Talks and Poster Presentations (with Proceedings-Entry):
M. Schmidt, W. Jörg, G. Gartner:
"Challenges in creating web base maps from distributed datasets";
Talk: 26th International Cartographic Conference ICC,
Dresden;
08-25-2013
- 08-30-2013; in: "Proceedings of the 26th International Cartographic Conference",
(2013),
ISBN: 978-1-907075-06-3;
1 pages.
English abstract:
Web base maps distributed as raster tiles and published as web services became very popular over the last seven years. While the first and still common base maps are tiles provided by large corporate web mapping services such as Google Maps or Bing Maps, base maps nowadays come in a broad range of types, content, styles and licenses, facilitated by the availability of open data (e.g. OpenStreetMap) as well as software and hardware specialized on the creation and distribution of base maps.
In Austria, an initiative by the Austrian province governments is working on a base map for the state of Austria, which uses the data created by the Austrian administrations. This paper deals with the cartographic (and partly also technical) challenges in creating such a base map from distributed, administrative datasets.
The administrative structure of Austria (state, provinces, municipalities) is reflected in the way geodata are created, managed and updated. Federal provinces are required by law to create and manage geo data in certain fields. Some of the provinces already produce province-specific renderings of their data. However, until now these services are restricted to each province territory; data and styles between the province versions are not mandatorily harmonized. Also, some of the services use different projections and zoom levels than Google Maps and similar services. The new project is aiming at offering the locally created geodata in a joint view for the user, which is implemented as a WMTS in Web Mercator Auxiliary Sphere (EPSG:3857), to ensure compatibility to existing web base maps. The final product (tile cache) is published under the Open Government Data license CC-BY 3.0 AT - a widely known and easily understandable license with clear conditions for the users, as well as the data contributing institutions.
Apart from the technical implementation, there are two main challenges from a cartographic point of view: data harmonization and generalization. Even though there are major efforts for standards in data creation (e.g. GIP - an intermodal transportation graph of the Austrian administrations and centralized transportation infrastructure organizations), a standardized visualization revealed province-specific differences in the data sets. Harmonization is also needed in terms of naming conventions, data formats and projections to ensure fast rendering. The need for generalization becomes apparent when considering the scale range of base maps; in this case approximately 1:400 to 1:3,000,000. However, the data from the contributors are usually created in one defined level of detail, fitting a certain scale. To ensure a legible and aesthetically pleasing result on all the other zoom levels, generalization is needed. Of course, this is not a new topic in cartography. However, it is still a challenge to implement generalization in an automated visualization environment, which allows incremental updates. Similar to harmonization, this is not just a technical issue; it also identifies requirements for the data. For example, in order to make a useful selection of features of a river network for different zoom levels, the initial geodatabase must include relevant metadata for this purpose. This is not available for some of the data sets yet. In the initial version, two zoom levels will be manually harmonized and generalized for all provinces. All other zoom levels will only use basic generalization methods - including a careful selection of features for each zoom level, appropriate symbolization and smoothing. While the first prototype does not meet all cartographic requirements yet, the project is working on methods for improving the cartographic quality of the final product in the next several months.
The experiences from this project will be helpful for other organizations aiming at creating and delivering web base maps from distributed datasets on the basis of an open standard.
Keywords:
basemap, GIS infrastructure, open government data, web mapping
Created from the Publication Database of the Vienna University of Technology.