2587 x 1430 - 828K


  • jdouglassjdouglass Administrator, NatCap Staff
    I'd guess that your biophysical table might be malformed.  Could you attach it here so we can take a look?
  • ok here is the biophysical table for the Sediment retention model. i will alos attach the one for Water Yield
  • this is the error from the water yield model
    1848 x 1419 - 792K
  • jdouglassjdouglass Administrator, NatCap Staff
    For your table (for SDR), you'll need to export your Excel file as a CSV.  All InVEST models require tables to be in the CSV format (and comma-delimited!) rather than excel files.

    For the water yield model error, the error message says that your spatial inputs don't overlap.  This happens sometimes when your inputs have different spatial references and/or projections, so a good first step would be to make sure that all of your inputs have the same projection.
  • ok, I will edit them all and try again. Thanks a lot jdouglass!
  • Hi jdouglass, the Water Yield model ran but the result came as empty. I mean, PET_mm, precip_mm, wyield_mm e.t.c. all came back as empty.
    2812 x 1580 - 2M
  • My SDR model ran for almost 5 minutes with .tiff output but could not complete the process and kept sending this error message.

    can somebody please take a look at this? I have re-edit my data several times but got the same result. I can't figure out what am doing wrong.
  • RichRich Administrator, NatCap Staff
    Hi @haliberry29, I'm guessing that there are no valid data overlapping some of the raster outputs. That could be because your DEM isn't pitfilled and the model is not able to route in regions where you may have a polygon, or it could be because you might have a watershed polygon totally outside your raster stack input.

    You could take a look at the raster output in your workspace to see if it's a routing issue, otherwise I'm happy to take a look if you can dropbox me your datastack to richsharp@stanford.edu.
  • RichRich Administrator, NatCap Staff
    Hi @haliberry29, I got your datastack but it did not include your DEM raster so I was unable to debug your issue. Could you send me your datastack again and include the DEM as well?
  • Yes Rich, right away.
  • RichRich Administrator, NatCap Staff
    Hi @haliberry29, it looks like your `Watersheds.shp` file is in an unprojected lat/lng coordinate system while all your data are in a linear projection. The InVEST models don't automatically do a reprojection, so can you project `Watersheds.shp` to the same coordinate system as your raster data and try again?
  • Hi Rich,

    for some reason, my study area which is West Africa seems only to run when I use the UTM projection and not Lambert equal area. So, I reprojected all the data and ran the model again. It gave out some result but with a lot of error message. I edited the data several times (i used the whole of today editing the data and running the model) I still got the same exact error message. Please find attached the error message from Water Yield.
  • For the SDR and NDR half of the result came out alright while half are empty for instance, sed_export.tiff, rkls.tiff

  • here's the log for NDR. In all fairness, this model gave better result than the previous model runs eventhough it has its own errors.
  • this is the link to the edited data. In all the model result only the "watershed_result" comes out with information, but the rasters are all empty.

  • RichRich Administrator, NatCap Staff
    Hi @haliberry29, I'm sorry to hear this is so much trouble...

    Looking through your NDR log, it looks like "luluc_2000_zero.tif" and "Prec_2000.tif" are both in different projections despite all your work (former is WGS_1984_UTM_Zone_33N and latter is SWA_LambertAzimuthal_Equal_Area). InVEST will need them to be in the same projection to calculate correctly. I'd expect you'd get pretty strange errors or results there. Can you ensure all the rasters are in the same projection and run it again?

    Despite all those warnings in the hydropower model, it looks as though it should have run correctly. The warnings are only associated with precision errors when writing to an ESRI Shapefile output. Do the results of the model seem otherwise okay to you?

    And you mentioned the SDR came out only "half right?" with the export missing? Or maybe it's blank? Seeing the warnings for the unroutable pixels made me wonder if the DEM you're using has hydrological pits that need filling, but the DEM you sent me seems to route okay, but it also has a different filename. "WS_DEM" vs. "DEIM_zone.tif"? If "DEIM_zone.tif" is not "WS_DEM" can you make sure to fill hydrolgical pits on it and give it another try?

Sign In or Register to comment.