Source code for objective trough/jet in gridded data sets.

 

Info.

 

After several requests, I’m putting the source code for the objective trough/jet lines on the internet to allow anyone interested to have a look or even use it.

The code is written in Fortran 90 and because of my computing skills (or lack thereof) it is quite long and clumsy. It reads data in GEMPAK format, although it can be easily modified to read other gridded datasets (e.g. netCDF, binary files etc) with the addition of appropriate read/write subroutines and libraries. The code has been successfully compiled with pgf90 and ifort on Linux systems and Sun f90 compiler on Sun systems.

 

Because the code is unsophisticated the input data must be in cylindrical equidistant projection (regular lat/lon) due to the way in which the spatial derivatives are computed (simple centred difference). The program can use streamfunction (computed externally) or the meridional and zonal wind components and a simple smoother can be applied to reduce noise before the fields are processed.

 

The output consists of several fields including shear and curvature vorticity, advection of curvature vorticity as well as trough and jet axes in both easterly and westerly flow. The trough/jet axis fields consist of isolated zero contours surrounded by missing data (assigned the value -9999.0), if only the zero contour is plotted the axes are shown. Included in the code is a very simple tracking subroutine that will output the trough/jet locations in separate text files. This is still under development so the reliability is untested in all situations.

 

 

 

The code is commented throughout (with variable quality) and credits for parts of the code must go to Profs. Anantha Aiyyer and Gary Lackmann at North Carolina State University in Rayleigh, NC. Please feel free to use the code and improve on it – I would appreciate a quick email message if you do use my code (merely from personal interest) and a copy if you make any upgrades. If you do use this work in any refereed work I would very much appreciate an acknowledgement!

 

Credit for ideas behind this program is given to Tim Hewson (UK Met Office), Chris Thorncroft and myself (SUNY Albany). This work is discussed in our paper in Monthly Weather Review:

 

Berry, G. J., C. Thorncroft and T.Hewson (2007): African Easterly Waves during 2004 - Analysis using objective techniques Mon. Wea. Rev. 135, 1251-1267.  Copyright 2007 American Meteorological Society.

 

Source Code download.

 

Here

 

 

FAQ.

 

How is the curvature vorticity computed?

 

Thanks to Tim Hewson the answer is relatively simple. In essence, we take advantage of the fact that:

 

Relative vorticity = Shear Vorticity + Curvature vorticity.

 

Which, rearranged gives:

 

Curvature Vorticity = Relative Vorticity – Shear Vorticity.

 

Relative vorticity can be computed directly from the NWP model meridional (v) and zonal (u) wind components:

 


 

The computation of shear vorticity is more involved, but essentially boils down to computing components of a vector. This is best described schematically. For an arbitrary wind field on a regular grid (grid spacing of Delta x and Delta y), the computation at each grid point ‘n’ uses the four surrounding grid points A, B, C and D:

 


 

The red vector shows the wind for the grid point being computed and the blue vectors are the winds at the surrounding points, which are used in the computation. The grey x’s are other grid points not used.

 

For the computation, the wind vector at point ‘n’ is used as the reference vector and the components of the wind relative to this are computed at points A-D (red dotted lines):

 

 

The components parallel to the reference vector (green vectors) are used to determine the shear vorticity:

 


 

These components are recast in grid relative coordinates to give the meridional and zonal components (‘vs’ and ‘us’):

 

 

From these we can the compute the shear vorticity:

 

 

 

Simple subtraction of the two fields (relative and shear vorticity) gives us curvature vorticity.

 

 

 

This material is based upon work supported by the National Science Foundation under Grant No. 0507976. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.