This forum is shutting down! Please post new discussions at community.naturalcapitalproject.org

Water Yield Error

Hi There, can someone please take a look at my Water Yield log and tell why the resultant images appear blank?

Thank you

Comments

  • RichRich Administrator, NatCap Staff
    Hi, from your log it looks like there were a few watersheds that had some nonsensical values, but the others seem to be okay. By "blank" do you mean your images are all nodata? And which images? Could there be nonsensical values that are driving the scaling to a weird state? If your visualization scales have strange values like -1e21, could you set them to some resonable ranges to see if you could get some valid looking results?
  • HI Rich,
     so I have attached a screenshot of how the result looks like with a lot of nonsensical values.
    1514 x 921 - 277K
  • RichRich Administrator, NatCap Staff
    That looks like a lot of nodata values that might have juuust rounded off so they don't match the real nodata values. Can you dropbox me your datastack to richsharp@stanford.edu? I'm hoping to start building in better checks for data and maybe this is one that can help with that.
  • RichRich Administrator, NatCap Staff
    Hi @haliberry29, the file you sent me contains almost all the files I need except for the depth to root restricting layer (or at least I can't find it). Can you pass me that one too? Or maybe look at it and see if it has a bunch of NODATA values in it with the pattern you see in your screenshot?
  • RichRich Administrator, NatCap Staff

    It looks like those blank areas are associated with areas where your precipitation raster is 0. I've attached a screenshot of your precip raster w/ 0 values turned magenta.

    So this all comes back to me now about how we decided to handle 0 precip when we were developing the model. Right now we set any areas with 0 precip to have nodata water yields. We could change that to report "0" yield but back in the day we decided it was nonsensical to have "0" precip so we did something extreme like this so we could spot it in the results if it was an issue.

    What do you think? Some of the regions where you have 0 in the prcip raster I'd expect to get some rain. Is that right? Are you intending to have 0 precip there? If so, would it be useful for us to report 0 yield instead of nodata in the cases where precip = 0?

    721 x 401 - 31K
  • there should be data in place of 0 rainfall and the missing values will affect the result of the model run. So, I will try to get better data without 0.


  •  I think the 0 values should reflect 0 instead of the no-data because the no data makes it hard to read the result from the model.
Sign In or Register to comment.