Strange T2 bias in newer WRF versions

[vaakos]

Your dveg option is 7.
The default is 4.
Could the problem be caused by the vegetation option?
I checked the changes from v4.2.2 to v4.3.1：Comparing release-v4.2.2...release-v4.3.1 · wrf-model/WRF
A snow related scheme and a irrigation scheme were improved.

Thank you for the recommendations.

I ran a test with the default dveg option (dveg = 4), but the cold patches still form this way.

The issue doesn't seem to be related to snow cover, as snow is not a factor in May in Central and Western Europe, and the snow cover is correctly zero throughout the domain.

I also do not use irrigation.

 

[haiqingsong]

Thank you for the recommendations.

I ran a test with the default dveg option (dveg = 4), but the cold patches still form this way.

The issue doesn't seem to be related to snow cover, as snow is not a factor in May in Central and Western Europe, and the snow cover is correctly zero throughout the domain.

I also do not use irrigation.

WRF4.2.2 seems OK

 

[meteoadriatic]

Good day,

What is your opt_snf value for NOAH-MP?
If opt_snf = 4, then try default (1) and check if that solves the issue.

 

[vaakos]

Good day,

What is your opt_snf value for NOAH-MP?
If opt_snf = 4, then try default (1) and check if that solves the issue.

Hello,

Sorry for the delayed reply. It took some time to complete the tests.

Switching from opt_snf = 4 to opt_snf = 1 seems to largely mitigate the issue. However, cold patches still remain, particularly over mountainous regions and, to a lesser extent, over plains.

I also ran a test with opt_snf = 1 and changed dveg = 7 to dveg = 4, so that all Noah-MP settings were set to their defaults. This did not lead to further improvement. In fact, the cold patches became even more pronounced.

The only change that consistently resolves the issue appears to be modifying the CBIOM value. Following a suggestion from a recent forum post (link), I changed CBIOM from its default of 0.02 to 0.0001. With this adjustment, the cold patches disappeared, even when using my original configuration (opt_snf = 4, dveg = 7).

That said, it is still unclear whether this change in CBIOM could have unintended side effects on other parts of the model output.

I have attached a figure to illustrate how these changes affect the T2 field.

 

[meteoadriatic]

I see, is there a snow cover where cold patches remain in opt_snf = 1, dveg = 7, CBIOM = 0.02 combo (second from the left) ?

If yes, I might have some further ideas; if no then not sure really.

 

[vaakos]

I see, is there a snow cover where cold patches remain in opt_snf = 1, dveg = 7, CBIOM = 0.02 combo (second from the left) ?

If yes, I might have some further ideas; if no then not sure really.

Since the simulations are for May, snow cover is limited to the highest peaks of the Alps in this Central European domain.
Therefore, the cold patches generally do not coincide with areas of snow cover.