Scheduled Downtime
On Friday 21 April 2023 @ 5pm MT, this website will be down for maintenance and expected to return online the morning of 24 April 2023 at the latest

Elevation difference in time series output (TSLIST) after restart

This post was from a previous version of the WRF&MPAS-A Support Forum. New replies have been disabled and if you have follow up questions related to this post, then please start a new thread from the forum home page.

userwrfch

Member
Dear WRF support team,

I am running WRF version 4.1.3. and I am getting time series output via TSLIST for a couple of weather stations in a mountain region. I found that the elevation in the .TS output file was very different after a restart:

Head of .TS FILE 1 (before restart):
AWS1 2016 3 1 c16_1 ( 50.823,-116.919) ( 81, 85) ( 50.819,-116.913) 2407.2 meters

Head of .TS FILE 2 (after restart):
AWS1 2016 3 1 c16_1 ( 50.823,-116.919) ( 81, 85) ( 50.817,-116.916) 2596.2 meters

I am running high-resolution runs with 5 domains for 8 days (restart after ~6 days), and the elevation difference does not occur in d05 (40 m resolution) for several weather stations, but does occur in d04 (200 m) and d03 (1 km). However, I compared the time serious output for several variables for the overlapping times after a restart, and the values are exactly the same, even for temperature (where the elevation difference should matter).

I didn't change anything, I only set .restart. = TRUE in namelist.input and modified the dates.

Can you please help me what to do? Which elevation is correct?

Thanks a lot!
 
Hi,
I tried to run a quick test to see if I could repeat the issue, but I couldn't. Do you mind attaching your namelist.wps and namelist.input (original run and a restart namelist) files, along with your tslist file so that I can set up your simulation? Thanks!
 
Thanks a lot for your help!

The issue occurred with a few different WRF runs, and I have attached the namelists of one of the runs as an example. I used SRTM 1 arc-second topography data and ERA5 reanalysis data as input. Moreover, I used "smooth_option = 1-2-1; smooth_passes=1" in GEOGRID.TBL for HGT_M.

My tslist looks as follows:

#-----------------------------------------------#
# 24 characters for name | pfx | LAT | LON |
#-----------------------------------------------#
AWS1 Conrad 2016 c16_1 50.8233 -116.9199
AWS2 Conrad 2016 c16_2 50.7822 -116.9120
 

Attachments

  • namelist.input
    4.9 KB · Views: 38
  • namelist.wps
    1 KB · Views: 27
  • namelist_restart.input
    4.9 KB · Views: 30
Hi,
After verifying what you're seeing, I did some research and found that the value you see in the initial run's *.TS file shows the elevation given in the wrfinput file for each domain. When WRF is run with nesting and feedback turned on, we average the nest grid cells that fall inside a parent grid cell. For e.g., if we had a 3:1 ratio, the 9 grid cells of topography in the area of the nest would be sent back to the center single parent grid cell to overwrite the original value, thus, slightly changing the value. Then the final value of the wrfout* and wrfrst* files may differ for the parent, than the value given during initialization. So this is expected behavior, and since all of your other values are identical, I think you can safely move forward.
 
Top