As previously announced (here), we recently worked in partnership with Green Cat Renewables Canada Corporation to bring the TrueViewVisuals augmented reality experiences to Alberta in Canada. This post explains how we sourced and processed the terrain data to make it accessible in our visualisation software.
Akin to the Ordnance Survey in the UK, the provincial Alberta government publish their terrain data as OpenData. This is available through the Altalis mapping agency website: http://www.altalis.com/products/terrain/dem.html. We downloaded the entire dataset of terrain data (5.5Gb data) at 1:20000 resolution.
The DEM is provided as an irregular grid. The distance between each data point is not consitent and the rows are not aligned to a grid. I suspect this is due to the way the data was collected and then projected. This makes for significantly more difficult processing than standard .ASC style height data grids.
Terrain Data Processing
Altalis data is organised using the National Topographic System grid (see more at http://www.nrcan.gc.ca/earth-sciences/geography/topographic-information/maps/9765). This scheme coveres the whole of canada, so only some of it is relevant to Alberta.
Alberta is then divided into big squares (e.g. numbered 72, 73, 74, 82, 83, 84). Each of these squares then sub divided into 16 smaller squares refered to as 82A to 82P. Data is provided for each squares in four files North East, North West, South East, South West.
Part of the processing was to merge all this data into a single format that could be imported into our terrain system. This included the following steps:
- Merging- all the data for large region together and processed it together. This helps to ensure we smooth joins between blocks of terrain data correctly
- Interpolation – as the data in not regular interpolation is used to “smooth” the surface and to fill holes
- Gridding – To ensure that the end result provides a data point every 50m as we require for visualsiation
This allows the terrain to be loaded into our cloud delivery system (where it occupies approx 20Gb data) in SQL Azure.
For our mapping overlays we chose OpenStreetMap as the source of data. This allows us to choose the important data to ensure that everything is algined correctly. There is not much mid-scales in Canada to use, so we have imported roads, waterways and housing to provide an overall context and navigation map.
TrueViewVisuals – Ready for Alberta
Now we have all the data loaded and ready, we can create visualisations in Alberta in minutes. From renewables to grid to buildings, TrueViewVisuals can help visualise anything anywhere.