Water scarcity problem - Missing sub watersheds output

Hi everyone, I would like please to ask for your help.
I am new with the software and for 
my thesis I
am running Water Yield model with water scarcity.

In the output folder, for water scarcity results, I can just find a shape for the watershed, there is no results per sub watersheds!

So I see
just one value for the entire area and can not compare between the

Does anybody know what may be the reason and how can I solve this?

Thanks for your time! Best regards, María


  • RichRich Administrator, NatCap Staff
    Hi Maria, did you provide a subwatersheds input along with the watersheds to run the model?  If so, could you attach the output log of the run here so we can help you debug further?
  • Yes Rich, I provided a subwatersheds input. I attached here the output log. 
    Thanks a lot for your help here! 
  • RichRich Administrator, NatCap Staff
    Thanks for the log.  I see you're running InVEST 2.5.6.  I suspect the missing subwatershed it might have been a bug in that version.  But it's been almost 4 years so it's hard to remember.  :) I'd recommend you upgrade to InVEST 3.3.3 and use the SDR model since we deprecated the sedimentation model some time ago.    

    But, I understand you might *need* to use 2.5.6, you can run the model again using the subwatershed shapefile as the watershed file.  The model only uses those files to aggregate output on the end, so you'll happily get the same result.  Note you'll have to introduce a "ws_id" field into the subwatershed shapefile to get the mechanics to work.  

    And here's a link to 3.3.3 if you do decide to upgrade (we recommend it!) http://data.naturalcapitalproject.org/invest-releases/3.3.3/InVEST_3.3.3_x86_Setup.exe ;
  • Dear Rich, thanks for your help. I tried using the subwatershed shape in the watershed space but the model gave me error (I attached here the log). I introduced "ws_id" instead if "subws_id", also creating a different column as "ws_id" but with both ways, an error is happening.

    I am thinking if a solution could be to run separately each subwatershed as a watershed, but then I will not know what to introduce as subwatershed. Will it be like the same shape files in both parts?

    Thanks again for your time here, considering this old version issue :)

  • swolnyswolny Member, NatCap Staff
    Hi Maria -

    You can probably create your own subwatershed output from the results that you have. Just use a GIS tool like Zonal Statistics to calculate the sum of  the water yield output raster (or whatever output you're interested in - precipitation/AET/etc) within your subwatersheds. This tool is available in both QGIS and ArcGIS.

    ~ Stacie

  • RichRich Administrator, NatCap Staff
    Hi Maria.  I sort of remember this model had a threshold table?  That was indexed by watershed id?  Does that sound right?  If so, I'd wonder if the watershed file you made from subwatersheds has an ID of 0 but no corresponding entry in the threshold table?
  • Dear Rich, the model is not asking me a threshold table. It requires me to enter only: 
    1. Root restricting layer depth (required). 
    1. Precipitation (required). 
    1. Plant Available Water Content (required). 
    1. Average Annual Reference Evapotranspiration (required). 
    1. Land use/land cover (required). 
    1. Watersheds (required). 
    1. Sub-watersheds (required).
    1. Biophysical Table (required). A table of land use/land cover (LULC) classes, containing data on biophysical coefficients used in this tool. NOTE: these data are attributes of each LULC class rather than attributes of individual cells in the raster map. Columns: lucode; LULC_desc; LULC_cat; root_depth; math:`K_c`;
    2. Demand Table (required). 
    3. Hydropower calibration table.
    I could not run the scarcity part properly in the invest 2.5.6 software BUT I could obtain the subwatersheds outputs running the invest 2.5.6 toolbox in ArcGIS. I attached this results in a excel file, comparing both ways (arcgis and invest). If you could have a look to this, I really appreciate your opinion. 

    On the other hand, my thesis supervisor and me have could not understand really well why invest asks for a value of evapotranspiration already calculated, to then calculate a potential value and then an actual value of evapotranspiration "AET". As we are having problem with this result, my professor considers is too low for the area of study, we would really appreciate if you can help us understanding this procedure that invest does.

    Again many thanks for your time and help, best regards Maria
  • RichRich Administrator, NatCap Staff
    Hi Maria, sorry sorry, I had misread and thought you were running the sediment model.  Looking back on that code it looks like it might be related to a bug we've fixed since.  We still do have a water yield model in InVEST 3.3.3 that is relatively unchanged since the 2.5.6 version.  Is there any chance you're willing to try that version?

  • Dear Rich

    After several runnings I have my outputs per subwatersheds for the Water yield model. What keeps giving me a problem, are the extreme low values of actual evapotranspiration "AET" calculated by the model.

    We want to ask you as for us it is not clear, why invest asks the user for a value of evapotranspiration
    already calculated, to then calculate a potential value and then it gives as an outcome an
    actual value of evapotranspiration "AET"? If the user is asked to calculate already the evapotranspiration and provide a raster, why is useful to calculate this again by the software?

    Thanks a lot for your time, we appreciate your help with this procedural doubts. Regards, Maria
  • RichRich Administrator, NatCap Staff
    From what I understand InVEST is asking for the *potential* evapotranspiration; something along the lines of what would evaporation be if you left an uncovered bowl of water out all day.  InVEST then calculates the actual evapotranspiration due to the habitat and other landcover on the pixel.  (See equations 1 and 2 and discussion in the user's guide: http://data.naturalcapitalproject.org/nightly-build/invest-users-guide/html/reservoirhydropowerproduction.html#the-model)

    Does that make sense for what you're seeing on your end?
Sign In or Register to comment.