Open footprints for England and Wales
This is our second CASA blog on the release of comprehensive open footprint data for the UK. In our earlier post we looked value of this dataset for the intelligent cities debate and the potential of OS Open Map Local (OML) to accelerate comprehensive cover. We also identified a quick method of creating detailed open footprints for local areas/local communities, by merging OML with out of copyright historical maps.
Using OML and INSPIRE to create open footprints for England and Wales
NB Since this blog was released OS has retracted its written confirmation that the dataset described below can be released without restriction. Despite INSPIRE having an Open Government License we have been told the merge can only be used 'internally' by our university department. Should you wish to use this merge please check conditions directly with OS.
Last week Adam Dennett at CASA took us a step further towards comprehensive open footprint release by realising that the Land Registry’s INSPIRE polygon dataset could be used to crack the problem of how to split OML’s generalised polygons into individual buildings.
We immediately wrote to OS to check that an OML dataset, adapted using INSPIRE, could be released and OS reponded that this is indeed the case. However OS did state that the INSPIRE dataset cannot be used for commercial purposes without its permission, and that polygons are available under an Open Government Licence (OGL) whose terms need to be met. This includes a requirement that the data source needs to be acknowledged and a link to this source provided. (OML terms also need to be adhered to). OS also noted that the Registers of Scotland have not as yet released their polygons.
OML generalized polygons split using INSPIRE being tested. Adam Dennett
Adam's finding means that we will be able, over the next month, to release footprint data for our test areas along with a methodology for release for the whole of England and Wales. We will then begin to upload our building age data for around 25,000 buildings within The London Borough of Camden and work on further metadata release.
It is worth noting that INSPIRE polygons cover freehold titles, not leasehold, and define the Land Registry boundary of the title. They may in certain cases be different to the Topographic data. Several terraced houses for example may be grouped under a single title. (As OS noted these are not legally guaranteed as they are based on Land Registry's Index polygons - not the actual title plan polygons which are definitive and which can be purchased from a re-seller of the National Polygon Dataset ie LandMark).
Building polygons created using OML/INSPIRE are also simplified, though they are more detailed than OS's commercial OSVectorMap Local product. For city metadata crowdsourcing purposes this level of simplification is not a significant issue. However for research/use where a detailed understanding of the physical make up of a locality is required, and for issues relating to planning, 3D modelling etc these outlines are insufficient. A possible solutions to this issue is described below.
Using LIDAR, INSPIRE & OML for localised areas
Duncan Hay, who is currently working on the CASA/ Survey of London Whitechapel project, last week tested another method of creating open polygons using the INSPIRE and OML data. This time he added LIDAR data (released last year by the Environment Agency) into the mix.
When the OML polygons are split with reference to both the INSPIRE cadastral parcels and the LIDAR building outline, a much more accurate vectorised footprint than possible with INSPIRE/OML only is created. This is comparable to that achieved using historical map outliines, identified in our first blog, which this method now supersedes.
LIDAR data in grey overlaid with OML generalised polygons in pink and INSPIRE polygons in blue. Duncan Hay
Flora Roumpani (http://en-topia.blogspot.co.uk/) is now looking at whether we can automate this process to enable vector footprints for England and Wales to be extracted from the LIDAR data. In the meantime the INSPIRE/OML method will be used for borough/citywide footprint/metadata release and the INSPIRE/OML/LIDAR method for local areas.
Urgent need for comprehensive cover and standardisation
We are however concerned that though open data releases (such as OML, LIDAR, and INSPIRE) are enabling us to build on and support Open Street Map’s long-term work in this area, and to move closer to a comprehensive open footprint dataset for the UK, we are now, along we others, spending large amounts of time mashing together different open products to try and create the open dataset we all seek.
In doing so, (and as more open products are released) we are likely to produce and release an ever greater number of open footprint datasets which are not interoperable.
As well as taking up significant amounts of time, we are also it seems storing up even greater problems for the future, as large amounts of metadata begin to be collected. As each month goes by without a single standardised open footprint format, time which could be used to merge and analyse data collected from multiple sources (to address sustainable city and other issues) will be wasted, and time needed to transfer data, in the long run, will increase.
OS is currently developing a new suite of APIs designed to give greater access to footprint data. For non-commercial use a number of free transactions are permitted per month, However, as each query has to be run individually for each building footprint, for even small local projects the transaction quota will be rapidly used up.
The OML/INSPIRE method now offers the opportunity to create comprehensive open footprint data, more detailed than OS VectorMap Local. Accuracy will be further increased in once footprints are extracted from open LIDAR.
However what is obviously required is a single, standardised comprehensive, regularly updated, OS open footprint product for the whole of the UK available to, and used by everyone.
OS's OpenMap Local is still a beta version. It could be released with building subdivisions and simplified outlines in a way that does not challenge the precision, quality, or market of its MasterMap product.
This would allow UK metatdata collection, visualisation and analysis to proceed immediately. Furthermore, as TfL has shown through its enlightened approach , it would also inevitably unleash a wave of innovation, within mulitple sectors, which OS would benefit from through association/accreditation, and through income generation from add-on products.