New ‘Data Layers’ for the OpenAQ Platform!
Note: We have updated our database schema with this data format, but it is not yet query-able via the API. Currently we just have one research-grade source, which is the SPARTAN data.
OpenAQ originally started as an effort to capture and make more easily (read: historically and programmatically) accessible data from government entities already shared publicly. A lot of these data would be nearly or completely impossible for the public to access otherwise.
But of course, there are other types of data out there, like research-grade data, collected by scientists at universities and research institutes. The ultimate destination of these data are varied and sometimes multi-pronged; often they are embedded in papers in academic journals and more and more frequently, available on open data repositories sponsored by universities or other entities. However, there is no data commons for accessing aggregate research-grade air quality data from different organizations and studies available programmatically, say with an API.*
And the advantages of creating such a commons are a) data are in a centralized place and b) they are query-able in aggregate (e.g. “Hey API, give me back all data, reported from any institution, in which PM2.5 levels fall between X and Y”). Or put another way, it makes it a lot easier to sift and sort through the data from different organizations and extends their usability past the original researchers’ intended purpose.
So! We’ve added some new layers to the OpenAQ data format to distinguish new types of data available on our platform:
All of the data we have aggregated to the platform to date is now tagged as “government” for sourceType.
Check out this post, which discuss SPARTAN (Surface PARTiculate mAtter Network) data — a research-grade data source supplying historical PM2.5 data for 9 different countries on OpenAQ. For sourceType, this is tagged as “research”. We’ve also added the capability to differentiate between mobile and non-mobile sources (see mobile parameter) and a sourceType of “other” in anticipation of adding in data from other sources, like mobile, low-cost sensor data too.
*To our knowledge. Please let us know if there is such a service!