I received multiple warnings while running the NDR model. The model did generate outputs and only one warning message was highlighted although there were many warnings about different issues. I am hoping you can help me clear up some of these warnings.
What do you expect to happen?
I think it may have something to do with my LULC raster. However, it is the same raster I used in the other models I ran, which all ran without issues.
What have you tried so far?
I tried using LULC rasters from different years, yet they all had the same errors, yet output results.
Hello @ARiggs thanks for providing your datastack! The key issue here is that your Threshold Flow Accumulation is set to 1. The TFA value determines how many pixels need to accumulate as water flows downstream before the flow is determined to be a stream. When you set TFA to 1, this means that all pixels will automatically be considered a stream pixel. Because the model does not handle in-stream processes, your IC0 raster ends up entirely filled with nodata, which leads to this warning.
The solution here is to use a different TFA value. We recommend experimenting with a range of TFA values until you find one where the streams raster closely matches a real-world stream network dataset.
Also, FYI, I noticed that your DEM is a 3-band RGB raster, which generally means that your pixel values are constrained integers in the range of 0-255. I would suggest that you revisit your DEM to make sure that the pixel values are not constrained to this range unless this is intentional and suitable for your study area. Also, please be aware that InVEST will always take the first band of your raster, so since RGB rasters can have different pixel values in each band, just beware that this can cause unexpected results. Your raster has duplicate information in the R, G, and B bands, so this isn’t an issue with the dataset you shared, specifically, but please be aware of this anyways.
Thank you for providing feedback regarding my issues with NDR.
I tried changing the TFA to what the user manual suggested (1000) prior to requesting assistance, so that was not the cause of the warnings.
However, after reading through the warnings, I noticed they mentioned missing data for some cells. Although I originally used the raster calculator to give cells value, my rasters still had “NoData” cells that I did not notice. Therefore, I went into my data and manually changed all the rasters’ “NoData” cells to -9999.
When I ran NDR again with a TFA of 1000 and the re-formatted rasters, the model ran perfectly without warnings and output meaningful results!
I thought I should let you know what I did to fix the warnings in case any other users run into the same issue.
Thank you so much for the time you took to answer my question. Have a wonderful evening!
Best,
Amanda
Thanks!
Amanda Riggs
MS Geography Candidate
East Carolina University
That’s great that you were able to fix the warinings @ARiggs ! To make it easier for folks to find this issue and your solution in the future, could you share the specific text of the warning?
button