Fast bucket discharge

[Jana Corominas]

Hello everyone,

Lately, I've been executing forecast simulations of streamflow with WRF-Hydro uncoupled with WRF model. However, there is barely no rain at this time of year and my streamflow results in a very fast bucket discharge, leading to almost 0.0 streamflow values at some points after a few days. 

I would like to achieve a more constant baseflow, with slower decay, however I already have Expon value set to 1.0 (minimum). I also have Zmax=40, OVROUGHRTFAC=10, LKSATFAC=300. 
I don't want to change much the basin behaviour, since this options give me a reasonable time to peak and intensity values, but decay is just too fast. I attached an image to example this. 

Could anyone give me any tips to improve this? How can I manage to have an almost flat baseflow even in the absence of precipitation?

Thanks a lot. 

[aubrey]

Hi Jana:

Just to make sure - you are using the baseflow exponential bucket formulation, you have confirmed that this is drainage from the GW bucket, and your model is properly spunup? If you are running the exponential bucket model, it is a pretty simple equation so you might try just setting up a little offline experiment (excel, python, etc.) to test the parameter sensitivity to see if you can create the behavior you are looking for. I know we have had folks in the past play with alternative formulations - those are pretty easy to plug in if you don't think the exponential equation is working. Let us know how it goes.

Thanks!

Aubrey

[Jana Corominas]

Hi Aubrey,

Thanks for your reply. Yes, I'm using baseflow exponential bucket formulation. Model has been running for various months now, so I am confident with my spin-up time. However, when using GFS as input data there is very low (or mostly none) precipitation, which leads to fast bucket discharge. 
I'm not familiar with pass-through baseflow model, would it make sense to use? would it make a difference?

Thanks.

Jana Corominas