Please allow choosing more than one map layer

[DenoBeno]

Related to #59, issue no. 5

• 5. allow user to overlay several layers and decide how to colour-code each
  Claudia has proposed this as a way to rapidly get an idea of where the main issues in the project area are. See CSIS Demo for European Week of Regions and Cities csis#105 (comment)

The easiest way to achieve this (at least the "show several layers" part) would IMO be by allowing the users to select more than one layer. This isn't a greatest thing in terms of the usability, but it should be easy to do.

The use case described by claudia should not be hindered by layer overlaps - it's either awood or it's a park or it's a lake or it's a house, never all of it together...

@p-a-s-c-a-l , @claudiahahn WDYT?

[claudiahahn]

The easiest way to achieve this (at least the "show several layers" part) would IMO be by allowing the users to select more than one layer.

I agree. I could imagne that is what a user is interested in. To see potentially hot and potentially cool places at once

The use case described by claudia should not be hindered by layer overlaps - it's either awood or it's a park or it's a lake or it's a house, never all of it together...

That is true. The layers were defined based on urban atlas data only and the layers don't overlap. Selecting an additional layer would just lead to another area being displayed with a different color.

[p-a-s-c-a-l]

This makes only sense for vector-based layers with transparent areas like, the Urban Atlas layers. These have to be treated differently than raster/grid-based layers (e.g. hazard) where an direct overlap does not work. A calculation of a 'differential' layer on client-side doesn't seem to be possible either.

So maybe we should pick up our initial idea to treat Urban Atlas Layers as additional Background Layers that are not only available in HC-LE but in any step.

Here is a list of potential useful background layers. We have to attach a special tag to the respective resources in the Data Package, so that the map component can decide whether to treat it as background layer or not.

[DenoBeno]

I think that this is a good idea. Indeed, if we can differentiate between layers that can be combined as a "background" (vector layers) and those that can't (grid/raster layers), it will be easier for the user.

If we could also assure that this choice is not forgotten on reloads and ideally even on change of the map, it would be even better IMO.

I do not believe that overlying the hazard layer on top of the already colorful background will work very well though. Not unless the users can change transparency of this layer. but this is a bit of a luxury problem that we can discuss at some later point.

[DenoBeno]

considering the "usable layers" from clarity-h2020/data-package#5

#	Layer name	Sources
1	Water	Urban Atlas
2	Roads	Urban Atlas
3	Railways	Urban Atlas
4	Trees	Urban Atlas + Street Tree layers
5	Vegetation	Urban Atlas + European Settlement Map
6	Agricultural areas	Urban Atlas
7	Buildings	European Settlement Map
8	Built open spaces	European Settlement Map
9	Dense Urban Fabric	Urban Atlas
10	Medium Urban Fabric	Urban Atlas
11	Low Urban Fabric	Urban Atlas
12	Public, military and industrial units	Urban Atlas
13	Basins	EU-Hydro
14	Flow accumulation steams	USGS - HydroSHEDS
15	DEM	EU-DEM
16	Surface Temperature	SL_2_LST_

1-12 are IMO useful for this purpose. 13 and 14, probably too(?). Not sure about the last two. Surface temperature doesn't sound like a vector layer and no idea what is DEM.

The layers from the second table, no idea. Some of them sound like interesting information, but I think we aren't even showing them at the moment. The third table is quite interesting (that's the same data as in table 2, just presented differently, I presume?), but we aren't showing any of this, are we? And I'm not sure if this is vector or grid data.

So those are the background layers that should be available (at least) in the HC LE Step:

Finally, the "Inventory Layers" sound very much like "exposure" to me. I never quite understood why we aren't showing such data under exposure.

[p-a-s-c-a-l]

List of multi-selectable background layers is implemented:

ATM, we use layers tagged with taxonomy term resource-type:local-effects-parameter to populate the list of background layers. In the future we may introduce a separate resource type, e.g. background-layer.

[DenoBeno]

Nice. Can we also have a legend for these layers? :-)

…

On Tue, 29 Oct 2019 at 14:27, Pascal Dihé ***@***.***> wrote: List of multi-selectable background layers is implemented: [image: grafik] <https://user-images.githubusercontent.com/1788226/67770822-6d0a4880-fa57-11e9-8046-3b9e1947828b.png> ATM, we use layers tagged with taxonomy term resource-type:local-effects-parameter <https://csis.myclimateservice.eu/taxonomy/term/1096> to populate the list of background layers. In the future we may introduce a separate resource type, e.g. background-layer. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#65?email_source=notifications&email_token=AAWTC7QTZ3L64NMRXDGXUPDQRA22LA5CNFSM4JD7K2IKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOECQPDLA#issuecomment-547418540>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAWTC7U325NWPJTDLBIM2A3QRA22LANCNFSM4JD7K2IA> .

[p-a-s-c-a-l]

@negroscuro Is it possible to add a legend to the HC-LE input layers?

[p-a-s-c-a-l]

done

[DenoBeno]

There is one strange effect in the map now. E.g. on https://csis.myclimateservice.eu/study/35/step/1524/view/maps, the background layers are listed once as background (multiple choice possible) and once as default (single choice)

[p-a-s-c-a-l]

Yes. Background layers are currently all HC-LE layers as explained here. Therefore they are duplicated in the HC-LE step. We could introduce an additional separate resource type, e.g. background-layer and then hide those layers in the single choice list.