omIndustrialStationaryTransportEmis.py

[NasimehShahrokhi]

Describe the bug

I ran the omIndustrialStationaryTransportEmis.py to map emissions on a finer resolution domain (3*3km2) but it was unsuccessful, and the output spatial pattern does not follow the pattern of nighttime light data. For example, when I run the scrip for urban scale domain around Melbourne, it doesn't suggest a hotspot around city of Melbourne but maps a hotspot offshore.

Failing Test

When I run the script, there are some warnings that some negative/positive indexes are out of bounds for my domain:

python3 omIndustrialStationaryTransportEmis.py Loading domain info processEmissions for Industrial, Stationary and Transport Clipping night-time lights data to Australian land border /path to lib/lib/python3.8/site-packages/rasterio/features.py:328: ShapeSkipWarning: Invalid or empty shape None at index 1 will not be rasterized.   warnings.warn('Invalid or empty shape {} at index {} will not be rasterized.'.format(geom, index), ShapeSkipWarning) Mapping night-time lights grid to domain grid index 670 is out of bounds for axis 0 with size 308 index 670 is out of bounds for axis 0 with size 308 index 669 is out of bounds for axis 0 with size 308 index 654 is out of bounds for axis 0 with size 308 .... index -570 is out of bounds for axis 0 with size 308 index -588 is out of bounds for axis 0 with size 308 27238 lit pixels were ignored Writing emissions data for OCH4_INDUSTRIAL Writing emissions data for OCH4_STATIONARY Writing emissions data for OCH4_TRANSPORT

Expected behavior

The spatial pattern for this sector of emission was supposed to follow the nighttime light pattern, however, it did not happen.

Screenshots

System

• OS: Linux
• Python version: 3.8.5
• Please also upload your poetry.lock file (first run poetry lock to make sure the lock file is up-to-date)

Additional context

[lewisjared]

Peter is taking a look at this now. Are you able to provide your the input domain file that you used?

[prayner]

Nasimeh supplied the file and I've been able to reproduce the problem and limit it to the rasterised fields read from tiff files. Next to see which reproj causes the error