RH much greater than 100%

[twhite3]

While running WRF v4.5.2 with the electrification module (WRF-ELEC), I’ve come across something strange. During a period of lightning activity during a lake-effect snow event, the model output shows dewpoint temperatures significantly higher than the actual temperature, with relative humidity values around 113%. Other than this, the simulation seemed to have run smoothly with no other odd or impossible values.

I’m not sure why this is happening, and I don’t believe the electrification module should be affecting these thermodynamic variables. For reference, I’m using the NSSL 2-moment microphysics scheme, RRTMG for both shortwave and longwave radiation, and YSU for the PBL.
Has anyone encountered this before or have ideas on what might be causing it?

Thanks in advance!

 

[Ming Chen]

Hi,

I believe this could be a bug, --- it is impossible that dewpoint temperatures can be significantly higher than the actual temperature.

Did this occur only at a single point ? Or is it common over a large area?

 

[twhite3]

I'll run a check to see if its a wide area during this timestep (and others), I only discovered this due to creating a simulated sounding at a lightning initiation location. I'd suspect its localized around the updrafts within the lake-effect band, explaining why lightning was initiated there.

 

[twhite3]

It seems RH values between 100% to 105% are extremely common around the lake-effect band, and > 105% is also found at an alarming rate at similar times, slightly less common. Would re-rerunning the simulation be the best course of action with a different version of WRF?

 

[Ming Chen]

I talked to our expert. He agreed that the result looks weird.

Please contact Ted Mansell and hopefully he can provide some help, --- we don't provide support for codes from individual user's branch.

 

[microted]

Residual supersaturation is actually a feature of the scheme, as it does explicit condensation on droplets rather than a saturation adjustment. Some amount of supersaturation is required for droplet nucleation (i.e., it is not 'impossible'). At low droplet concentration and reasonably strong updrafts, the SS can be substantial. This is related to the 'warm-cloud invigoration' effect where added CCN keep the droplet concentrations higher and the SS lower and thus produces more latent heating. The scheme has the same behavior in regular WRF. WRF-ELEC just adds the electrification interaction (and may be more recently updated than the WRF version).