Driver Blocks Trains On Elevated Tracks at Mount Baker Station

96 Replies to “Driver Blocks Trains On Elevated Tracks at Mount Baker Station”

1. Does anyone know why the driver drove up on the tracks?

   1. There’s zero information about “why”. To be honest, based on the Utah plates, she was from out of town, unfamiliar with the area and out of sheer stupidity ended up on the tracks when making a turn. And then just kept going. In one IG video, a bystander who was on the platform as it occurred, said “maybe she has dementia”. Doubt it. She was just dumb AF

      https://www.instagram.com/reel/DZHXg-VoLHH/?igsh=MXh6Mzh6bHlrYjVkZw==

   2. I think I saw a mention about people following GPS directions and ending up on the tracks when they don’t quite understand a complicated intersection situation. That’s not positive confirmation that that happened in this case, but it may be a possibility.

      1. Can’t forget Michael Scott from the office driving his car into the lake over his GPS telling him turn right vs making a bear right
         https://youtu.be/DOW_kPzY_JY

      2. Makes sense. It’s no different from looking at routing on GPS going through Boylston Ave but the map looks like you’ll be on I-5. Or GPS tracking you along the express lanes but you end up just taking the standard throughway because the alignment is nearly identical. I myself, and I know very well every other driver, has once or twice take a turn (usually left turns) into the wrong lane because the intersection is so massive or the markings aren’t apparent. But there’s quick realization of the mistake and a correction in the moment. This driver, barring dementia or a medical episode, lacked any common sense.

         Any one here 70 years old and have mental lapses when doing a simple task (serious question)?

         1. haha I’m almost 65 and have common sense to know I am not driving on a road. I also come from an era of not needing GPS to direct me to someplace or at the very least for me to see something isnt right. This is just a stupid person not paying attention to the area around them and relying wayy too much on a GPS. I saw her wheels as she came into the platform and they were turned into the rails probably ruining her transmission.

2. So – we’re now going to build a new downtown tunnel to make the line to Tacoma more reliable (and duplicate monorail service to Seattle Center), but leave the at-grade section as is, while making no reliability improvements to the original tunnel carrying the east and north lines. Bravo – a political solution with zero technical merit!

   1. There’s a parallel process to increase safety on MLK, and leanings toward maybe grade-separating it long-term. But that’s not in ST3’s project scope so ST3 money can’t be used for it, so it’s no relevant to this decision, except in the sense that Seattle’s promised cooperation on Graham could include some street safety improvements right in the vicinity, compatible with the strategies ST is favoring in the general MLK safety task force.

      1. Graham street is an enormous waste of money already. So you’re saying the $250 million towards a redundant station on an already slow segment is “leanings toward maybe” having the farce rebuilt in the sky later?

         Well at least the rest of it will be faster….

   2. Of course, what else are you gonna froth about if they didn’t?

3. They reopened the tracks quickly. That’s not necessarily a compliment. Did rail specialists or engineers have time to thoroughly examine the portion of the tracks that were driven on before reopening it?

   Also, check out a Street View of BelRed Station at 132nd and Spring Blvd. 100% someone, someday, will mistakenly turn onto, and drive down the tracks.

   1. In 2022, someone drove onto the tracks on 136th St in Bel-Red. This was before the starter line opened so there were no operational impacts. From the police report:

      “V1 was making a right turn out of a parking lot onto south bound 136th PL NE. This area of the roadway is divided by two ways of light rail. there is a full height curb, then a dip to the rail tracks. V1, when making a right turn, drove over the curb and over the first rail line, getting stuck halfway onto the south bound track. […] D1 was extremely nervous and was not completely sure as to how she got the car onto the tracks.
      Superior tow removed the car from the tracks. Sound light rail was notified of the incident and came to the scene to inspect for damage.”

   2. I would respectfully disagree the tracks were not opened quicky. It took 2-3 hours for service to resume. If the driver was able to successfully drive a regular SUV with no visible damage all the way up to the platform, crews could’ve easily reversed the car back down the tracks.

      And yes, I strongly believe someone will drive onto the tracks at Bel Red area for sure.

   3. Sam, I doubt that an SUV could damage the trackway in any way. Steel rails would make short work of anything on the undercarriage of that SUV-let that were rubbing them.

