Isn't this what Jesus did?
Canonical_Warlock
You still can. My dad bought a 2010 F150 for $400 a couple years ago. Sure it used to be a salt truck and therefore had more rust than metal left on its body. Sure it had 4 bald flat tires on it. Sure you have to disconnect the battery every time you park it or it dies. Sure the CD player ocasionally makes grinding noises and starts smelling like smoke every once in a while until you whack the dash hard enough to make it stop. Sure it has no shocks left whatsoever and it feels like you're driving a trampoline. But who cares about minor things like all that?
Mine is a ball python. They seem to prefer basking under the heat lamp after eating. Also they haven't ever pooped in their water dish that I can remember, but I have had them horf up a partially digested rat in there before. That incident was made even more pleasant by the fact that I had their heat lamp positioned over the water dish at the time (to try and keep humidity up) and it happened while I was sleeping so I didn't catch it for several hours. So I awoke to my entire house being filled with the miasma of a partially digested rat which had been stewing under a heat lamp for several hours. The smell was indescribable and beyond the imagination of any sane individual.
Oh yeah, basically the easiest way to do it would to pump water out of each basin, through a heat exchanger, and then back into the basin. That way you could have your whole temp control aparatus located outside of the terrarium. Plus that would also easily enable automatic water level management to keep the water levels identical. For your heat exchanger you could just use a CPU water block and a peltier device. You regulate the power going to the peltier device by monitoring a temp sensor in your water return pipe and just pulsing the peltier device on and off at different rates to control the heating or cooling rate. Plus with the peltier device you can just reverse the polarity to switch from heating to cooling to enable the shuffling of the basins. All of this would be controlled and charted in a csv file by a raspberry pi. Additionally you could connect a simple motion sensor so the pi could flag the times the snake was using one of the basins to make it easier to read the data.
Rather than monitoring ambient temp or humidity you would probably be better off just keeping them tightly controlled and constant via other systems. That would further reduce variables for the initial test.
If that's your fetish, I guess.