Studies on the Impacts of 3D-VAR Assimilation of Satellite Observations on the Simulation of Monsoon Depressions over India - Data Assimilation for Atmospheric, Oceanic and Hydrologic Applications - page 650

a b c

Fig. 26.6 Spatial distribution of forecast impact (FI) for ( a ) MODIS, ( b )ATOVS,( c )SSM/I
calculated against quikscat observations. for 19-22 September 2006 depression

corresponding to the observation location. The improvements if any, in the results
of the assimilation experiments are established by comparing its RMSE values
with that of the CTRL run. At 950 hPa level, the RMSE value of the wind speed
is found to be higher for all the assimilation experiment when compared to the
CTRL run for the forecast valid on 20 September 2006 12 UTC. The ATOVS
experiment gives lower RMSE values for wind speed at 920, 850, 780, 700, 500
and 300 hPa levels while the MODIS experiment gives lower RMSE at 850, 780,
700, 620 and 300 hPa levels when compared to CTRL run. The RMSE values of the
SSM/I experiment is found to be higher than that of the CTRL run and the other
two experiments in almost all the levels for 20 September 2006 12 UTC. On 21
September 2006 12 UTC, it can be seen that the RMSE of the wind speed of the
assimilated experiments have reduced in almost all the levels when compared to
the CTRL run. On the third day of the forecast (22 September 2006 12 UTC), the
wind speed simulated by the MODIS experiment has exhibited lower RMSE values
up to 700 hPa level from the surface. However, above 700 hPa level, the RMSE of
the wind speed rapidly increases and persists till the upper levels of the troposphere
when compared to the CTRL experiment. The rms error values of the wind speed
are lower for the ATOVS experiment when compared to the CTRL run in the lower
and upper levels of the troposphere for all the three mentioned times of the forecast.
For all the three times the SSM/I experiment shows higher RMSE values of wind
speed as compared to the CTRL run at almost all the levels. Though the results of
the MODIS and ATOVS runs did not yield lower RMSE values for all the times and
at all the levels, they do show a consistent reduction in rmse of the wind speed for
most of the times, thus showing some positive impact due to assimilation.

Rainfall

Due to its convective nature and complex interactions with the terrain features
and vegetations it is difficult to forecast the precipitation fields very precisely.