[Qgis-user] Raster Calculator Miscalculations - User error or bug?

[Burley, James (ASC/CSA)]

NON CLASSIFIÉ / UNCLASSIFIED

Hello to the QGIS user's community,

I'm a planetary scientist working for the Canadian Space Agency who is working with lunar south pole data. I'm struggling with an issue when running a simple multiplication raster calculation that I've included below (sorry for vague layer names, data is not public):

"Layer1" * "Layer2" * "Layer3" * "Layer4" * "Layer5" * "Layer6"

The output raster is set to the spatial resolution of the highest resolution layer, as I'm trying to gather contextual data from coarser datasets at very specific locations.

Most cells are calculated correctly, but many cells (<10-20%) are miscalculated to the tune of 10-30% of the expected value (sometime values are calculated lower than they should be, sometimes higher. Even when the highest spatial resolution cell is fully contained within a single cell in all other layers (no overlapping cell boundaries that might lead to different resampling values under a nearest neighbour protocol), the calculated value can still be incorrect.

I'm running QGIS 3.44.7, my system is using an i7-1265U with 32 gb of ram. The rasters are large, 5000x5000 cell 32-bit float Geotiffs. All are using the same coordinate system (ESRI:103878 - Moon_2000_South_Pole_Stereographic).

I've exported the input layers so I can run the same calculation in ArcPro, and the cells are all calculated correctly. Not sure if that helps narrow down the issue.

Cheers,
James

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-user/attachments/20260303/ede5544e/attachment-0001.htm>