[Talk-GB] Solar Power mapping update Q2 2020

Gregory Williams gregory at gregorywilliams.me.uk
Thu Jul 2 18:15:29 UTC 2020

Thanks Jerry, and thanks to everyone that's continued to contribute
more coverage.

The next quarter's update to the FiT register should be published in
the next few days. So I hope to find time to update the site to use
that soon.

I continue to be amazed at the steady progress in the coverage. Though,
as you say, there are quite a few areas where the imagery either just
isn't clear enough to untangle the ambiguities, or is clear but isn't
recent enough.

Personally, I've recently been trying to concentrate on a mixture of
areas with less than 10% coverage, and on the lightly-mapped LSOA
hotspots that my tool picked out.



On Thu, 2020-07-02 at 18:56 +0100, SK53 wrote:
> We passed a couple of milestones a few days ago:
> 20% of FIT totals
> 170k individual panels mapped (excluding those in solar farms)
> In terms of coverage there are now well over 50 LAs (all in England &
> Wales) with more than 50% of solar installations mapped, with around
> 10 exceeding 80%. Areas with good coverage are:
> Scottish Central Belt: helped no doubt by more atomic data much of
> the Central Belt is around 20% mapped.
> North-East (former Tyne & Wear): Newcastle, Gateshead, Sunderland and
> North & South Tyne.
> North Wales: Conwy, Flint, Denbigh & Wrexham. Most panels in the
> first three are in the coastal resort towns, but reasonable rural
> coverage.
> North West: recent activity has been around Preston, Blackburn Wigan
> and Chorley.
> East Midlands: mainly Leics & Notts. Improved & recent imagery for
> Leicester made a huge difference.
> West Midlands: Warwickshire, Worcestershire & Herefordshire are
> roughly in the 20-30% zone. ALso extending into the South Wales
> valleys. brianboru's detailed mapping in the latter is another good
> index of rural coverage. 
> South Coast: Bournemouth area & Southampton, all at over 50%
> More rural areas continue to be challenging: older imagery which is
> often difficult to interpret doesn't help. I've experimented in
> places where every building is already mapped by stepping through
> each building, but still one may only find 20% of the number in FIT.
> London and immediately adjacent areas also have relatively little
> mapped. Imagery can be a problem, but also finding panels in older
> and/or larger housing with more complex roof shapes is hard.
> One thing I'm continually amazed at is how many places have buildings
> mapped, which is very helpful for this task. However in a couple of
> places: Ribble Valley & Leicester - it is clear that better imagery
> would allow existing building outlines to be improved, but also that
> plenty of buildings have been extended, demolished or replaced. This
> type of activity lends itself to combined work using tools such as
> Tasking Manager or MapRoulette and might be worth considering in the
> future for a quarterly project.
> There's still no shortage of places where a lot of panels can be
> mapped quickly, although more systematic mapping of a single LA often
> requires a couple of passes over imagery. 
> Looking forward to achieving the next milestones of 200k & 25%.
> Jerry
> Personally, I'm concentrating on areas adjacent to the existing well-
> mapped (50%+) areas with the aim of extending these areas.
> _______________________________________________
> Talk-GB mailing list
> Talk-GB at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-gb

More information about the Talk-GB mailing list