The file data.json represents the data repository accompanying the publication "How to realise a homogeneous dipolar Bose gas in the roton regime" 
by Juhász et al., published in Physical Review A.

The file contains a table where each row represents results from a single run of the dipolar Gross--Pitaevskii equation (dGPE) ground state solver for 
a dipolar Bose gas near the dipolar collapse stability boundary in traps of different power-law geometry. The columns of the table contain the following 
information (as used in the publication):

p ------------- trap power-law exponent
gamma --------- trap aspect ratio
eps_dd_inv ---- inverse of the dipolar to contact interaction strength ratio
D ------------- dimensionless dipolar interaction strength
gs ------------ dimensionless contact interaction strength
psi ----------- ground state wavefunction
n2D ----------- ground state areal density in the x--y plane
ncrit --------- critical areal density for dipolar collapse in an infinite system
E ------------- dimensionless ground state energy
mu ------------ dimensionless ground state chemical potential
S ------------- dimensionless trap wall steepness parameter
nu_TF --------- nu parameter (see publication) derived from the maximum density in the Thomas--Fermi approximation
nu_max -------- nu parameter (see publication) derived from the maximum ground state density
nu_avg -------- nu parameter (see publication) derived from the average ground state density
hist_bins ----- areal density histogram bins, scaled by ncrit 
hist_vals ----- areal density histogram values, scaled by ncrit
H3 ------------ homogeneity parameter calculated using a 3% range around ncrit
H5 ------------ homogeneity parameter calculated using a 5% range around ncrit
H10 ----------- homogeneity parameter calculated using a 10% range around ncrit
grid_size ----- size of the numerical grid used in the simulation
grid_r -------- spatial coordinates of grid points along the radial direction
grid_z -------- spatial coordinates of grid points along the axial direction

Recreating figures from data:
- Figs. 2(a)--(c) can be recreated by plotting "n2D"/"ncrit" against "r" for the relevant values of "gamma" and "p",
- Figs. 2(d)--(f) can be recreated by plotting "hist_bins" against "hist_vals" for the relevant values of "gamma" and "p",
- Fig. 3(a) can be recreated by plotting "H5" against "p" for the relevant values of "gamma" and "eps_dd_inv",
- Fig. 3(b) can be recreated by plotting "H5" against "S" for the relevant values of "gamma" and "eps_dd_inv",
- Fig. 3(c) can be recreated by plotting the "p" that maximises "H5" against "gamma" for the relevant values of "eps_dd_inv",
- Fig. 3(d) can be recreated by plotting the maximum value of "H5" (for a given "gamma") against "gamma" for the relevant values of "eps_dd_inv".


Further notes:
- The dGPE is solved assuming rotational symmetry around the z-axis and mirror symmetry around z=0. As a result, the wavefunction can be represented 
using a 2D slice in the radial and axial directions (r, z), using only one half-plane along z (z<=0). Therefore, psi is a 2D array represented on a 
grid defined by "grid_size" and points "grid_r" and "grid_z".
- Data are given for two cases---a purely dipolar gas ("eps_dd_inv" = 0), and a dipolar gas with significant contact interactions 
("eps_dd_inv" = 0.333). In each case, ground states correspond to just below the stability boundary (i.e. the highest possible "D") for different 
values of "p" and "gamma".
- The .json file can be loaded using the Pandas package in Python, calling the

pandas.read_json(filepath) 

command, which returns a Pandas DataFrame with the data used in the publication.