re: Needs to be a lot colder for CO2 to work
Ah, no.
If it got colder, the CO2 would freeze, as it does at the Martian poles already. What you need for CO2 to flow like water is the present temperature, but more pressure.
You can do this underground. It's very hard (impossible) to do this with any beleivable model of atmosphere post early Noachian bombardment times.
What CO2 offers is a way to enhance the fluidisation of dust by changing phase from solid ice to vapour in a collapse event, as I model for the Malin and Edgett gullies.