Forest Plantations

Hello forum,

I’ve been doing some mapping from Bing imagery in Scotland recently. There are some large forest plantations I have started to map. What I’m not clear on, however, is how to:
(a) map areas which have either been cleared (but which are likely - presumably - to be replanted in the future), and
(b) map other areas of very young tree growth.

The Bing imagery is very good quality in the areas I’ve been working on, so it’s pretty easy to identify these areas.

Should I use the landuse=forest tag in both instances? Or is there some other treatment?


you may need to check the bing tile dates for this type of classification.

See this page by Martijn van Exel

a bit of Scotland specifically

Pics seem recent at hi res.

Thanks. Blimey. The footage I’ve been using seems to be 10-14 years old! Aaargh!

In which case…those areas which are obviously cleared forest (you can spot them a mile off) have probably been replanted now. Those appearing to be full of young trees are now probably full of not-so-young trees. And some heavily wooded areas will now have been cleared. It’s the circle of life…etc. I suppose (?) I should tag them all as landuse=forest and accept that on any given day/month/year, some areas will be wooded & some not. Not sure what the alternative is.

At low zoom levels the 2004 images seem common, but once you get closer in to about zoom 12 and closer they seem to be mainly 2011/12 for the bit i checked initially.

zoom 11
zoom 12

but then again for another area at zoom 19
they are from 2000

so must depend on the commercial demand for the sat pics for different areas I guess.

Aaaaaah…my mistake. I was looking at the dates of the low resolution blurry imagery that you see when zoomed way out. The hi res stuff pops up when you zoom in. That stuff’s from Apr-Jun 2009.

The wooded bits around here: - are what I’ve been looking at.

You can also cross-check against the Ordnance Survey Open Data StreetView layer in the editor (beware they have a somewhat liberal interpretation of wooded).

I haven’t used it yet, but I would suggest using recently_cleared=yes (or if possible with the imagery date instead of yes) for the recently felled areas. This seems to be a widely used convention in various National Forestry inventories, and will make anyone it easier for anyone looking to check if a new survey /update is required. I think typically 5-10 years after felling is a suitable time to find out how the land is now used.

(I have been spending quite a lot of time recently looking how we can improve the detail of tagging for woodlands)

Thanks. Yes, I’ve often cross-checked with the OS Streetview layer. Seems to be pretty accurate although tends to be “maximal” in that it includes areas seen on the Bing layer as having been cleared.