National Scale Land-use and Transport Modelling: the Mars Austria Model
G Emberger, P Pfaffenbichler, R Haller, R Koelbl, Vienna University of Technology, AT
A land-use/transport model with nation-wide scope is set up and validated for Austria. The paper discusses appropriateness of the model for rural regions and generality issues. Forecasting results are juxtaposed with traditional transport forecasts.
Integrated modelling of transport and land-use is a theoretically and methodologically sound approach implemented in several operational land-use/transport interaction (LUTI) models. However, its application has to date been mainly limited to urban regions. While this is understandable?many related problems are most apparent in urban areas?, there is no fundamental theoretical or empirical reason to neglect rural areas. Moreover, most operational urban models appear as relatively custom-tailored implementations of generic modelling environments. The resulting questions of generality, including e.g. issues of transferability to other areas, have rarely been studied in depth. We expect that the implementation of a LUTI model on national scale can help to address these issues.
This paper presents experiences made in the setup of a national land-use/transportation interaction model for Austria (?MARS Austria?). The model covers the entity of Austria and 121 fairly diverse model zones. Despite this aggregate layout, even relatively small urban agglomerations are differentiated into core and hinterland areas. At the current stage, data from the 1991 and 2001 population are used, but we intend to extend the time series to 1971/81 for extended model testing.
Work progress currently does not yet permit to present concluding results in relation to the research questions outlined above, but first tentative conclusions are drawn for the MARS model. We identify two concrete conceptual shortcomings of the model for national scale applications. First, migrations flows are treated in a too aggregate way (?pooling? of migrants over the case study) and, second, the costs of migration are neglected in the model whereas they play a central role in migration theory. Even though the model results are not satisfying at the moment when compared to observed development, we conclude that implementing LUTI models higher spatial scales does have the potential to raise their generality and theoretical underpinning, which may in the end also benefit urban models.
Association for European Transport