This is the problem with the transportation infrastructure here in the US. In Japan those trains could easily make 100-200 round trips a day. And there would be a bajillion other train routes to other cities as well.
Seriously, trains? Between two neighboring major cities? Three times a day? That's laughable at best in most parts of the world.
If the demand isn't there, why spend untold billions filling it?
Tokyo's station serves 3000 trains daily and some 415K passengers per day as of 2013.
The Chinese high-speed rail system also deals with similar levels of passenger traffic. Shanghai's Hongqiao station for instance has no less than 30 tracks, can hold 10K passengers at the same time, and comfortably serves an average of 210K passengers per day (much more on holidays), and it is only one of a few railway stations serving the area.
Trains scale well if you do them right. Technology has come a long way.
But let's take your numbers: 415K passengers per day. Let's keep the same three day evacuation period. That's ~1.2 million people. Leaving 5 million in Houston.
Here's the point: Evacuating millions of people over a weekend is a gargantuan task. Magic words like "train", "plane", or "automobile" don't instantly solve the problem.
That's what life would look like if people in the US actually used trains on a regular basis.
In any case, all I'm saying is that if a massive number of people need to be moved, trains can scale well, additional reserve capacity can be deployed during evacuations, and it's also a lot safer than having hundreds of thousands of people taking to the roads in bad weather.
Tokyo is crammed and that's why trains work great over there. They wouldn't work in US city with suburban sprawl.
[1]http://www.npr.org/sections/thetwo-way/2017/08/28/546721363/...
The headline has a question mark at the end; that's the extent of the "questioning". The rest of it lists every point made to support the decision to stay. It even notes at the end that both FEMA[1] and the state government not only refused to question the mayor's decision but backed it.
It's not a critical story, and it's disingenuous to suggest that simply because "even NPR" ran an explainer story, that there's some widespread official or credible base of observers "questioning the present stance".
[1] Literally:
> The head of FEMA, Brock Long, "did not question the decision," Reuters reports.
>But at some point, of course, those floodwaters do become deadly. If lower floors flood completely, Houston's Office of Emergency Management cautioned, residents should never take refuge in the attic unless they have a way out. They could risk being fatally trapped.
> Attempting to drive through flooded streets is a common cause of death during heavy rainstorms.
Basically, you are damned if you stay and damned if you drive. Maybe building a float from furniture would work, but I think the Office of Emergency Management knows better than random internet commenters how to handle a flood.
[1]http://www.npr.org/2017/08/29/546978194/with-much-of-houston...
That train could evacuate 15K people a day.
Imagine if each state maintained one of those trains. In a crisis, Texas could borrow trains from NM, OK, KS, LA, AR, MS, and AL. They could move a combined ~120K people per day (no pets).
Planning ahead a week and beginning the evacuation 3 days before the storm, you could move 360K people. That would leave only 4.6 million in Houston.