
-----------------------------------------------------
Update: August 31, 2006
Add changes so that all data from WPS can be processed
	Only file effected is sr/ripdp_wrf.f
-----------------------------------------------------
-----------------------------------------------------

Bugs fixed
==============================

- fixed missing k-loop in brnshr.f
- fixed problem in capecalc3d when no EL exists
- fixed problems with non-unique lat/lon variables in calls to locinterp 
- fixed problem in dBZ calculation for low dBZ values
- fixed wrong title of "vageo" field
- fixed wrong units of "clg" field
- fixed problem in pseudoadiabat lookup table when p < min value of p in table
- fixed problem with direction labels ("N", "NE", "E", "SE", etc.) in
  cross sections

Improvements
==============================

- made improvements to make RIP more robust for arbitrarily large domains

   - removed hardwired dimensions in smooth.f
   - increased values of niam, ncs for area-filling subroutine calls
   - removed hard-wired work array from smooth.f.  Now it requires a work array
     to be passed in, which can be dynamically allocated.
   - changed definition of nscrs (the horizontal dimension of a vertical
     cross section) so that it does not get outrageously large, and
     correspondingly limitted the dimension mabpl
   - defined parameter iwrklen in hcondraw and vcondraw, for dimensioning of
     area-fill workspace.  It is now set to a much higher value than the
     previously set dimensions.
   - increased hardwired dimensions in v5df.h
   - increased the axis label character string to 5 characters in hticdraw
     and vticdraw.f

- upgraded ripdp_wrf to handle idealized WRF model output

   - set nproj (map projection) to 0, which indicates idealized model output
     (i.e., no map projection), but RIP treats this as Mercator (with some bogus
     map/domain parameters), just to make sure everything goes smoothly if user
     accidentally asks for a map.
   - set initial time to 00 UTC 1 January 1940, just to make RIP happy

- upgraded ripdp_wrf to handle model input files

   - added new code to calculate pressure, which is not available as a 3-d
     array in the WRf input file

- upgraded ripdp_wrf to handle WRF version 2+ output with proper
  map definition parameters

- upgraded ripdp_wrf to produce improved, multi-line .minfo file, and rip
  can print out however many minfo lines are available.  Max number of
  lines printed on plot can be set by user with nmin=N

- upgraded ripdp_wrf and rip to handle nonzero true latitude with Mercator
  projection

- incorporated cell-filling option (in addition to the standard area-filling)
  for color-filled contours

   - made necessary changes in cpmpxy.f and fy.f to accomodate cell filling
   - created new integer array ipslab, required to hold color indices for cells 

- improved domain rotation ("rota") functionality

   - cleaned up and simplified code
      - removed "rota" stuff from routine maptform
      - created routines rotpslab and rotipslab to handle repeated task of
        rotating pslab arrays
   - changed "rrota" to "irota" (integer) to avoid any exactness errors
   - set up lat/lon labeling to work with "rota" now
   - made "rota" work for all horizontal plot types (exception still remains that
     rotation of the map background only works for polar stereographic
     projections)
   - added temporary arrays pslabt and ipslabt, needed for rotating data

- incorporated a smoothed surface pressure field (sfpsm array) for SLP reduction
  and below-ground interpolation--seems to improve results 

- added new fields:

   - k-index (feld=k-index)
   - geostrophic absolute momentum (feld=amtg)
   - d(theta_e)/dp (feld=stbe)
   - horizontal vorticity components (feld=vox, feld=voy)
   - some minor changes/clarifications of "precipitable water" (pcp...)
     and "column-integrated" (int...) fields
   - maximum reflectivity in a column (feld=maxdbzX, where X is either c, v
     or omitted - see RIP User's Guide)
   - surface dew point temperature (feld=tdsff, tdsfk, tdsfc), computed using 
     2 m water vapor
   - surface temperature calculation using 2 m output (feld=tsfc,tsfk,tsff)
   - horizontal bulk wind shear between surface and up to 6 km (feld=bshXk,
     where X is the kilometer, 1 to 6)

- created "vertical bar" (feld=vbar), which can be drawn at a specified location
  in vertical cross sections

- added dash capability for boxes and lines (ptyp=box or ptyp=line)

- upgraded wind barb drawing so that each barb is handled on an individual
  basis with regard to northern/southern hemisphere protocol for which
  side of the arrow the barbs are on.  Previously the entire plot was
  treated as either N or S hemisphere.

- added 'm' (Temperature) as another choice for vertical coordinate

- combined scondraw.f and scondraw_polar.f into one routine (with a polar
  flag).  Same for svecdraw.f.

- reworked routine frtitle.f to make it more sensible for the various possible
  choices of information to be plotted in the main frame title

- added function lennonblank.f, which replaced redundant code in several
  locations

- made it so that contour labels and hi/lo labels are always the same size,
  instead of being dependent on the size of the viewport.

- removed several routines from ripdp_mm5 and ripdp_wrf that also existed
  as stand-alone in the RIP source code, to remove redundancy.  (e.g.,
  mconvert, writefile, fillarray, xtodot, premaptform, maptform, virtual).
  The Makefile now links the necessary routines to the ripdp programs.

- added height scale to polar skew-T background.

Other changes
==================================

- made several additions/changes to top-level and source-level Makefiles

- made slight changes to text size and line spacing for various text lines in the
  RIP plotting frame (you'll notice characters are slightly smaller, but other
  than that, little difference)

- removed "flight track" code

- made it so that default indicator values of true latitudes (truelat=91.) are
  no longer carried in rip.  ripdp_mm5 passes the actual default values.

- changed specification of vertical level "from bottom" to "fb".  For example,
  if you want the lowermost model level, now use "vcor=s; levs=1fb", whereas
  before it was "vcor=s; levs=b1".

- removed remnant references to "fullsigma" and "halfsigma" from ripdp_mm5.f

- grabbed updated versions of v5d.h, v5d.c, and vis5d.h

- switched to using dashsb exclusively for drawing dashed lines/curves
  (previously, dashdc was used in one or two places).

- Added new colors to the color.tbl. These color mimic the standard NWS NEXRAD radar plot

- Add a new example of using "feld=dbzv", with colors mentioned above to rip_sample.in

- Add a new directory called "custom_maps", where users must put their custom
  map definitions if they want to use them. A sample of how such a map should look like has
  been provided in this directory. In addition hmapdraw.f has been modified to look for
  the custom_maps in the ${RIP_ROOT}/custom_maps directory.

- New routine "dbzcalc.f" added. This is an update to the old "dbz" routine.

- Improvements added to the capecalc3d.f routine.

- Add an fx function to eliminate vector plotting problems on some LINUX computers. 
  Thanks to Bob Conzemius of Colorado State University to reporting this.


RIP Users' Guide (Doc/ripug.htm)
============================================

- Upgraded the Users' Guide to reflect all of the changes herein









In the src directory:
=====================
* dbzcalc.f       (New routine, essentially does the same thing as the 'dbz'
code that was in fields.f, with some minor modifications.)
* fields.f        (New code for feld=dbz, and feld=vor3d.)
* hmapdraw.f      (changed so that it looks for custom maps in the
${RIP_ROOT}/custom_maps directory.)
* capecalc3d.f    (changed as per discussions with Jim Bresch.)

