These .tgz are compressed tarballs of two categories: the 100 surface and the 111 surface
The 111 surface is split into thickness of N as in the paper
The 100 surface is split into charges e that are the artificial surface charges as in the paper
Each is further subdivided into field strengths, where charge.xvg and potential.xvg files are (potential is then hacked to include the applied field as presented in the paper) calculated by gromacs.
These results are extracted with simple in-house python (to manipulate as .csv type data) and plotted against the field parameter to define the points of compensating charge, and have been saved as .png files in the relevant directories. 
The capcitance*.png files were point estimates of the capacitance, they didn't make it into the paper but have been left in the repository for completeness.
Hopefully all this is the most up to date versions, this is my first dataset as a 1st year PhD student, so I'm sure I messed something up somewhere, if in doubt, what is written in the paper is definitely correct.
--Thomas E Sayer