Page 2 of 2

Re: Question regarding FOREST_COVERAGE

Posted: Mon May 06, 2019 10:58 am
by jrcraig
Hi Julian,

I think this is as simple as removing the square brackets and correcting your interval- the command should look like this:

Code: Select all

:TransientParameter IMPERM LANDUSE
  1980-11-01 12:00:00 0.5 4
  0.5
  1.0
  1.0
  0.5
:EndTransientParameter


Here, 4 is the number of data points in the time series, and 0.5 is the time interval, in days, i.e., 12 hours (you had 30 in your example, which implies an interval of 30 days, which is unfortunately not supported at this time, thus the warning - the largest interval is 1d). For permafrost representation, you'll likely want to set an interval of 1 day and just create a long annually repetitive time series. Note that v2.9.1 of Raven, appearing any day now - likely this week, will simplify the process of representing annually cyclical monthly time series.

Re: Question regarding FOREST_COVERAGE

Posted: Mon May 06, 2019 12:38 pm
by Julian
Thank you kindly for the correction and the heads up. I will keep my eyes open for the update! :)

If I would want to make such changes to HRU-Groups I guess it's not directly possible? So I will have to create seperate soil-classes for the Permafrost-HRU, so I can select them with the :TransientParameter -Command?

Re: Question regarding FOREST_COVERAGE

Posted: Tue May 07, 2019 7:13 am
by Julian
ERROR : CTimeSeries::Initialize: time series forcing data not available at end of model simulation

With your corrected code below I'm getting the error above. I'm not sure why it is appearing now since it was running without :TransientParameter smoothly.

Re: Question regarding FOREST_COVERAGE

Posted: Mon May 13, 2019 9:15 am
by rchlumsk
Hi Julian

On the time series error, please check that you have sufficient forcing data to fill the simulation time. That error typically indicates that there is less data than simulation duration.

Re: Question regarding FOREST_COVERAGE

Posted: Thu May 23, 2019 7:21 am
by Julian
jrcraig wrote:[...] Here, 4 is the number of data points in the time series, and 0.5 is the time interval, in days, i.e., 12 hours (you had 30 in your example, which implies an interval of 30 days, which is unfortunately not supported at this time, thus the warning - the largest interval is 1d). For permafrost representation, you'll likely want to set an interval of 1 day and just create a long annually repetitive time series. Note that v2.9.1 of Raven, appearing any day now - likely this week, will simplify the process of representing annually cyclical monthly time series.


Is this function already integrated in 2.9.1 and how would I use this?

---------------------------------

So far the HBV-EC Emulation is working great for my case, but I have one question regarding the structure of the model. In D2 of the Raven-Manual it says that SLOW_RESERVOIR is a LakeStorage and in the Raven Walkthrough #3 - PDF (HBV-EC Overview) it's the FAST_RESERVOIR. In both cases I can't find information in the literature. I wasn't familar with the Canadian variant of the HBV-Model before so probably I'm missing something. I checked Hamilton et al. 2000, Stahl et al. 2008 & Moyer et al. 2016 and can't find this modification.

Thanks in advance,
Best Julian

Re: annually repetitive time series & question regarding HBV-EC Emaluation

Posted: Thu Jun 06, 2019 10:32 am
by jrcraig
The easy solution using the most recent Raven is to use the :AnnualCycle commnad in the time series, e.g.:

Code: Select all

:TransientParameter IMPERM LANDUSE
  :AnnualCycle 0.5 0.5 0.5 1.0 1.0 1.0 1.0 1.0 0.5 0.5 0.5 0.5
:EndTransientParameter


Each of the values are monthly parameter values, which will vary based upon the :MonthlyInterpolationMethod method you use. They will be applied continuously every year regardless of the simulation duration.