During calibration, how precise should the AWY results be? A case in point, I have annual realised supply volume at 3,270,677,117m3 and the river discharge volume for a hydropower plant at 4,535,507,520m3.
Copied from userguide, “Outputs in the per_pixel folder can be useful for intermediate calculations but should NOT be interpreted at the pixel level, as model assumptions are based on processes understood at the subwatershed scale”.
If I lay the wateryield output per pixel raster over the LULC raster in QGIS using “raster layer zonal statistics”, I get the wateryield output for each LULC class per pixel. This helps me determine how much each LULC class contributes to wateryield. Does it make sense to interprete these results following the warning in the userguide?
Thank you
I’d be interested to hear what our scientists have to say (@RafaSchmitt, @adrianvogl?) but I have a feeling that it’s “as close as you can get”. If you’ve done a good calibration exercise, and that’s as close as you can get, then it is likely that this very simple model is not capturing something important that’s going on in your watershed.
As for the per pixel outputs, it is ok to process them like you did, taking the mean or whatever statistic of each LULC class to compare their contributions. Our warning is meant for people who might want to choose one pixel over the next to do, say, restoration, based on the exact amount of water being produced in that pixel. The model just isn’t made for that.