With an average of ten million visitors each year, Great Smoky Mountains National Park experiences many visitors who get lost in the park because of inaccurate Location-Based Services (LBS) or outdated maps. Park-issued maps are available at visitor centers, but many visitors rely on navigation assistance from their mobile phones or other GPS devices. This is a major problem because the map data used by many LBS providers does not reflect authoritative data - causing many visitors to follow poor navigation directions.
Great Smoky Mountains National Park is a 276,344 acre park located on the border of North Carolina and Tennessee.
In Comes OpenStreetMap
When the NPMap team announced several initiatives built around OpenStreetMap (OSM), the Smokies quickly saw an opportunity to get accurate maps and data into the hands of an increasingly technologically-savvy public. All National Parks must make their authoritative map data available through the NPS Data Store, but few park visitors possess the knowledge or the software required to manipulate the data or load it onto their personal electronic device. OpenStreetMap, on the other hand, is built specifically for non-technical users, and it gives them the ability to edit and interact with the data in a platform of their choosing.
During the summer of 2013, the Smokies approached the OSM community through the Imports-US mailing list with their proposal for uploading the park's authoritative data to the OSM database. The park engaged with the community because any significant addition or alteration of OSM data usually requires a discussion to:
- Ensure that the data is a good fit for the OSM model
- Reach agreement with the community on how the data will be entered into the database
Initially, the park struggled with some of the feedback received during the import discussion process. Like any federal agency, the Park Service must follow strict rules when publishing map data in any format. For example, federal regulations prohibit a government entity from publishing the name of a geographic location unless that name has been approved by the United States Geological Survey's Board of Geographic Names. Given that the OSM database is based on “crowd-sourced” input, many OSM members felt that the park's official names for many features were not inline with local, common, or well-known geographic names that members of the public might recognize. Ultimately, consensus was achieved when the park agreed to add an “Official Name” tag to data it was contributing to OSM.
Experienced OSM mappers were also able to help the Smokies work through some challenging “tagging” questions, such as how to tag buildings, deal with TIGER data, and document National Park Service tagging guidelines on the OSM Wiki. The value of this feedback emphasizes the importance of reviewing import proposals with the OSM community.
Following agreement on how the park would proceed with its OSM import, the park converted its authoritative base data into shapefile format - allowing it to be imported via the Java OpenStreetMap (JOSM) application. For the many features that already existed in the OSM database, park mappers simply conflated (replaced old with new) geometries, such as road and trails, with the more accurate park data, then attributed the tags according the NPS tag wiki. They completed a portion of the park and made the OSM database available for public review before completing the import and posting all of the data to the production OSM database.
A Lesson Learned
One important lesson was learned when the park ran into significant data validation issues related to the amount of data they were uploading at one time. This was related to the length of time it took to edit such a large geographic footprint. During the few weeks that the park was uploading data, many changes were still occurring to the “live” OSM database, which led to an increase in time spent re-validating data. In addition, the park experienced the ubiquitous computer crash during an upload of several hundred thousand objects (each node edit is counted as an object), resulting in a hair-pulling, all-night-long correction of a duplicate upload to the OSM database.
Because of this experience at the Smokies, the NPMap team is now discouraging bulk uploading and editing of NPS data in the OSM database - at least for the time being. The team now recommends that no more than a few features be edited or uploaded at once. The team is investigating ways to simplify the process of getting data from internal NPS databases into OSM.
The entire process, including import discussion and editing, took about 400 hours for the park to complete. The NPMap team and Smokies staff spent much of that time documenting tagging guidelines on the OSM wiki. These guidelines are meant to make it much easier for other National Parks to contribute their data to OSM. The NPMap team expects that parks following those guidelines would be able to complete a park-wide update of their data in OSM in 40 hours or less.
The park is well-mapped in OpenStreetMap now.
The Smokies update to OSM has been complete for over a year now. Present in the OSM database, available for public use, is what the National Park Service considers essential base data for visitor services: transportation networks, points of interest, and visitor services infrastructure. For example, anyone can consume OSM data from the Smokies and make a map showing the location of visitor centers, ranger stations, campground, and hiking trails - no specialized skills required!
The continued maintenance of the Smokies data in OSM is a dynamic process. Through the OpenStreetMap editor and specialized tools like MapRoulette, other OSM mappers continue to improve upon the park's contribution by, for example, updating feature that have too many nodes or making cartographic improvements to feature geometry.
The National Park Service does not monitor or endorse commercial mapping applications, but the Smokies has noticed that many smartphone and web-based applications are using the park's OSM data. This demonstrates that OpenStreetMap is a great way for the Smokies to make their authoritative data available for public usage in an open way - with no restriction on how the data is used or distributed.