experimental MAF reset switch: which circuit to interrupt? more complications

[workingonit]

After years of having problems with fuel trim readings being higher than normal, which resulted in setting codes for lean conditions on both banks, and eventually setting a CEL on the dash, I have searched, tested, prodded, and done work-arounds for keeping the abnormal readings at bay. Also, I've checked the tailpipe for bad smells (indicating a real problem), haven't had but one misfire in 3+ years (the engine starts and runs great, always), and monitored the O2 sensors as well (AOK). I've used three different MAF sensors (to see if there was any mis-reporting...the OEM unit was reporting lower than the two, cheap, replacements I now use...one at at time, of course), and done several sight inspections, and more smoke testing (and never found anything amiss), I finally gave up looking, and have resorted to just resetting the fuel trim reporting system by unplugging the MAF, temporarily, and have done so for almost three years.

Since I very seldom drive the X5 (2600 miles over 3 years and 3 months), and 99% done in stop and go, low speed, short hops (maybe 10 miles max, 5 miles or less, usually), I just reset the fuel trims whenever the eventually climb near "10", before any codes might possible get set, and forget about actually fixing the phantom, unfound, underlying problem causing the fuel trims to rise. However, on the very few longer, higher-speed trips I've driven, the fuel trims actually go lower, my overall fuel economy rises, and all is well...until the short-trip cycle starts afresh.

So, just yeasterday, going to the Dr. during a rainstorm, my LTFT's were rising again, and I wondered if I could find a way to do the MAF reset, if I nneded to, during a pouring rain, without going underhood. I thought that I could use an on-off switch to interrupt the circuit (for 10 seconds or so, only), after I pulled oot of traffic, of course, to reset the fuel trims, without getting soaked. My M54 has a 3-wire MAF (ground, signal, and positive), so my first thought would be to interrupt the ground or positive feed (any signal wire going to the DME seems to present the chance of shorting out the DME, in my mind). But, maybe I'm looking at it wrong. Any ideas?