Non-uniform grid interpolation

[wtbarnes]

For the time-distance plots, we've interpolated onto a uniform grid thus far. @jwreep has suggested that we instead interpolate onto a grid composed of all of the possible unique positions of the grid cells so as to preserve all of information from the original grid resolution.