Geoscience Reference
In-Depth Information
2005-11 are considered. In all cyclone cases, the model is initialized at 00
UTC with corresponding date mentioned in Table 1 and integrated up to 96h.
For each tropical cyclone case, two types of experiments namely control (CTL)
and assimilation (3DVAR) are conducted.
3 Data and Methodology
As the objective is to quantify the performance WRF model with and without
assimilation in operational forecasts, CTL experiments are initialized with
National Centers for Environmental Prediction (NCEP) Global Forecast System
(GFS) analysis at 1.0/0.5 degree resolution and boundary conditions driven by
corresponding GFS forecasts. To study the impact of observational assimila-
tion, operationally available PrepBufr global observations from NCEP
Atmospheric Data Project (ADP) archives are used in the study. These data
comprise land surface, marine surface, radiosonde, pilot-balloon and aircraft
reports from the Global Telecommunications System (GTS), Atmospheric
Motion Vectors (AMV) from Geostationary satellites, profiler and radar derived
winds, SSM/I oceanic winds and NESDIS satellite wind data which are
operationally collected by the NCEP and more details about the data can be
obtained from http://rda.ucar.edu/datasets/ds337.0. In the assimilation
experiments, observations are ingested (shown in Table 1 and Fig. 1) using
3D-VAR in single assimilation cycle with +/-3h assimilation window. The
domain specific background error coefficient data used for 3D-VAR experiments
are generated by running the WRF model for a 30-day period during November
2009. The model physics as optimized in Srinivas et al. (2012) for the BoB
cyclones is used in all the experiments which include Kain-Fritsch scheme for
Cumulus convection, WSM6 for microphysics, RRTM scheme for longwave
radiation, Dudhia scheme for shortwave radiation, Hong & Pan (YSU) non-
local PBL diffusion scheme and NOAH land surface model. Model results
from CTL, 3DVAR runs for cyclone track and intensity parameters are evaluated
using the India Meteorological Department (IMD) best track data and the IMD,
TRMM gridded rainfall data sets. The details of the period of each cyclone,
initial and model integration time, resolution of GFS data sets, density of
observations of each type used in assimilation experiments are provided in
Table 1.
4 Results and Discussion
Figure 2(a and b) shows the differences in the initial temperature (shown
shaded), relative humidity (shown in contour) and winds between Control and
3DVAR experiment both in horizontal (a) and vertical cross-section (b) for
one cyclone case study “Jal” at 00 UTC of 04
th
November 2010. A higher
warming of 0.6 to 1
o
C and drying up to 4% of relative humidity in the region
of cyclone is seen in the assimilation experiment. Also the difference plot shows
stronger northeasterly (~by 3 m/s) winds in the region of cyclone and the strength
Search WWH ::
Custom Search