Of all the commuter rail lines being considered for Houston, the one to Fort Bend is the most studied. It was one of two lines proposed as part of a 1993 plan that failed to get federal funding. It was the only commuter rail line included in the 2003 METRO referendum. And, in 2003-2004, it was the subject of 9-month study (1.7mb PDF) sponsored by HGAC.

The 2004 study concluded that a 15-mile commuter rail line along US90A from Fannin South to Rosenberg was feasible, but that it would need its own tracks alongside the existing freight rail tracks. It estimated 12,100 daily boardings for a capital cost of $380 million. (These are 2004 costs, and there has been a lot of construction inflation since.)

This year’s HGAC regionwide study looked at the same corridor as part of its “Principal Corridor” system, but, because the goal was to bring all lines into one central hub, modified it to run to the Intermodal Center instead of Fannin South. The study also extended all of its proposed lines to the edge of the HGAC 8-county region, which put the end of the line in Kendleton, TX (pop. 537) instead of Rosenberg (pop. 24,043). These changes resulted in a much longer 50-mile line which would also be considerably more expensive (over $600 million) due to the additional rural trackage and the cost of upgrading the busy Terminal Subdivision inside the Loop. However, the projected ridership — 7,500 — is much lower than the earlier study. That may be due to some differences in the models used, but it likely has more to do with the lower level of service in the newer study (8 trains a day each direction instead of 22) and with the fact that a lot of the people who live in Fort Bend work in the Medical Center.

In the final stage of the study, the system was modified based on railroad input. Because of the same freight rail congestion that lead the 2004 study to study separate freight rail tracks, the 90A corridor was dropped altogether from the modified “baseline” system, replaced by a new “Popp” corridor. This has the advantage of a good connection to the Medical Center. However, it requires creating an entirely new rail corridor inside 610, following 288. The study found a ridership of 8,406, better than the Principal Corridor plan but still much less than the 2004 study.

At the public meeting on the HGAC plan, Missouri City Mayor Leonard Scarcella reacted angrily, calling the revised plan the “Loop-De-Doop.” It’s not hard to tell where he’s coming from. In avoiding freight rail congestion, this option also avoids most of Fort Bend’s population. Overlay it on population density, and we see it’s a classic case of rail where the people aren’t:

So how did we get here?

The problem is not in the study itself, but in the basic parameters it was based on. The political leadership of HGAC wanted to study only a single mode (diesel locomotive-hauled trains sharing tracks with freight rail) with the lines running into the Intermodal Center and extending out to the edges of the HGAC region. No study was ever commissioned to see if those were the right assumptions.

So let’s question those assumptions. First of all, mode: with congested freight rail tracks and no good connection to Downtown, this might not be a corridor suited to commuter rail. In fact, the 2004 study considered light rail, running alongside the freight rail tracks as in the photo above. Because light rail trains are smaller, they are suited to more frequent operation: every 15 minutes at rush hour instead of every 30. That alone, they found, would increase ridership from 12,100 to 17,200. If those trains were run onto the Main Street line, eliminating a transfer, ridership would increase further to 21,800. Light rail would come at a higher capital cost: $756 million compared to $383 million. But the cost per rider would actually be lower.

Now let’s question another assumption: do we need rail to Rosenberg? Population density drops off dramatically once we get past Sugar Land. The line outside of there cost the same per mile, but it picks up many fewer people. So let’s determine the extent of the line based on where the people are, not on political boundaries:

With the outer end at Sugar Land, we have only a 12.6 mile line. Assuming the same cost per mile, that cuts the capital cost to $376 million, less than commuter rail to Rosenberg. But even if we assume that the people who would have boarded in Richmond and Rosenberg would not ride at all (when in reality many of them would simply drive a bit further and catch the train anyway) we still get a ridership of 13,900, higher than commuter rail to Rosenberg.

The bottom line: based on these studies, light rail to Sugar Land cost less, carries more people, and offers more convenient service that commuter rail to Rosenberg or Kendleton. And, of course, the light rail line could always be expanded outwards in the future.

But the most important point is this: we have a variety of transit modes to chose from — commuter rail, light rail, commuter bus. We can’t assume that one of these will be the right answer everywhere. We have to consider all of them in light of the demographics and existing conditions in each corridor. And, in the end, we’ll likely find different answers in different corridors. HGAC indicated two weeks ago that that was indeed the intent: studies will be done in each corridor, considering multiple modes. We need to make sure that political pressure to move fast, however well-intentioned, does not override that process.

The forums are the right mode for your comments.