[Talk-GB] Mapping of Kielder Forest(s)

[Nick]

As a follow up, an informative thesis is 
https://labor.beuth-hochschule.de/fileadmin/labor/gem/arbeiten/Keil_Master_Arbeit.pdf 
<https://labor.beuth-hochschule.de/fileadmin/labor/gem/arbeiten/Keil_Master_Arbeit.pdf> 
that highlights both some of the challenges (e.g. data quality) and also 
opportunities.

On 03/09/2021 10:10, Nick wrote:
> Hi Russ and James
>
> A forest is and has always been more than just trees, so designations 
> become even more complex when the land use is mixed e.g. 
> agro-forestry. The point about mapping is surely that there are 
> multiple uses that maps are put to. So for example, it could be used 
> for estimating biomass in which case species, age, percentage tree 
> cover etc. are important data sources. If we map large areas the 
> assumption is that it is uniform unless additional data (tags) 
> indicate otherwise. Conversely if small areas are mapped (sub 
> compartment approach) but lacks precision and accuracy, then any 
> subsequent analysis could be flawed. The point I am trying to make is 
> that what and how we map affects the usefulness of the maps for a wide 
> range of users - the more detail, the more potential benefit. 
> Therefore micro-mapping to my mind has more potential benefit than 
> coarse-mapping but also depends on the additional linked data provided.
>
> Cheers
>
> Nick
>
> On 02/09/2021 14:26, Russ Garrett via Talk-GB wrote:
>> We clearly don't have a consensus here, so I will concede defeat. I've
>> reverted all my changesets which have touched Kielder forest.
>>
>> Unfortunately I don't think there is a workable, maintainable way of
>> tagging the names of these forests at the moment.
>>
>> Russ
>>
>> On Thu, 2 Sept 2021 at 13:47, James Derrick <lists at jamesderrick.org> 
>> wrote:
>>> Hi again Russ,
>>>
>>> On 30/08/2021 19:59, Russ Garrett wrote:
>>>   > First off I would like to emphasise that I am not suggesting the
>>>   > removal of any detail here.
>>>   >
>>>   > I am suggesting that small gaps in the forest should not be
>>>   > represented by gaps in the forest polygons, but should be replaced
>>>   > with man_made=cutline (and/or a highway tag), which would reduce 
>>> the
>>>   > number of individual landuse polygons. Larger gaps (such as the old
>>>   > farmsteads, etc) should certainly be preserved, potentially as 
>>> holes
>>>   > in a multipolygon.
>>>
>>> It may be me, but these two statements seem to be incompatible.
>>>
>>> In several areas I have taken the time to (start to) micro-map
>>> individual stands of trees with a single area. As cutlines/ firebreaks/
>>> haul roads are seldom linear features (river ravines, farmsteads,
>>> bothys, stone walls, shedding rings, gullys, etc), this is additional
>>> information with real features useful for navigation - remember not all
>>> maps are consumed on a GPSr.
>>>
>>>
>>> Rejoining these areas into one larger area and adding linear ways to
>>> represent (a few of) the gaps in trees in my mind absolutely is 
>>> removing
>>> detail.
>>>
>>> I'd all it **macro-mapping**, at the risk of creating a hashtag.
>>>
>>>
>>> In areas where the mappping complexity is low, an outer area
>>> landuse=forest plus ways man_made=cutline makes a lot of sense - it
>>> gives a good balance of effort / useful detail.
>>>
>>> This isn't the case in Kielder, however. A lot of detail was 
>>> imported in
>>> the past from other sources with individual stands of trees, and broken
>>> into small areas / import viewport clipping fixed later.
>>>
>>> As a simple example of imported data showing pre-existing areas:
>>>          G3YAC imported from OS_Opendata_Natural_Line
>>>          https://www.openstreetmap.org/way/131625930/history
>>>
>>> Would you join these relatively rectilinear areas and add cutlines?
>>>          https://www.openstreetmap.org/way/131625842/history
>>>          https://www.openstreetmap.org/way/131625810/history
>>>          https://www.openstreetmap.org/way/131625930/history
>>>
>>> I was about to reference a more complex case but...
>>>
>>>
>>>   > I'm not suggesting that the entire area should be tagged as 
>>> "Kielder
>>>   > Forest".
>>>
>>> Sadly, after looking around Kielder observatory, I see you've already
>>> started merging areas and removing detail:
>>>          https://www.openstreetmap.org/way/383346121/history
>>>          https://www.openstreetmap.org/changeset/110350625
>>>
>>> A single 50km2 (roughly 6km x 4km) multi-poloygon is a poor replacement
>>> for the previous detail, and I ask you to stop **macro-mapping**.
>>>
>>>
>>> James
>>> -- 
>>> James Derrick
>>>       lists at jamesderrick.org, Cramlington, England
>>>       I wouldn't be a volunteer if you paid me...
>>>       https://www.openstreetmap.org/user/James%20Derrick
>>
>>
>
> _______________________________________________
> Talk-GB mailing list
> Talk-GB at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-gb
>