Here are the steps: 1) digitize the four corners of the slanted plane; 2) assign x,y,z to each of the corners, reading off the grids which form the axis planes; 3) use regression to arrive at three simple equations which go from the x,y on the slanted plane to the three coordinates (psi, temperature, moisture); 4) cut out the slanted plane with data point and straighten it; 5) digitize the data points; 6) apply the equations to arrive at the final result: three dimensions for each point. You can use my free digitizing and curve-fitting tools. While you could deskew the plane using Paint Shop, it might be easier to just fix it on a pixel-by-pixel basis.
This is what the data plane looks like deskewed. It looks like the points at the top are slightly above the plane and the ones at the bottom are slightly below so you would have to fix that manually. The rest of them (above or below) is just a guess.
This spreadsheet contains the digitized points in the deskewed plane. The next step is 2D->3D based on the slant of the plane. The last step is tweaking the pairs of points slightly above and below the plane. I would then recreate the plot in 3D using Tecplot as a final check. I may work on it some more tomorrow.
I haven't yet tweaked the points for some to be slightly above and some slightly below the slant plane but here is the first draft: data, Tecplot layout, spreadsheet, and what it looks like in 3D
This is a rotated (i.e., moving, animated) 3D view. RG no longer animates GIFs but your web browser should be able to display it properly, showing the rotation.