README file for "Sea Surface Temperature Constrast in Tropical World:  Part 1 Mean State

The AM2.1 GCM code was downloaded from GFDL.  The only modifications we made were to change the vertical or horizontal resolution.
The 1D radiative transfer model RRTMG was downloaded from the web site of AER rtweb.aer.com/rrtm_frame.html
Our Matlab wrapper for this code and the Matlab code for doing the convective adjustment is included in the file RCE1D.zip
The final year of data for each of the seven control simulations are given in separate .nc files
The Matlab programs we used to analyze the data are in the file called analysiscode.zip
Netcdf files for the final year of monthly data for each control run are also included.


All the Matlab programs that were used to process the data were concatenated in a text file called allcode.txt  The use of these programs is briefly noted below.

NNotes for tworld.d directory 

gmeans.m  computes global means from the model output  This is no longer relevant!!!!

trendm.m takes the input files from the model simulations and generates some time series
  (gmstc.m is new and was copied over trendm.m to make the latest version of trendm.m)
  when you run trendm and trendmp you need to use exactly the same years, since the compositing in compp.m uses the output of both.

trendn.m  I think these are new versions of trendm.m  writes an output file called TimeSeries3''''.mat

trendnp.m  and trendp.m  the n versions fine the hottest fraction of area, not the hottest fraction of points. makes an output
          file called PrTimeSer3''''.mat that is required by plotz2.m

tspec.m uses the output of trendm.m to compute power spectra of lots of things
tspecc.m uses the output of trendm.m to compute power spectra of lots of things, Produces an unnormalized spectrum

compm.m uses the output of trendm.m to make composites,, but it needs the results of tspec.m to choose the filter window. 
  It makes plots of stuff versus lag.

trendp.m  makes vertical profile time series    trendnp.m?  uses proper area-weighted SST difference

compp.m  makes plots of the out put of trendp.m  trendnp.m?

gmeans.m  Computes global means of stuff  I think all the gmeans*.m programs require the output of a trend*.m program. 
 It does not appeaer that gmeans.m writes an output file.  It uses basic model output as input.  Runs slowly. Makes a ton
 of diagnostic time series plots.  writes no putput file.

gmstc.m  Computes the horizontal part of GMS from advection.  Writes a file called TimeSeries2_casenum.mat, same as trendm.m??

gmeanss.m  surface stuff  writes a file called casenumSurf.mat

gmeansv.m  vertical profiles of stuff  writes a file called casenumGM.mat

plotz.m  Makes vertical profile plots  reads GM.mat files that are produced by gmeansv.m
   Focuses on T, RH, omega^2, CF, IWC, LWC
plotz2.m  
  Focuses on heating rates, radiative, convective, large-scale LW, SW  vs p, z and T.  Requires output from trendnp.m and whatever 
   produdes the Stats'''.mat file

plotc.m  Makes line plots as a function of lagged composites, requires 'Composites*.mat" input that comes from compm.m, 
  which depends on trendm.m

gmplot.m  Makes line plots of global-time averages as funtion of SST, requires "TimeSeries2_*.mat" input, which comes from trendm.m, 
   I think, or gmstc.m is the latest version of trendm.m??  NOTE new version is trendn.m and writes file TimeSeries3_*.mat
   It produces an output file Stats'''.mat that is used by other programs like plotz2.m

tbin.m  uses basic model output files, makes plots of line plots versus SST composites of various things.  Does not produce 
   a data output file, maybe it should 190801.  It produces plots of Area, NCRE, LWCRE, SWCRE, GMS and LE + SH.

tbin2.m  same as tbin.m, except it saves every month individually into a file for later compositing 

tbin2p.m  same as tbin2.m, except it saves vertical profile instead of single level data

erasst.m  This is actually in cersst.d  it provides the observed comparison plot for TANGMS plot But it only uses ERA data and 
   does not combine CERES radiation data with ERA data. For that we need to combine the output, which is done by cereraplot.m
   but that does not include the GMS and LE terms.  Aack.  But the data are there in cereraplot.m to do it, so I will add it.