I'd unbolt that track from the roadbed.
What you see is the result of change between 2 fixed points.
I only use clamps where I have to, I much prefer the screw down rail joiners. Sure those tiny screws can be a pain, but using the web of the rail makes a smooth railhead path a lot easier to achieve.
I'm in S AZ, my temps range from -4 to 114 degrees, I use Aristo SS. I see lateral movement all the time, but the track stays flat by it's own weight.
Half my track floats in ballast and the other half sits plain on wooden planks raised a foot or so. I don't bother removing the screws that hold the rails to the ties. I figure the strength of the steel will force the plastic to conform and so far it has.
Due to illness, the plain elevated track was left alone. Many a day was merely pondering, my 2 fav pondering spots have sightlines down my longest tangents, never saw a kink. My feet are harder on track than nature. I don't leave gaps nor do I have any expansion devices.
Unlike real RRs, our track is stronger than ballast and will grow sideway, as a unit, if allowed.
For the track in ballast, half of a return loop, is in a cut and held fairly ridgid, the other side moves laterally on the bridges. Most of my expansion is resolved by the curves that bulge a tad or so it seeems.
I'd venture that your overnight buckling was caused by your roadbed shrinking underneath the track. Is it elevated? Have cold winds? Bobs yer Uncle. Ya we like to work when it's warm......
John