NOTE: It is useful to look at the readme files in the yyyymmdd_traj directories 
to understand how the trajectories were computed and how they are organized.

There are three types of files in this directory:

CONINF_yyyymmdd files contain convective influence information along
the flight in GAINES-Hipskind ascii format.  The header is self explanatory.

The flightlevel_[broad,fine,med]_yymmdd.pdf files contain plots of convective influence
at the aircraft flight level, with each page containing plots for each
one-hour segment of the flight.  Fine area contains information along
the flight track, with large squares indicating convective influence, and
small points no convective influence.  Medium area includes the squares
of convective influence plus parts of the trajectory.  Broad area includes
the entire trajectory, including the locations of convection influencing
the flight track.

For diehard convective influence fans, the c_attre_yymmddhhmmss_eraint.. files
are idl save/restore files containing the convective influence information for
all the calculated trajectories.  Calculated trajectories are in the
yyyymmdd_traj directories.  Read the documentation in those directories for
a description of how the trajectories are organized.

Each file has the following variables.  cldlatest is the time to most recent convection,
in days for each trajectory (how many days prior to the observation time the convection
influencing the parcel occurred).  11 days means no convection, since we ran trajectories
back 10 days.  cldlatestlat, cldlatestlon, and cldlatestp are the latitude, longitude, and pressure where
the most recent intersection with convection occurs for each trajectory.  cldtimtrajtot
is the total number of days that each trajectory is influenced by convection.  Lastsatc
is the time of the latest satellite image considered in the convective influence
calculation.  Latin, lonin, and thein are
the latitude, longitude, and theta of the initial (latest in time) points in the
trajectory.  (Note that we calculated trajectory clusters for each aircraft point that
we chose to do trajectories from -- 9 points per cluster, surrounding the aircraft
sampling point).  The "sat" variables are the satellite images considered in evaluating
the convective influence (doy -- julian day; fils -- file name; tim -- yyyymmddhhmm date).
timepa(i) and timena(i) are the times, in doy, along the trajectory where we use the
ith satellite image to determine the convective field.  timindlatest is the time index
along the trajectory where the latest convective influence occurs.

CLDLATEST       FLOAT     = Array[558]
CLDLATESTLAT    FLOAT     = Array[558]
CLDLATESTLON    FLOAT     = Array[558]
CLDLATESTP      FLOAT     = Array[558]
CLDTIMTRAJTOT   FLOAT     = Array[558]
CLDTIMTRAJTOTL  FLOAT     = Array[558]
DECTIM          FLOAT     =       14.0000
DEPTH           LONG      =           24
EXPNAM          STRING    = 'attre'
LASTSATC        STRING    = '2014031415'
LATIN           FLOAT     = Array[558]
LATLL           FLOAT     =       34.4312
LATUR           FLOAT     =       35.5461
LONIN           FLOAT     = Array[558]
LONLL           FLOAT     =       240.440
LONUR           FLOAT     =       242.931
NDAYS           INT       =       10
NUMCALC         INT       =        3
RELDIST         FLOAT     =      0.250000
SATDOY          FLOAT     = Array[81]
SATFILS         STRING    = Array[81]
SATTIM          STRING    = Array[81]
THEIN           FLOAT     = Array[558]
TIMENA          FLOAT     = Array[81]
TIMEPA          FLOAT     = Array[81]
TIMINDLATEST    LONG      = Array[558]
TRAJFIL         STRING    = '/Volumes/hd4/ATTREX/lenxdr/tattre_140313384201_eraint_347_diabatic_10_55'...
YYMMDDHH        STRING    = '14031338'

Method:  Given the diabatic trajectories in the yyyymmdd_traj directories, run each trajectory to
a time varying field of cloud top theta (field of cloud top theta varies in longitude, latitude, and
time).  We have 3-hourly convective cloud top theta fields based on: (1) the 3-hourly quarter degree
TRMM/Microwave/IR rainfall information (TRMM site at NASA/GSFC); (2) 3-hourly global IR (11 micron)
brightness temperature information from geostationary satellites; and (3) ERA-Interim model level
temperature analyses.  The rainfall information is used to define convection (as opposed to extended
cold anvils).  The brightness temperature information defines the cloud top temperature.  The temperature
analyses are used to relate the temperature to altitude (and theta).  We use a simple mixing algorithm
to deal with the coldest brightness temperatures (near the cold point tropopause).  

Convective influence is said to occur if the trajectory point comes within .25 degrees of a point where
the cloud top theta is greater than or equal to the trajectory theta.