Thanks a lot for this extremely nice overview of airGR. I have a question: Is it possible to have multiple hydrological response units (HRUs) as GR4Js in a catchment using your package? Do you have an example code? Thanks a lot in advance!
Usually, GR4J is used either in a lumped way, i.e. on a total catchment, or in a semi-distributed way, i.e. on sub-catchments, with GR4J models flowing on downstream GR4J models. For the second option, we usually use streamflow stations to mesh the total catchment and the measured data at this station to calibrate the model. Using GR4J on HRUs is theoretically possible, if the HRUs respect the flowing directions, even if we have never done it. The only issue is that to do so, a method to estimate the GR4J parameters on ungauged HRUs would be necessary. There is plenty of litterature on this, but we do not provide the regionalisation tool. Finally, please note that the airGRiwrm package, which is an extension of airGR, should be prefered as it automatizes the use of GR4J models in a semi-distributed way.
Thanks a lot for this. Very informative. I am however not able to run the airGRteaching GUI, i get a shiny app but every option seemx to be greyed out. Also the first command throws the following error: PREP
GR models are continuous hydrological model; no missing data in precipitation and PET times series are allowed. Please note thBasinObs column names respect the following order: DatesR: dates in the POSIXt format [NO MISSING TIME STEPS] P: average precipitation [mm/time step] [NO MISSING VALUES] T: catchment average air temperature [℃] [OPTIONAL] E: catchment average potential evapotranspiration [mm/time step] [NO MISSING VALUES] Qmm: outlet discharge [mm/time step]
![2022 Session 8 "Long Short-Term Memory networks for rainfall-runoff modelling" [Frederik Kratzert]](http://i.ytimg.com/vi/7gDPVa74Flk/mqdefault.jpg)
![2022 Session 8 "Long Short-Term Memory networks for rainfall-runoff modelling" [Frederik Kratzert]](/img/tr.png)
![2022 Session 2 "Improving Reproducibility and Efficiency - Use of R in Hydrology"[T. Recknagel]](http://i.ytimg.com/vi/IfARakdzpR0/mqdefault.jpg)
Thanks a lot for this extremely nice overview of airGR. I have a question: Is it possible to have multiple hydrological response units (HRUs) as GR4Js in a catchment using your package? Do you have an example code? Thanks a lot in advance!
Usually, GR4J is used either in a lumped way, i.e. on a total catchment, or in a semi-distributed way, i.e. on sub-catchments, with GR4J models flowing on downstream GR4J models. For the second option, we usually use streamflow stations to mesh the total catchment and the measured data at this station to calibrate the model. Using GR4J on HRUs is theoretically possible, if the HRUs respect the flowing directions, even if we have never done it. The only issue is that to do so, a method to estimate the GR4J parameters on ungauged HRUs would be necessary. There is plenty of litterature on this, but we do not provide the regionalisation tool.
Finally, please note that the airGRiwrm package, which is an extension of airGR, should be prefered as it automatizes the use of GR4J models in a semi-distributed way.
Thanks a lot for this. Very informative. I am however not able to run the airGRteaching GUI, i get a shiny app but every option seemx to be greyed out. Also the first command throws the following error:
PREP
GR models are continuous hydrological model; no missing data in precipitation and PET times series are allowed.
Please note thBasinObs column names respect the following order:
DatesR: dates in the POSIXt format [NO MISSING TIME STEPS]
P: average precipitation [mm/time step] [NO MISSING VALUES]
T: catchment average air temperature [℃] [OPTIONAL]
E: catchment average potential evapotranspiration [mm/time step] [NO MISSING VALUES]
Qmm: outlet discharge [mm/time step]