WRF-Hydro run fails when lake_option = 1 is enabled

[Negusu]

Dear WRF-Hydro users,

I am having trouble running WRF-Hydro when I enable lake routing with:

lake_option = 1

The model runs successfully when lake routing is disabled, but it fails within a few seconds when lake_option = 1 is activated.

I tested this in both WRF-Hydro v5.2 and WRF-Hydro v5.3, and I see the same behavior in both versions.

I used the following files in my setup:

LAKEPARM.nc

Route_Link.nc

The lake is also linked in Route_Link.nc through NHDWaterbodyComID, and LAKEPARM.nc contains a lake entry with lake_id = 1.

However, when I turn on lake_option = 1, the run stops almost immediately with the error:

HYDRO_nlst namelist error in read_rt_nlst

Without lake_option, the simulation runs successfully.

I would appreciate any suggestions on what could cause this failure, or what additional settings/files I should check for level-pool lake routing.

Thank you very much.

[Arezoo RafieeiNasab]

Hi Negusu, 

The lake_option in the hydro.namelist was added in the most recent version of WRFHydro, v5.4. If you would like to use that option please upgrade to using the WRF-Hydro v5.4. That being said, the earlier version will do the same exact modeling routines for the lake_option of 1 if you just specify the path to the lakeparm file. In short, in the earlier version you do not need the lake_option in the hydro.namelist. 

Thanks!

Arezoo 

--
You received this message because you are subscribed to the Google Groups "wrf-hydro_users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to wrf-hydro_use...@ucar.edu.
To view this discussion visit https://groups.google.com/a/ucar.edu/d/msgid/wrf-hydro_users/579c313e-0c55-4e11-8776-60651c3815e5n%40ucar.edu.

--

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

My working day may not be your working day. Please do not feel obliged to reply to this email outside of your normal working hours.

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

Arezoo Rafieei Nasab, Ph.D.

Scientist V

NSF NCAR Research Applications Laboratory

are...@ucar.edu

[Negusu Tarekegn]

Hi Arezoo,

Thank you very much for your clarification.

Based on your explanation, I removed `lake_option` from my earlier version of the setup and kept only the path to `LAKEPARM.nc`.

However, I am still not seeing any lake-related output, even though I set:

`outlake = 1`

I expected to see files of the form:

`YYYYMMDDHHMM.LAKEOUT_DOMAIN{X}`

At the moment, the model runs successfully without the `lake_option` setting, but no `LAKEOUT` files are produced.

The lake is present in the geospatial input files, including `LAKEPARM.nc`, `Fulldom_hires.nc`, and `Route_Link.nc`. Since the lake information is already represented in these inputs, I am wondering what the possible reason might be for the absence of lake output.

My current output settings are:

* `CHRTOUT_DOMAIN = 1`
* `CHANOBS_DOMAIN = 1`
* `CHRTOUT_GRID = 1`
* `LSMOUT_DOMAIN = 1`
* `RTOUT_DOMAIN = 1`
* `output_gw = 1`
* `outlake = 1`
* `frxst_pts_out = 1`

Do you know what might cause `LAKEOUT_DOMAIN` output not to be written in this case?

Thank you again for your help.

Best regards,
Negusu

To view this discussion visit https://groups.google.com/a/ucar.edu/d/msgid/wrf-hydro_users/CALWrc65esRz%2BcDtBLBnehBsKj%2B8RSjJCZmSD1ey3%3DaWDHBiddg%40mail.gmail.com.

[Arezoo RafieeiNasab]

Hi Negusu

From your email chain it seems that you are running with channel option set to 2 (MC), correct? If that is the case, then the lakes are identified on the Routelink. Could you double check the content of the variable NHDWaterbodyComID and make sure it matches the feature_id in the LAKEPARM.nc file? In other words, if you have a lake with the ID of 2 on the Routelink, you need to have a feature_id of 2 in the LAKEPARM.nc file. Would you be able to share your domain folder and the namelists you are using with us? I could take a look and try it on my end. 

