NAM WRF-NMM humidity and convection issues

This message was edited 1 time. Last update was at 26/06/2006 09:25:04

WFO Salt Lake continues to see major issues with the NAM-WRF
in regards to relative humidity/dew point/temperature depiction
over the Great Salt Lake. This was noted in the parallel NAM testing as well.

An example included demonstates the nearly saturated layer that the
NAM-WRF displays over the GSL. Based on observations over the the lake
the NAM-WRF forecast is 50 to 70 percent too high for RH. This is also
affecting KSLC (Salt Lake City) BUFR data as depicted by the stable
marine layer in the attached sounding. We have had RH issues (too high)
before with the NAM-Eta but they were inconsistent and often in the
mountains. This appears to be tied to SST temperatures or other changes
to the GSI for the NAM-WRF. A local version of the WRF-NMM shows
considerable improvement. Perhaps it is possible to incorporate AVHRR for
the GSL or even climatology?

I will send another email with additional images.

Regarding convection, overall the test period was acceptable but we have
yet to see the NAM-WRF during a true monsoon. Recent examples during
"dry" westerly flow showed that the NAM-WRF produced unrealistic rainfall
over a small area when precipitable water was less than 0.40 inches. An
attached example was observed on several consecutive runs of the
NAM-WRF from June 23-24 2006. Though high based moist convection
was observed rainfall was minimal. NAM-WRF QPF was as high as 3 inches
over a small area. Runs of the local WRF-NMM using the KF scheme
predicted no precipitation. The behavior of the NAM-WRF in this situtation
looked similar to explicit convection results using the local model.

Any comments would be appreciated. Thanks!

Alex Tardy
WFO SLC

Additional images attached.

Refer to prior email.

WFO Salt Lake
Alex Tardy