I've been casually following the thread and I suppose it's time to jump into the fray. The unstated reason for why TETRA is more beneficial than other technologies is talkpath gain. For one 22 kHz wide channel (a 20 kHz channel for PowerTrunk's "reduced power" TETRA), you get 4 talkpaths. For one repeater station, you have 4 timeslots that can be each used for a separate conversation or data transaction. In DMR ("Mototrbo" and others), for 1 repeater station, you have 2 timeslots. In P25 Phase II, you also have 2 timeslots. So, for deploying a 4 talkpath system, a DMR or P25 Phase II system will be twice as expensive because you need an extra repeater, not to mention those use 12.5 kHz wide channels and half of a channel might be orphaned. This becomes critical on 700 MHz narrowband channels, because they can't effectively be reused in proximity to the other half of the channel.
There is a caveat - P25 Phase II still relies on a dedicated control channel, so one repeater is dedicated to 9.6 kbps signaling and control. A three repeater P25 Phase II system allows for 4 talkpaths, not 6. A three repeater TETRA system allows for 12 talkpaths.
That's most likely the reason the other systems were not considered. TETRA represents the way to maximize traffic through the system with the greatest efficiency and least amount of money dumped into fixed-end infrastructure (no, I don't sell this stuff).
The parallel to TETRA is OpenSky, with one exception - TETRA has a proven history of success where it's been deployed. Some OpenSky fans might say it actually "works."
As for interoperability, no one ever said "interoperability" could or should be a free-for-all. An agency has the prerogative to contain its internal system. On 700/800 MHz, interoperability can be had on a number of frequencies dedicated to the purpose. It doesn't have to live on the agency's internal system, and an internal-only system does not need to be held hostage to the "I word," especially if it's a heavily worked system just based on day-to-day operations.
There is no reason why an SDR subscriber unit can't accommodate multiple waveforms - that includes TETRA, P25 Phase I, P25 Phase II, and analog.
As for a TETRA scanner, it has to be a very niche product nowhere near worth the return on investment in North America. That doesn't mean the very talented people in the Open Source community couldn't take a swing at it.