I am using miniconda and the Open Data Cube project. I am trying to ingest some Landsat_5 data into a datacube, where the geotiff are converted to netCDF files. Using the datacube code I used the ingest tool:

datacube -v ingest --executor multiproc 12 -c C:\envs\DataCube_Files\LS5_USGS_UTM17N.yaml

It creates NetCDF files and they look correct, however when running through my jupyter notebook code where I load the landsat dataset with:

landsat_dataset = dc.load(x=x_extents, 
                      platform = platform,
                      product = product,
                      measurements = ['red', 'green', 'blue', 'nir', 'swir1', 'swir2', 'pixel_qa']) 

I encounter an error where its looking for a LS5...6_48...nc file that doesn't exists. I have the LS5...6_49...nc , 6_50, and onwards files but not the 6_48. It occurs for multiple files and not just one. I think maybe its an error when converting the geotiff to NetCDF but not sure how to fix it. I've tried making sure my libraries are updated and it looks all good. Error is as follow:

RasterioIOError: 'NetCDF:D:\Cameron\CMAs\Toronto\NetCDF_Tiles\LS5_TM_USGS_SR\LS5_TM_USGS_SR_32617_6_48_19890111152725138088.nc:red' does not exist in the file system, and is not recognized as a supported dataset name.

Any suggestions? I've reduced my images to 15 from 8000 and have completely dropped the datacube and made a new one with only those 15 and still encounter the same problem.


Figure it out after some trial and error. I had Rasterio version 1.0.18 installed and when I downgraded it to version 1.0.13 the error disappeared and everything works. Note that it still did not create the LS5...6_46....nc file but I guess it also didn't the file in the database so it doesn't occur as an error.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.