Hello everyone,
I have a problem with the outputs of the annual water yield model: the values of PET and AET are equal to 0 for each watershed/subwatershed, both in CSV tables and in TIFFs (aet, fractp). While there are no problems for wyield and precip. (The Z value is 16).
How can I fix it?
Input Data.zip (191.6 KB)
Workspace.zip (349.0 KB)
Thanks.
Hi @DavideD, and welcome!
Thanks for posting your data.
When I bring the input rasters into either ArcGIS or QGIS, they all appear to be 3 band rasters, with RGB values. Usually, we use single-band rasters whose values are actual AWC values, root depth values, precipitation values, or the same “lucode” values that are in the biophysical table (for the land use/land cover map). This just makes things easier.
The models can take multi-band rasters, but the AWC/root depth/precip/LULC values must be in the first band. I’m guessing that the problem with 0s in the outputs could be that the first band of the LULC map does not contain the correct mapping to “lucode” in the biophysical table. Looking at the files in the intermediate Workspace folder, kc.tif has values of all 0, root_depth.tif is all 1, and veg.tif is all 0, which makes me think that the values in the LULC raster are not getting correctly mapped to values in the biophysical table. It is then also possible that the other input rasters (AWC etc) do not have the correct values in their first band, so even if the model does produce results, they may not be based on the correct input values.
So you’ll need to go back to your data sources and try to export them to have the correct format. Look at the Annual Water Yield sample data for an example of what these look like.
Also, looking at the biophysical table, I see that the values are separated by semicolons, not commas, as is required by InVEST. So you’ll want to re-save the biophysical table with values separated by commas.
~ Stacie
Hi, I think I solved part of the problem, thank you so much: now the results seem much more consistent than the inputs, but now nothing appears in the AET item in the results (not even as 0, there is just a space between the other values).
What may have happened?
Here my inputs and outputs.
Input Paradiso.zip (503.4 KB)
annual_water_yield_workspace.zip (704.3 KB)
(these are data relating to another area than those of the message a few days ago, but the organization of the data and the problem are the same)
Glad to hear you’re progressing @DavideD. Looking at the Paradiso data, and the Evap Para.tif file has all very small, negative values in it (usually the values are significantly larger, and lowest value is 0). This will definitely cause problems for the AET calculation. So I suggest reviewing that raster’s values and see if an updated version produces better model results.
~ Stacie
Hello, the problem was exactly that and now the model works perfectly and produces correct data.
Thanks for all the support you have given me!
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.