TheCorrelationDictionary.pkl is a Python pickle that contains the data used to generate
the figures in Olsen et al 2024b. It contains a single variable that can be read using:

with open('HCl_data_JGR_Olsen_2024b.pkl','rb') as pkl:
   corrdict = pickle.load(pkl)

The data is also stored as an hdf5 file HCl_data_JGR_Olsen_2024b.hdf5 with in identical
structure of nested keys. 

This variable is a dictionary who's keys correspond to the correlation table in Fig. 1.
For each key, a second, a second key can be used to identify what is being compared.
e.g., corrdict['HCl']['dust'] contains the data for the HCl-dust comparison.
The main keys in the comparison are then:
 'Xvar': Definition of X variable, 'MCS dust (nearest profile)', in this example
 'Yvar': Definition of Y variable 'MIR HCl', in this example
 'X': x-values 
 'Y': y-values
 'Z': corresponding altitudes
 'Ls': corresponding solar longitudes 
 'LT': corresponding local times
 'lat': corresponding latitudes
 'Lsdiff': Differences in Ls between Xvar and Yvar
 'dist': corresponding distances between Xvar and Yvar 
 'Pearson': Pearson correlation coefficient between Xvar and Yvar
For MCS comparisons, thee sets of data are included: the nearest profile, all co-located 
profiles (indicated with _all in the key), and the mean profile of the co-located MCS data

Note that the dictionary does not contain redundant information.
e.g., if corrdict['HCl']['dust'] is complete, then corrdict['dust']['HCl'] will be empty.

Note, also, that occationally the Ls range or altitude range is restricted before computing 
the Pearson correlation coefficient is computed. This is also contained here, usually 
named 'Pearson_lowZ', and will bear the value given in Fig. 1.