[Wrf-users] 2 meter temperature problem.

[jruiz at cima.fcen.uba.ar]

Hi
We have been running WRF model every day since september 2005 over South
America and we are concerned about the behaviour of the temperature at 2m.
The amplitude of the diurnal cycle is underestimated by the model (the 72
hours evolution sometimes has the aspect of a damped oscilation which
means that the problem becames worst as the forecast time increases). (see
for example this forecast compared agains the GFS forecast and the METAR
observations
http://www.master.iag.usp.br/graf_phplot/comp_prev/comp_prev_comum.php?resp=Mostrar%20a%20figura&data_ens=0602220000&rm=-1&pontos=0&quintis=0&dias=7&est=87166&var=temp&modelos=WRFAR1%20AVNxx1&datas=%200602190000%200602190000
)
We are currently running 2 domains one with 60 km covering South America
and a smaller one (20 Km) nested over north-eastern Argentina. We are also
using the following physics options:
RRTM scheme for LW radiation.
Dudhia scheme for SW radiation.
Noah land-surface model.
Mellor_Yamada-Janjic pbl scheme.
Grell cumulus scheme.
Has somebody found the same problem?
Any hint will be appreciated.
Thanks in advance.
Regards.
Juan Ruiz.
University of Buenos Aires- Centro de Investigaciones del Mar y la
Atmósfera - CONICET/UBA.