REAP

Perform the same calculations as in 3562 and 3563, but for each instance in time for a run.

Peter writes:

The user would select the time of interest and the results would be displayed similar to the general stats. In addition, it would make sense to display the differences in color visually. What I have in mind here is an image of the means for all data sets for the given instant in time. I'd choose the number of rows and columns in this image based on the number of data sets being compared. Specifically, I'd make the image either square if the number of data sets is close to, but a little less than or equal to, a perfect square and rectangular if not. For example, if the number of data sets is 13, I'd make the image 4x4 with the means of the first 4 data sets in the first row, the 2nd four in the 2nd row, the 3rd four in the 3rd row and one element in the 4th row. It the number was 11, I'd make a 4x3 array. I'd either use MxM or MxM-1 elements. It might be good to augment this with one additional element that is an image of where the tiles are in the image of this time. This might be the element in the upper right hand corner of the MxM (or MxM-1) image with the other elements corresponding to the various data sets. I understand that each element in the overall image is a single value (color), but I still think that it would be good to display the data this way - one gets a sense at a glance for how the data compare. (Of course, one could present the same image for the General Stats. The color of each tile for a data set would be based on the mean SST for the SST of that data set at that instant in time; it would be summed over all tiles at that instant in time - the sum over the first dimension of the data matrix for matrix elements a, b and c).