Open Patrick1G opened 9 months ago
I was not aware that more classes were delivered. Thanks for the heads up. We will check the data that we have and adapt the style as necessary.
Hello @Patrick1G looking at the data, it is difficult to how we should adapt the style.
We currently visualize grpot_Kat1
(for which the values are either 0 or 1).
We show the building as green, if the grpot_Kat1
value is 1 and grexisting
is 0.
There are two other Kat values, for which we do not know what they mean. There is grpot_Kat2
and grpot_Kat3
.
Could you let us know how the expected classes should be distinguished visually and if these are the parameters we should focus on?
@slumnitz or @FranLeonelli Could you check with DLR colleagues how these values should be used in the tool?
After receiving feedback from Annekatrin, the 3 classes we have are: GRpot_Kat1 Area of roof(-part)s where the slope between 0° and 5° is greater than or equal to 5m²; 1: true, 0: false GRpot_Kat2 Area of roof(-part)s where the slope between 5° and 20° is greater than or equal to 5m²; 1: true, 0: false GRpot_Kat3 Area of roof(-part)s where the slope between 20° and 45° is greater than or equal to 5m²; 1: true, 0: false
We can define 3 different classes. When there is an area of >=5m2 and: <= 5 degree slope: high suitability 5-20 degree slope: satisfactory suitability 20-45 degree slope: low suitability
Since these classes can be overlapping on a single roof, we should choose the highest-level suitability for those roofs which present more than one suitability class.
As Annekatrin suggests it should be mentioned in the description that the suitability refers to the whole roof or roof parts depending on the structure.
@FranLeonelli ok, thanks for the inputs. Could you send the color legend we should use so we can include it and know the expected colors? Just to clarify, only roofs larger then 5m² can have a true value, so we don't actually need to check if the area is below that size when applying the style, correct? Should we (as we did previously) only show the potential of rooftops that are not marked as already having green roofs?
You can upload the new color legend and update the text in the assets, here: https://github.com/eurodatacube/eodash-assets/tree/main/collections/SOL1_3_Green_Roofs_grpotential
the current layer for green roof suitability only shows binary information as suitable/not suitable./..
According to DLR, 4 or 5 classes were distinguished and delivered.
It would be good to expose these different classes, using a categorical color scale (and not using the green that was used for green roof detection)