The ROI, except for masked-off data is fitted by a 2-D polynomial of user input order in X and Y. The chosen surface is the least squares solution to the data. Generally the order of the polynomial should be much lower than the number of data points in each direction to avoid that the surface fits noise in the data.
The RMS residual of the fit is output. This figure may be compared to the error estimates of the data to see if the fit is to loose or fits the errors too much. If variance estimates are present, the fit will be weighted, and the RMS residual output will be a reduced chi-squared figure.
The fitted surface is created in the memory, so may be recovered after leaving the FIT sub-menu.