 to 2007. It assumes that around 6 Tg(N)/yr are emitted by international shipping globally, resulting in 90 Gg(N)/yr in the region of interest when using Automated Mutual Assistance Vessel Rescue System (AMVER) as spatial proxy. The results do not support some previously published lower ship emissions estimates of 3-4 Tg(N)/yr globally.