ARTS/RTTOVSCATT Comparisons.

This working group met to discuss the latest intercomparisons between the
scattering ARTS and RTTOVSCATT RTMs. Comparisons were made at AMSU-B frequencies
using a Met Office Mesoscale forecast case study over the UK.

Initial clear sky results show a good agreement between the models.  For the
scattering case, however, RTTOVSCATT does not simulate the lowest brightness
temperatures that are seen in the corresponding NOAA-16 AMSU-B observations. 
ARTS shows a good agreement with the observations, although the effective radius
used in the size distributions had been tuned to achieve this.  

The funadamental difference that most likely leads to the disagreement between
the RTMs is the treatment of microphysics.  ARTS has a fixed effective radius of
100 microns; RTTOVSCATT has a variable effective radius. The depressions seen in
the brightness temperatures may be arising from a small number of large ice
crystals, and it seems that RTTOVSCATT is not getting enough large ice
particles to simulate this.  

It was decided to step back from the RT and to undertake a more detailed study
into the microphysical assumptions and size distribution parameterisations. 
Most current parameterisations focus on the smaller end of the size spectrum,
and set the distribution by the mean crystal size.  It was thought that in this
case it may be more appropriate to set the distribution by the maximum crystal size instead.  

Other differences arise from the parameterisation of surface emissivty.  ARTS uses a fixed emissivity over the ocean while RTTOVSCATT uses FASTEM-2.  However, it is thought that the differences arising from this are small compared those arising from the different microphyiscal assumptions. Future work will concentrate on size distribution parameterisations.  


Written by Una O'Keefe