true, although both have grown. this also means i am not going to copy it to the spare 2TB drive i do have (i have only one spare USB3 port and this would take all day on USB2). i will limit any solutions being considered to those that either don't change what's there (in existing trees) or add new trees that i can "rm -r" later.

my latest idea:

a python script will keep a dictionary indexed by each file's i-node and store a set of the top-level directory name(s) the file is found in. another dictionary indexed by i-node will store the file's sub-path under the top-level tree. after a scan finishes building these two dictionaries a loop iterating over all the i-nodes will handle each that has a size of its set of top-level names different than the number of trees by reporting which trees it is in and which trees it is not in, along with its list of sub-paths. then this script can figure out if any tree has every file and list them as critical trees to keep. it can also rerun the scan over all trees looking for the problem sub-paths and report all the relations (which sub-path in which top-level).