Progress update 1 – geospatial data review
It’s now been just over a month since U· Geo kicked off, so I think the time is right for a progress update. Most of my time as project officer has been divided over two major tasks – a review of geospatial data at the UK Data Archive and the mapping of Archive metadata to various geospatial specifications. Across two posts in the coming weeks, I will take a look at each of these in turn and give some details of the process’ involved and what they are yielding.
Review of geospatial variables in national survey data held at the UK Data Archive – or how I learned to stop worrying and love geographies
At the core of WP1 is an assessment of geospatial data in the Archive collections. This is partly exploratory, but also forms the basis of a database that we hope will eventually support an Archive geo-service.
This fairly mammoth task breaks down into:
1) locating the geospatial variable(s) in the data files, and if applicable, any variable coding
2) ascertaining which geography (i.e. spatial unit) they (hopefully) conform too
3) and finally, potential linkage with available map resources (e.g. ESRI shape files).
The main challenge thus far has been the general vagueness with which geospatial variables are defined within data sets and documentation. This is common across all the major social and economic surveys, and can make GIS use challenging to say the least. In an ideal world, each of these variables should have a time-stamped unit definition.
Testament to the inextricable complexity of UK boundary geography, even the humble GOR (there are 9 in England – see graphic below) is subject to temporal variation, with three historical changes to the classification in 1996, 1998 and 2001.
You can begin to imagine then, the complexity of smaller resolution historical geographies, as we move down through counties, output areas, divisions etc. The thought of the 10,000+ electoral wards, with yearly altered boundaries and name changes is enough to send shivers down any GIS users spine! So already, the areas for improvement in the accessibility and documentation of these data are becoming starkly apparent. Thankfully (for my sanity), retroactive linkage is proving manageable with a bit of lookup table wizardry, and the database grows steadily. I am careful to keep in mind the 1st commandment of GIS at all times (thanks James at #jiscgeco).
So far my work has centred on boundary geographies, as opposed to point geographies. One of the most exciting recent developments at the Archive is the advent of the Secure Data Service. Part of their remit is to make accessible lower level geographies than have typically been made available for national survey data. This includes national grid referenced data, which I am hoping to be able to look at as part of this review – it will be very interesting to see how useable this data is ‘out of the box’ as it were.
Check back soon or subscribe to keep updated – next time, metadata standards and more tenuous pop culture references!