Description of the
AMIE-Gan objective analysis of large scale forcing data
1. Overview
The constrained variational objective
analysis approach described in Zhang and Lin 1997 and Zhang et al. 2001 was used to derive
the large-scale single-column/cloud resolving model forcing and evaluation data
set from the observational data collected during ARM MJO Investigation
Experiment (AMIE), which was conducted during October 2011 to March 2012 on Gan
Island in the Maldives. The forcing data represent an average over the analysis
domain centered at ARM mobile facility with a diameter of 300 km as shown in Figure 1.
Due to the lack of a sounding array at
Gan Island, the forcing data for AMIE-Gan was developed based on ECMWF
analyses, which were then constrained with observed surface rainfall through
the constrained variational objective analysis. Three versions of the forcing
data are provided, which were constrained by surface precipitation retrieved
from SMART-R, S-POL, and TRMM precipitation radar measurements, respectively,
to address uncertainties in surface rainfall. With the ongoing effort to
address some calibration issues with the SMART-R and S-POL in the AMIE/DYNAMO
radar community, the large-scale forcing data expect to be updated in the near
future once the improved radar precipitation data are available. However, we
expect that only the magnitude of the forcing fields will be affected, not the
timing and the vertical structures.
The data here are in both ASCII and
netCDF formats for three domains.
Standard vertical resolution (25mb)
data
There are two standard resolution
(25mb) ASCII data files for layered variables and surface variables,
respectively for each of the domain. They are:
S-POL based forcing (version v0):
gan180varanaecmwfanaspolsurfaceC1.c1.20111113.000000.dat
gan180varanaecmwfanaspollayerC1.c1.20111113.000000.dat
SMART_R based forcing (version v1):
gan180varanaecmwfanasmartrsurfaceC1.c1.20111002.000000.dat
gan180varanaecmwfanasmartrlayerC1.c1.20111002.000000.dat
SMART_R adjusted based forcing (version
v1):
gan180varanaecmwfanaadjsmartrsurfaceC1.c1.20111002.000000.dat
gan180varanaecmwfanaadjsmartrlayerC1.c1.20111002.000000.dat
TRMM based forcing (version v1):
gan180varanaecmwfanatrmmsurfaceC1.c1.20111002.000000.dat
gan180varanaecmwfanatrmmlayerC1.c1.20111002.000000.dat
These ASCII data files can be read
using following FORTRAN files
read_layer.for
read_surface.for
The netCDF files that include all the
variables contained in the two ASCII data files are also provided:
gan180varanaecmwfanaspolC1.c1.20111113.000000.cdf (S-POL based)
gan180varanaecmwfanasmartrC1.c1.20111002.000000.cdf (SMART_R based)
gan180varanaecmwfanaadjsmartrC1.c1.20111002.000000.cdf (SMART_R adjusted based) gan180varanaecmwfanatrmmC1.c1.20111002.000000.cdf
(TRMM based)
To see the quick look plots of the data
please go to:
http://www-pcmdi.llnl.gov/ARM/scm-forcing/gan-amie/html/preview_amie.html
2. Some details of the analysis
The objective analysis domains used for
analyzing the AMIE-GAN data are shown in Figure 1. The domain-averaged
surface and TOA radiative fluxes and surface heat fluxes constraints required
by the variational analysis were obtained from ECMWF analysis data.
TRMM precipitation data is based on
3B42 data. We use the RATE_ZH for precipitation retrieval from SPOL data. This
is just one of many rates that S-Pol is able to calculate, and there is some
active discussion going on concerning which rate is the "best" rate
to use for the AMIE campaign. Due to a blockage issue, the areal averaged rain
rate applies only to about one-third of the S-Pol domain, from about 38 thru
158 degrees, and for a radar range of 17 to 100 km. SMART-R radar retrieved
rain rate is also only using the 180 deg sector from 338 to 158 deg because of
blockage to the west. If future improvement is made on these retrievals,
forcing dataset may change accordingly.
3.
Notes on Newly Release of Version 1
In this new
release,
Time period
is extended to cover both 1st and 2nd MJO (00Z 10/02/2011
– 21Z 12/31/2011)
(Updated S-Pol
precipitation is still in development, thus forcing based on S-Pol is not
updated in this release, version 0 is kept for forcing based on S-Pol covers
00Z 11/13/2011 – 21Z 12/13/2011 )
PSmartR_adjusted = PSmartR * PTrmm_150km_domain/PTrmm_SmartR_domain By doing this
the adjusted SMART-R precipitation rate retains variability mostly from TRMM precipitation
rate, however retains the magnitude mostly from the original SMART-R
precipitate rate.
The resulted new precipitation rate and omega field
are shown in Figure 2 and 3.
4.
References
Zhang, M. H., and J. L. Lin (1997), Constrained
variational analysis of sounding data bases on column-integrated budgets of
mass, heat, moisture, and momentum: Approach and application to ARM
measurements.,
J. Atmos. Sci., 54, 1503-1524.
Zhang, M. H., J. L. Lin, R. T.
Cederwall, J. J. Yio, and S. C. Xie (2001), Objective analysis of
ARM IOP Data: Method and sensitivity, Mon. Weather Rev., 129,
295-311.
AMIE Homepage: http://campaign.arm.gov/amie/
AMIE forcing data page on LLNL ARM Team
page: http://www-pcmdi.llnl.gov/ARM/scm-forcing/gan-amie/html/preview_amie.html
5. Acknowledgement
We would like to thank Bob Rilling,
Steve Williams (NCAR), Courtney Schumarcher and Aaron Funk (TAMU), Chuck Long
(PNNL) and Chidong Zhang (FSU) and all AMIE participants who collected and
provided the valuable field data for this study.
6. Contacts
For questions, or to report data problems,
please contact:
Shaocheng
Xie: xie2@llnl.gov
Renata McCoy: mccoy20@llnl.gov
Yunyan Zhang: zhang25@llnl.gov
Figure 1: Analysis domain for AMIE-Gan,
with diameters of 300 km. The red triangles denote sounding locations.
Figure 2: Precipitation Rate for 1st and 2nd
MJO. Notice the double-peak
behavior in SMART-R becomes smaller in the adjusted SMART-R precipitation rate.
Figure
3: Omega field based on different precipitation rates from TRMM, SMART-R and
adjusted SMART-R during 1st and 2nd MJO periods.