The WRF Variational
Data Assimilation System (WRF-Var)
1. Introduction
Data assimilation
is the technique by which observations
The MMM
Division of NCAR supports a unified (global/regional, multi-model, 3/4D-Var)
model-space variational data assimilation system,
WRF-Var, for use by NCAR staff and collaborators, and
is also freely available to the general community, together with further
documentation, test results, plans etc., from the WRF-Var
web-page http://www.mmm.ucar.edu/wrf/WG4/wrfvar/wrfvar.htm.
The documentation you are reading is the "Users' Guide" for those
interested in downloaded and running the code. This text also forms the
documentation for the online tutorial. The online WRF-Var
tutorial is recommended for people who are
·
Potential users
of WRF-Var, and who want to learn how to run WRF-Var by themselves;
If you are a new
WRF-Var user, this tutorial is designed to take you
through WRF-Var-related programs step by step. If you
have familiar with 3/4D-Var systems, you may find useful information here, too,
as the WRF-Var implementation of 3/4D-Var contains a
number of unique capabilities (e.g., multiple background error models, WRF
framework-based parallelism and I/O, direct radar reflectivity assimilation).
If you don't know anything about 3D-Var, you should first read the WRF-Var tutorial presentations available from the WRF-Var web-page
Goals Of This WRF-Var
Tutorial
In this WRF-Var tutorial, you will learn how to run the various
components of the WRF-Var system. In the online
tutorial, you are supplied with a test case including the following input data:
a) observation file, b) WRF NetCDF background file
(previous forecast used as a first guess of the analysis), and c) background
error statistics (climatological estimate of errors
in the background file). In your own work, you will need to create these input
files yourselves.
The components of
the WRF-Var system are shown in blue in the sketch
below, together with their relationship with rest of the WRF system.
·
Running other,
more computationally intensive, case studies.
·
Experimenting
with some of the many namelist variables. WARNING: It
is impossible to test every code upgrade with every permutation of computer,
compiler, number of processors, case, namelist
option, etc., so please modify only those that are supported. The ones we
support are indicated in the default run script (DA_Run_WRF-Var.csh).
Good examples include withholding observations and tuning background error coefficients,
·
Running with
your own domain. Hopefully, our test cases will have prepared you (and us!) for
the variety of ways in which you may wish to run WRF-Var.
Please let us know your experiences.
As you are going
through the online tutorial, you will download program tar files and data to
your local computer, compile and run on it. Do you know what machine you are going
to use to run WRF-Var related programs? What
compilers do you have on the machine?
Running WRF-Var requires a Fortran 90 compiler. We currently support
the following platforms: IBM, PC/Linux
Tutorial Schedule
We recommend you
follow the online tutorial in the order of the sections listed below. This
tutorial does not cover parts of the larger WRF system, which are required if
you wish to go beyond the test case supplied here, e.g., the WRF Pre-processing
System (WPS) and real
The online tutorial
is broken down into the following sections.
a) Download Test
Data: This page describes how to access test datasets to run WRF-Var.
b)
c)
d)
e)
f)
g)
Once you are able
to run all these programs successfully, and have spent some time looking at the
variety of diagnostics output that is produced, we hope that you'll have some
confidence in handling the WRF-Var system programs
when you start your cases. Good luck!