Overlaying Tables

To Overlay Tables is to choose a column from each table and then perform a merge operation where a row for the overlaying table is merged into a row of the target table if the value for the chosen columns from the row in the overlaying table and the row in the target table are identical. If a value for a cell of the overlaying table is blank, then it is not merged into the target table. If the overlaying table has new columns, then those columns are automatically added as part of the merge. If multiple rows from the overlaying table match the same row in the target table, each match will create a new row in the resulting merged table. 

The interesting part of this process is determining the propagating effect of a change in a row of a table. So if 10 tables are being overlaid on top of each other and a row is modified in one of the middle tables, is it possible to determine how the resulting merged table is modified without having to perform the merge process again? This problem is at the heart of differencing calculations in the RDD project.