Co-curating an open point cloud timelandscape



View from Castle Hill, Cambridge rendered with Potree from Environment Agency LiDAR point cloud open data

The Environment Agency’s LiDAR survey is a magnificent open data resource and an impressive leap of faith from the previous business model. For a while I’ve been wondering, how might a platform for co-curation increase the value of this asset within a national/global data infrastructure?

I obtained the digital terrain and surface models for Cambridge in January 2015 as part of an artist-led maker challenge run by Collusion which involved combining the data with OpenStreetMap to produce a prototype city model in Minecraft.


At the time the Environment Agency Geomatics data was neither free nor open but was easily discoverable via and generously and swiftly provided for our prototype with 100% discount.


We exhibited the prototype at the 2015 Cambridge Science festival and a handful of times subsequently prompting some interesting and intergenerational conversations (and solutions) around land use, cost of living and placemaking.

One of the recurring questions from the prototype was:

“Could this LiDAR data really form a reference 3D city model within which to play out city challenges?”

However, In September 2015 that changed:

Laser surveys light up open data and one year on even all the raw point cloud data is open: 


Yes— in theory. The prototype revealed 3 challenges:
1. The EA LiDAR data wasn’t open . 





2. Geographic coverage: In the case of Cambridge, much of the Coleridge area (incidentally the leader of the city council’s ward) is missing.

3. Temporal coverage: the data for Cambridge was collected from 2006–2010. So the centuries old university buildings look great but all of the new development is missing — and there’s been a lot of that since 2010. In fact the last three properties I’ve lived in are missing from the data.

This didn’t seem to detract from the interest and I’d typically get the following questions:

“When’s it going to be updated?”

“What is the cost to fill in the gaps?”

to which I had no answer.

I learned enough from the prototype to believe that

a) playable models offer an opportunity to help local government shift from consultation towards co-design.

b) there’s potentially both the demand and opportunity to create a more complete open LiDAR (or general point cloud) dataset.

Registering Demand

Mapping demand might enable additional funding to be crowdsourced locally to address gaps in the dataset. Existing networks of open data advocates and smart city initiatives are well placed to register this demand. In Cambridgeshire we have the Cambridgeshire Insight partnership and Connecting Cambridgeshire’s Smart Cambridge programme.


There are other sources of supply that could contribute to a point cloud dataset. For example release of data captured for scientific research and archaeology.


Better visibility of current gaps and registered demand might even allow forward flight planning to be optimised to incorporate filling survey gaps where there is demand and funding.


Capturing and processing LiDAR data has its complexities. Curation would be important and with the right approach to curation additional point cloud sources could be incorporated with any limitations clearly understood by data consumers.

Drones, (or kites or balloons or helikites)

With a platform and curation, new opportunities for supply could emerge with potential for localised UAV surveys to fill gaps or add to the dataset.

Back down to earth

While the price of LiDAR sensors remains out of reach for most of us (butgradually coming down), combining open point cloud LiDAR survey data with other techniques such as photogrammetry can create a good basis for models whether physical, virtual or somewhere in between (augmented).

Sensor lending

Coordination of equipment lending could facilitate community projects adding to the dataset. Maybe this could plug into existing initiatives like, maker spaces, or even public lending libraries. 


A combined point cloud dataset could form the basis for a reference digital timelandscape, providing a temporal and spatial context for decisions today and an archive for tomorrow’s digital archaeologists. It could also help to reduce the gap between data insights available to all and those only available to big business such as Google and Uber:

Uber starts mapping UK city streets for best pick-up and drop-off points via @cbronline #UK #cities #uber #data

Moreover it would support the development of data science skills and literacy around increasingly prevalent active and passive sensing technologies.

Let’s talk

It feels like the community, expertise and goodwill is there to explore the possibility of a collaborative point cloud. Perhaps this is a conversation that folks involved in OpenDefra and the Open Data Institute could facilitate?

In any case I’d like to know what it might cost to update the survey for Greater Cambridge/Cambridgeshire and whether any of our patch is on the Geomatics forward flight schedule.

If you’re exploring any of the above, please let me know as I want in!

Richard Hall, Cambridge, September 2016

This article was originally published on Medium on 29/09/2016. The original blog post is available at: