streamflow overestimation

[Aarti Soni]

Hello everyone,

I am trying to compare daily streamflow (CHANOBS) with ground observation. For daily streamflow computation I have converted 3-hourly CHANOBS to daily time scale. As you can see in the attached figure, wrf-hydro-derived streamflow for January month is showing a continuous smooth curve. Why it is not showing any fluctuation?

Second, why wrf-hydro-derived streamflow is showing too much overestimation? Most of the time I am noticing values are very high.

* I have not yet done the calibration. 

Thank you.

[aubrey]

Hi Aarti:

Are you saying you have precipitation falling in the basin, it is rainfall and not snow accumulation, and you are not seeing a streamflow response? The first 2 things I would check are to make sure you have your stream gauge on the correct reach and to make sure the model is properly spunup. If you do have rainfall occurring in the basin above your gauged reach, you can check the water budget components to see where that water is going (e.g., soil moisture, a reservoir, groundwater).

Let us know how your checks go.

Thanks!

Aubrey

[Aarti Soni]

Hi Aubrey,

I apologize for the late response. I checked the spinup period, which is correct. Rainfall, I did not check yet. I will share rainfall plots also once it is prepared. 

I have plotted daily streamflow and compared it with observation. In the results, I have noticed three things, 

first, the streamflow magnitude in model simulation is much higher than the observation, while I have averaged over the time instead of the sum.  (the model and observed streamflow comparison plot are attached for three stations, S1, S2 & S4). 

Second, from an upstream station (S1) to a downstream station (S6), streamflow is continuously decreasing (A plot is attached daily_2006, which shows a comparison of model-simulated streamflow in all the stations). 

Third, in all the stations (S1-S6), streamflow follows the same pattern.  

How to figure out this problem? 

Thanks.