This forum is shutting down! Please post new discussions at

Input for evapotranspiration (Hydropower production)

JeanJean Member
edited July 2013 in Terrestrial Models
As I'm reading the InVEST user guide for Hydropower production, I found a discrepancy regarding to which data should be used for the evapotranspiration input (Eto). The three terms "actual evapotranspiration" (as occurs in the explanation for the parameter AETxj ), "potential evapotranspiration" as occurrs later on explaining the same parameter and "reference evapotranspiration" as in the module input are used interchangeably. My colleague conducted a literature review and found they have different meanings and the data associated with each term are remarkably different from each other.
Does anyone know the data source of the Eto provided by InVEST or has experience using other evapotranspiration dataset? Thanks very much.


  • RichRich Administrator, NatCap Staff
    Hi Jean, let me ping our hydrologist about this issue.
  • bkeelerbkeeler Member
    edited July 2013
    Hello Jean, You're correct that the documentation on ET is inconsistent in the Users Guide. We are working to fix this!

    To answer your question: The correct evapotranspiration raster to use is reference ET (ETo) based on a reference crop such as grass or alfalfa. In many applications of the water yield model we use FAO reference ET data, which are available globally and represent annual average reference ET from ‘61-’90.

    The ETK or Kc coefficients in the biophysical table adjust the reference ETo values specific to the crop or vegetation type in each pixel of your land use/land cover map. For example, if your landscape is covered in forest then you need a Kc coefficient that adjusts the reference ET (based on grass or alfalfa) to represent the potential ET of forests. This is used to estimate actual ET (AET) for the watershed, one of the outputs of the water yield model.
    Post edited by bkeeler on
  • JeanJean Member
    Thanks very much guys for your prompt response. I really appreciate it.
  • Hi,
    I have another question regarding to the algorithm used to calculate Eto (as shown in p272 of the InVEST_2.5.5 document.) Can you point me to a reference of the algorithm? I looked up some papers on the modified Hargreaves's equation and found disagreements in terms of the parameters and numbers. Thanks very much
  • PerrinePerrine Moderator, NatCap Staff
    Hi Jean,
    You are right, there are several "modified" Hargreaves' equation in use in the literature. The expression used in InVEST is based on the following paper, and was found to provide good results with rather inaccurate data:
    Droogers P. and R.G. Allen. 2002. Estimating reference evapotranspiration under inaccurate data conditions. Irrigation and Drainage Systems, Volume 16, Issue 1, February 2002, Pages 33 – 45
  • Hello,
    I'm confused with regards to which data to input for the Kc factor within the model's Biophysical Table. Should the Kc values already have been multiplied with their respective ETo (as shown under the Kc calculator excel file sheet provided in the appendix) and thus represent ETks, or should they be annual averages of Kc (FAO table excel file sheet) and the model will later link them to ETo? Thank you
  • PerrinePerrine Moderator, NatCap Staff
    Hi Cecilia,
    The values of Kc in the biophysical table should be the FAO Kc value multiplied by 1000. This factor of 1000 is not the ETo, it is simply a requirement of the model code --only accepting integers.
    Apologies for the inconsistencies in the notations used in the user's guide: we are editing it to be consistent with the FAO notations (Kc), so we understand the spreadsheet is a bit confusing at the moment.
  • Hi Cecilia,
    Re the Kc factor used for InVEST, could you please send me the link to download the FAO table excel file sheet you mention in your post of November 2013?
Sign In or Register to comment.