issue with writing multiband GeoTiff using dask array

[h3raklit]

Hi there,

im quite new to this and looking for some advice. Following this example from the rioxarray documentation using dask to parallelize reading/writing large datasets:

with LocalCluster() as cluster, Client(cluster) as client:
xds = rioxarray.open_rasterio(
"../../test/test_data/compare/small_dem_3m_merged.tif",
chunks=True,
lock=False,
# lock=Lock("rio-read", client=client), # when too many file handles open
)
xds.rio.to_raster(
"dask_multiworker.tif",
tiled=True,
lock=Lock("rio", client=client),
)

I am trying to load multiple single band tiffs into dask arrays and concatenate them to one 3-D dask array. However the resulting array does not have the rio attribute anymore, which makes it impossible calling rio.to_raster() to write the stack on disk.

Any advice why that might be and how I could circumvent the issue? Thanks a lot!

Actual code:

f = 'filename...'
sList = []
sTags = {'date': dList}

for band, layer in enumerate(fList, start=1):

load single band rasters

dLayer = rioxarray.open_rasterio(layer, chunks='auto', lock=False,)
sList.append(dLayer)

concatenate array list to new 3-D dask array

stackLAI = da.concatenate(sList, axis=0)

write 3-D dask array to disk as multi band GeoTiff

stackLAI.rio.to_raster(f, tiled=True, tags=sTags, driver='GTiff', dtype='float32', lock=Lock("rio", client=client),)

[snowman2]

https://corteva.github.io/rioxarray/stable/getting_started/manage_information_loss.html

[snowman2]

Also, should note that rioxarray is a xarray extension, so it only works with xarray.DataArray and xarray.Dataset objects.

[h3raklit]

Indeed, thanks! The main issue was using dask.concatenate() and not xarray.concat() to build up the 3-D array. Using the later works (spatial attributes are preserved), as the dask arrays are wrapped by xarray data.arrays.