Data Sourcing

Urban Mapping has the ETL expertise required to source and create whatever data you need for your enterprise. These skills in custom sourcing high-value data allow Mapfluence to tackle inquiries from noise pollution to demographics and everything in between. We use them to curate our proprietary neighborhood boundaries and public transit datasets, and to maintain third-party demographics data available through our data catalog. All of this can be licensed as extracts or through our APIs. In addition, we source and create custom data sources for clients. In each of the examples below, the Urban Mapping data team proved we are the best in the business at sourcing custom data.

Noise Contours

When a real estate customer inquired about data to help determine noise pollution near airports, the Urban Mapping data acquisition team sprang into action. The best data they could find were reports conducted by airport authorities detailing noise exposure around runways. After identifying the correct departments within operating authorities and making numerous public record requests, the data acquisition team followed the paper trail and discovered a program administered by the US Federal Aviation Administration (FAA). After pouring over copious federal regulations, the team uncovered a well-defined data model. Whenever a new airport runway is constructed, federal law requires a comprehensive study of surrounding noise levels and mitigation plans. This is submitted to the FAA for approval, however, while the FAA acts as a public overseer, it only approves these reports. The substance of the reports is available only from the operating authority. Some airports had publicly posted the reports submitted to the FAA, others were available upon request, and some denied the public record request citing vague concerns about security.

When possible, Urban Mapping always seeks source data in a machine-readable format. However, because many airport operators engaged outside engineering firms to conduct the studies, source data was not always available. This meant ingesting data in a variety of media formats - digital images (PNG, JPG, PDF etc.), large five by seven foot posters, and sometimes even scanned photocopies of reports. The ingestion process is made up of several steps, and the underlying media drives each set of workflows. At a high level, this means digitizing, extracting or tracing contours, creating comprehensive metadata (to describe our process and explain the FAA requirements to the layperson) and testing the database by creating a variety of maps and spatial queries to better understand how the data behaves.

After defining the problem, acquiring raw data, ingesting appropriate formats, creating descriptive metadata, and testing the database, Urban Mapping was able to release the project into a production database available to Mapfluence customers.


When a customer approached Urban Mapping to build a view of the world that would be accurate for its customers across ten languages and hundreds of countries, the team developed a comprehensive process to accurately source, define, and render the data necessary to create a basemap containing the highest level of accuracy and local knowledge. In so doing, we had to reconcile with all the world's boundary disputes and manage the translation of ten sets of worldwide place names.

Boundary data was sourced from a combination of national mapping agencies, commercial vendors, and community generated mapping projects to create a unique repository of boundaries that were then merged where appropriate and categorized into specific views depending on which country and language the user selected for their project.

Place name data was compiled using Urban Mapping's propetiary gazetteer of canonical place names. Where we did not have pre-existing coverage, we used a crowd-sourced approach to survey, adjudicate, and audit place names contributed by local residents. This process is enormously complicated. Non-native language place names are sometimes exonyms, literal translations, or a combination of both.

Finally, fonts and text are one of the most important and one of the least understood choices cartographers make when designing a coherent map, however many fonts do not contain characters across all languages. Working with commercial font vendors and open source projects, Urban Mapping successfully developed a character set appropriate for rendering across the regions and languages supported by the basemap.

To ensure the final product was rendered accurately, Urban Mapping harnessed the power of the cloud to recruit local residents to evaluate and provide feedback on the rendered product. Users were surveyed to determine accuracy, coverage, and potentially sensitive areas.

Learn more about Data Sourcing

© 2014 Urban Mapping Inc. Privacy Policy