PET raster file has NoData

[Lalalashp]

I generated multiple PET raster for the same area and looking into them i realized there are No Data pixel while the input data were not NULL.
I was wondering if there's any know problem that can fix this or is just how the plug-in works.

PET input data:
age: 25
Weight: 60
Height: 165
Activity: 195 ( i tried even with 160 but i have the same results)
Clo: 0,25 (i tried with 0,9 too and had the same problem again)

FileInput.zip

thanks in advance!

[biglimp]

Can you please also upload you input grids (DSMs, land cover etc.) and you met forcing file.

[Lalalashp]

input2.zip

There should be everything. I also added my PET output for reference, if needed.

[biglimp]

I cannot identify what geodata you have been using. Please specify:

1. DSM
2. DEM
3. CDSM
4. Land cover

[Lalalashp]

sure!

1. DSM: p_dsm_buil.tif
2. DEM: p_DTM5.tif
3. CDSM: p_cdsm.tif
4. Land cover: p_atlasRecl.tif

My bad, i just putted everything i a had together.

[biglimp]

I see two problems:

1. Your land cover is not correct. It should have the same spatial level of detail as the other data. Have i look at the tutorial data for SOLWEIG. There you see that each building is separated from whet is in-between your buildings. In your land cover, your just have large neighborhood areas that is buildings.
2. Your DSM is very caothic. I guess your tried to derive this from a LiDAR dataset? You need to make sure that the vector buildings used to calculate wind speed in URock are the same as the buildings identity if you would substract your DEM from your DSM.

[biglimp]

Can I convert this to a discussion item instead, as this is not an issue with the tool itself?

[Lalalashp]

Good afternoon,

I tried refining the land cover raster by adding more details, but unfortunately, the results didn't change. I have attached the new file, which you can find inside the Land Cover folder.

I also included the folder with the new PET rasters and an example of the missing pixel in my area, in case that provides any useful clues. regarding the DSM, I was provided with the file directly, but I believe it was derived from a LiDAR dataset.

Do you still think the issue lies with the dataset itself? Or could it be related to how UMEP processes data close to buildings (or something similar)?

Thank you very much for your help!

Data_updated_new.zip

[biglimp]

Yes, it is still your data that is causing the issues. The land cover itself is not creating the nodata around buildings. That is the differences between your DSM and your vector building dataset. Did you have a look at the tutorial data?