RUC vs. ETA (NAM)

[Glenn Rivers]

Has anyone noticed this list of quirks about the 2 models

RUC-

-underestimates dews when initiated in the morning
-overestimates daytime heating
-overestimates mixed layer depth
-over-estimates daytime convection
-misses many nocturnal convective events
-QPF's too low (even for 3 hour periods), esp for convective precip.
-underestimates frictional backing of surface winds

ETA (NAM)

-over-estimates dews when soil is dry and moist advection is less
important than local evapotranspiration
-almost always underestimates daytime heating unless day is cloudy
-very high CAPE forcast beyond 48 hours often does not verify.

has anyone else noticed these annoying quirks, or others that I have not mentioned?

[Jeff Snyder]

Good list of observations, and I generally agree.

Much of the precip, CAPE, etc, issues involve the use of convective parameterizations in the various models. There are a few good sites that explain this pretty well: http://www.comet.ucar.edu/nwplessons/preci...ectiveparam.htm (http://www.comet.ucar.edu/nwplessons/precipproclesson3/convectiveparam.htm)
http://www.met.tamu.edu/class/metr452/mode...convection.html (http://www.met.tamu.edu/class/metr452/models/2001/convection.html)

The HPC also has a model bias page at http://www.hpc.ncep.noaa.gov/mdlbias/biastext.shtml