Thanks!

Arezoo

To view this discussion visit https://groups.google.com/a/ucar.edu/d/msgid/wrf-hydro_users/CACGuQy4fXWRUZ0xqU-b80yv8LmLnGPYhGSK2ZR90usWnVLtarw%40mail.gmail.com.

[Negusu Tarekegn]

Hi Arezoo,

Yes, I am using channel option = 2 (MC). I have attached the required files for your review. I am also trying to identify the issue on my end, as you suggested.

Thank you for your support.

Negusu 

 Domain_Files.tar.xz

To view this discussion visit https://groups.google.com/a/ucar.edu/d/msgid/wrf-hydro_users/CALWrc67eRaTW2-DkSXO%3D_Xp3uZ_7iJj6ZJkaa2LcU10tiSH5Nw%40mail.gmail.com.

[Arezoo RafieeiNasab]

Hi Negusu, 

I think the LINKID in the Fulldom should match the link in the Routelink, and it seems some of the ones in the Routelink does not exists in the Fulldom file. I displayed the LINKID in arcmap and focused on the lake outlet (there should be only 1 outlet for a lake in WRF-Hydro model), and it seems the link 4647 drains into the link 4694 and then 4694 drains to 4703 that does not exists on the Fuldom LINKID. The other branch of the lake is also seems to drain to the link 4703 which does not exist on the Fuldom file. I am not fully sure if this is the problem, but might be worth checking into it. I will check with others in case they would know the problem. 

Thanks!

Arezoo

To view this discussion visit https://groups.google.com/a/ucar.edu/d/msgid/wrf-hydro_users/CACGuQy4V4E5eQTZm0Y8nNgKPT%3Du9ptKcMjoFmCqF5Ws2-TFEpw%40mail.gmail.com.

[Arezoo RafieeiNasab]

Hi Negusu, 

Did you create the Routelink you are using based on the Fulldom.nc file you are usine? Is there a chance that the Routelink you are using was generated for a different Fuldom, or has been tweaked after running GIS pre-processing tool? 

Thanks!

Arezoo

[Negusu Tarekegn]

Hi Arezoo,

Thank you very much for your reply and for checking this.

Yes, I generated the Route_Link file from the same Fulldom.nc currently used in my simulation, using the WRF-Hydro GIS Preprocessor v5.2.0, and I did not modify it afterward.

I have also noticed multiple outlet candidates around the lake outlet area. Increasing the stream threshold reduced them, but did not remove them completely.

One additional point is that, after preprocessing, I did not get a zipped output folder. Instead, the required files were written directly into the ‘scratchdir’. I also get the Lake_Problems.csv file that mentions potentially multiple outlets.

Thank you again for your guidance.

Best regards,
Negusu

To view this discussion visit https://groups.google.com/a/ucar.edu/d/msgid/wrf-hydro_users/CALWrc64oLcqE_w8vNcs2ZGgH_pZDqC%2BTMu6yfQtfBa%3Deb%2Bmucw%40mail.gmail.com.

[Arezoo RafieeiNasab]

Hi Negusu, 

I ran your problem and reply with our GIS expert and below is his response. 

That sounds like an error with the pre-processing tools. I would suggest rerunning the tool and making sure that you specify a zip file in the Output Zip File path. This is a common error where users specify a directory rather than a zipped file. If you were running the tool multiple times, there could be multiple versions of the data in the scratch directory. I guess is that when you pulled data from the scratch directory, you pulled data from two different runs.  You need to get a zipped output file to make sure your data is correct.  

Could you please clean up the scratch directory and request a zip file and re-do the pre-processing once again? Hopefully this helps!

Thanks!

Arezoo

To view this discussion visit https://groups.google.com/a/ucar.edu/d/msgid/wrf-hydro_users/CACGuQy47KQf85%3D5dCi4ZoeU%3DK6mYnmG2w_VUyLBNSNFh8L1y5Q%40mail.gmail.com.