Seasonal Water Yield - Rain events vs Rain days

JessieWellsJessieWells Member
edited March 2017 in Freshwater Models
I'm just trying to make sure I understand each calculation in the model, and I'm slightly confused by the descriptions accompanying the first equation (not by the equation itself)

I'm looking at the documentation for the latest nightly build

Under 'How it Works'
The first variables are described as:

ai,m= [Pi,m/ni,m] /25.4 and
ai,m (mm) is the mean rain depth on a rainy day at pixel i on month m [mm].
ni,m is the number of events at pixel i in month m [-]
Pi,m is the monthly precipitation for pixel i at month m [mm].

But isn't ai,m just converting from P/n (in mm per event), into inches?

If so, then wouldn't ai,m be in inches, and be defined as "the mean rain depth per event, in inches" ?
(The results for QFi,m in the next equation convert back into mm, by multiplying the calculations by 25.4 mm/inch)

Rain events vs Rain days:
For it to also be the "mean rain depth on a rainy day", there can only be 1 event per day. This would mean that ni,m is the number of events, but also the number of rainy days.
Am I correct that the model makes this assumption? (1 rain event = 1 day on which rain occurred)
For places such as the wet tropics where rain events can happen more than once per day, it would be helpful to make this explicit, so we know to convert from N events to N days containing events. (otherwise subdaily data such as TRMM could give multiple events in one day)
Similarly, in the Data needs section, the 'Rain events' table, you could change from "A rain event is defined as >0.1mm." to something like "A rain event is defined as a day with >0.1 mm rain". (So ni,m = the number of days on which >0.1mm fell, for pixel i in month m).

Best wishes
- Jessie
Post edited by JessieWells on


  • RichRich Administrator, NatCap Staff
    Hi Jessie!  Yes, I think you're right about the units of ai,m.  I'll tentatively change them in the doc and wait to hear if Perrine yells at me about it.  I'll talk to Perrine, but might make sense for readability to keep all the basic values in mm then do conversions directly in the larger equations...

    And sorry, I'm not sure about your rain events vs. rain days question since I'm not familiar with the original sources.  But what you're saying makes sense.  I'll let Perrine comment on this directly. 

    p.s. sorry for the delay, and maybe expect a little bit more delay.  we just finished our large annual meeting here at Stanford.  It's been a busy week and prep leading up to this week.
  • PerrinePerrine Moderator, NatCap Staff
    Hi Jessie and Rich, 

    Absolutely right about the units of ai,m. We are actually submitting a paper describing the model in details and I had just found out about this typo last week: thanks for fixing it, Rich!

    Re: the rain events, you are correct, we usually equate event precipitation to daily precipitation. The User's guide terminology is not always consistent because of this approximation. Note that the theory is developed based on the distribution of events over the month, which means that if a user has information on the number of events per month (instead of rainy days per month), this number should be used.
    I will add a note to clarify this in the User's guide.

    Thank you!

Sign In or Register to comment.