This folder contains ASCII files for the intensity colour maps in Figs. 2A-T.

The supplied MATLAB code plot_data.m plots 20 figures when run in the same directory as the unzipped data files.

To obtain the plots in the paper, use the matlab code plot_data.m. Running this will plot all 20 figure panels. 
This takes about a minute to run. Each figure panel in the paper is displayed in a separate matlab figure window, 
with the paper figure panel number, i.e. 2A to 2T, displayed in the figure window top banner. 
One can also run separate sections of the code for each individual figure.

plot_data.m contains the subroutine 'plot_color_fig' which plots the colour data specified in the first argument,
the second argument specifies whether the data contains a column for errors, and the third argument
specifies whether smoothed intensities (default, flag=1), raw intensities (flag=0), or errors (flag=-1) should be plotted.

The RGB colour map in colmap.dat was used for all colour figures.

For all panels, energies are given in meV and l in rlu.
Intensities and Errors are given in arbitrary units for experimental data, absolute units of meV^-1 for calculations. 


Colour map experimental data panels
++++++++++++++++++++++++++++++++++++++++++++++
The first row of the file gives the dimensions of the plot arrays in the form [nRows, nCols].

The intensities are smoothed between neighboring bins by convolving a matrix with relative weights 
A=[0.1  0.2  0.1; 0.2  0.8  0.2; 0.1  0.2  0.1] to have a smoother colour variation between neighboring 
bins in the plotted intensity map. The plotted intensities with experimental data in panels 2A, 2E and 2M are 
produced by applying this smoothing matrix twice whereas those in panels 2I and 2Q are produced 
by applying this smoothing matrix once (the calculation plots are unsmoothed).
Unsmoothed data can be plotted by editing the plot_data.m script as described below.

The code uses an internal function to reshape the data into 2D matrices, 
convert from bin centers to edges, and then use pcolor with appropriate axes, 
colour limits and colour map.

For experimental data files, the format is 
5 columns [l Energy Intensity Error(Intensity) SmoothedIntensity]

For each experimental panel, there are two data files (_1 and _2). For panels E, I, M, Q, these correspond 
to the angle between the c-axis and the incident beam direction
psi = -60 and -25 respectively. 

Panel A comes from a different experiment (reported in Fig 3B of ref [5] R. Coldea et al, Science 327, 177 (2010))  
For this panel, data files _1 and _2 correspond to the angle between the a-axis and the incident beam direction
psi = 155 and 90 respectively. For this panel, the data have been mirrored in l=0 for display purposes
and the intensity scale used is different.

In each case, the plots from these two files are superposed to generate the final image.

The files for these are:
panel A: 0T_1.dat, 0T_2.dat
panel E: 0T5_1.dat, 0T5_2.dat
panel I: 1T0_1.dat, 1T0_2.dat
panel M: 1T5_1.dat, 1T5_2.dat
panel Q: 2T5_1.dat, 2T5_2.dat 


Colour map calculation panels
++++++++++++++++++++++++++++++++++++++++++++++
The first row of the file gives the wavevector bins for which data is provided in the file in the form start:spacing:end.
The second row gives the energy bins for which the code will calculate intensities in the form start:spacing:end. 
The third row of the file gives the width used for convolution. This is the FWHM/(sqrt(log(2))*2).

The calculation colour map files, the format is 
3 columns [l Energy Intensity]

The energies given are the exact calculated energy levels, and the intensities are convolved with Gaussians before plotting.
This is done using the plot_color_fig subroutine, which also uses an internal function to 
convert from bin centers to edges, and then use pcolor with appropriate axes, 
colour limits and colour map.

The files are:

panel B: 0T_ED.dat
panel C: 0T_twosoliton.dat
panel D: 0T_analytic.dat
panel F: 0T5_ED.dat
panel G: 0T5_twosoliton.dat
panel H: 0T5_analytic.dat
panel J: 1T0_ED.dat
panel K: 1T0_twosoliton.dat
panel L: 1T0_analytic.dat
panel N: 1T5_ED.dat
panel O: 1T5_twosoliton.dat
panel P: 1T5_analytic.dat
panel R: 2T5_ED.dat
panel S: 2T5_twosoliton.dat
panel T: 2T5_analytic.dat

Hatched patches for localized analytic calculation
++++++++++++++++++++++++++++++++++++++++++++++++++++
The .xy files give the boundaries of the hatched patches indicating the continua in the right-most column of Fig. 2.

For these, the format is
7 columns [l E1min E1max E2min E2max E3min E3max]
where E1, E2 and E3 denote the top, bottom and middle continua respectively, and where the min and max are the lower
and upper boundaries respectively of each continuum at that l value.

plot_data.m contains the plot_patches subroutine which plots these boundaries and adds hatching in between them.

The files are:
panel D: 0T_analytic.xy
panel H: 0T5_analytic.xy
panel L: 1T0_analytic.xy
panel P: 1T5_analytic.xy
panel T: 2T5_analytic.xy