   4. My concern is for the portion of the tracks that were driven on that are not embedded into or flush with the concrete. The raised rail portion, as can be seen in the top post photo, at and near the station. Not sure how far it extends beyond the station. Anyway, my concern is for where the car possibly scraped its undercarriage along the tracks, possibly damaging rail clips and other rail parts. But, if specialists came out and thoroughly examined and tested the rails before resuming service, that’s good enough for me.

      1. It’s understandable to be concerned about the fasteners (the “rail clips”), but the truth is that they are 1/2″ annealed steel rods and (mostly) enclosed within their anchors, either to the rail or the plinth. They would be among the gleeful wrecking crew disassembling anything that reached down that far from the undercarriage of the SUV-let.

         1. Tom, and that’s the attitude I’m afraid of. A sort of arrogant confidence that the rails are indestructible, and therefore it’s pointless to carry out a thorough inspection. An arrogance so great that maybe an ST employee just stood on the platform, looked down at the rails, and said “Looks ok to me. Resume service.”

         2. They’re not “indestructible”. But they can’t be put out of gauge by a passenger car bouncing along them.

            For sure, the car might have dropped something on the rail that would be a derailment hazplinth. I bet that the inspectors were concerned about that.

            But a car just can’t move a modern trackway by driving on the concrete ties, even if the tires hit some of the plinths. It doesn’t produce enough force.

            A front loader with the bucket in scraping position hitting it directly from the side? Thst’s a problem. A car driving on the ties? Nope.

         3. Tom, I hope there aren’t people in ST/KMC rail that have that attitude. It’s how people arrive at the conclusion no thorough inspection is needed, because it’s impossible for the car to have damaged any part of the track. That attitude is arrogant and ignorant.

         4. Question: “Can a car damage a rail fastener?”

            “Yes, a moving car can absolutely damage a rail fastener. This typically happens at level crossings when a heavy or low-clearance vehicle bottoms out, strikes the rail directly with its undercarriage, or collides with the track infrastructure.

            If a car collides with the track (e.g., losing control at a crossing or high-speed impact), the massive forces can easily shear, bend, or dislodge the bolts and clips that secure the rails to the underlying ties or concrete slab.

            Because damaged or missing fasteners can cause gauge widening (the tracks spreading apart) and subsequent derailments, railroads require thorough track inspections after any vehicle collision at a grade crossing.”

   5. “It took 2-3 hours for service to resume”

      It’s a bit surprising that after the car was cleared, it took an hour for trains to resume. You’d think it might take ten minutes. But we aren’t transit-accident logistics experts. Three hours is better than six or nine hours. At least ST managed to keep it down to three hours.

      1. Track Inspection

         Given the distance she covered, the track inspector needs to be absolutely sure the tracks are not out of gauge for her whole travel segment.

         1. No way, Jim. An LR train exerts a hundred times as much rail spreading force as does anything on that tinny SUV-let. [ed note: Yes, “tinny” as in made with thin, cheap metal, not “tiny”, though it is that, too]

         2. @Tom Terrific

            Yes, the physics as you describe is true. However, without an inspection, whoever authorizes a train to proceed over said un-inspected tracks is in Big Doo-Doo if the train derails.

         3. Jim, that’s certainly, so of course they inspected for kinks. But they would have been VERY surprised to find one.

            Tracks are HEAVY……

         4. And trains are loud,
            That’s why people feel it’s safe walk in the middle of the ROW, even on systems at-grade with their own exclusive ROW. (BNSF Mainline, for instance)

            The point, within this discussion, is never take anything for granted.

            Doing a track inspection is a minor inconvenience in a relatively rare situation such as this.

      2. @ Mike…I can see your viewpoint that indeed we aren’t train-accident experts. And an inspection probably took place. But for context (I’m an aviation nerd), when planes have “hard landings” or engine fires on the runway, air traffic control closes the runway for inspection of debris. It usually takes less than 30 minutes and they have to cover a much longer distance.

         1. How many people are involved in that inspection?

         2. As many as the industry-standard policy says it requires, using the tools and methods that the industry-standard policy says it does.

            Look, I work in transportation, I get that people are nervous about something that’s out of their control and, if not done right, could kill them. There’s entire volumes of documentation for how these things work, though, developed over decades of practice and implemented by people who actually care about their jobs.

            There’s no magic involved, it’s just that nobody wants to be the weak link whose crappy work kills a dozen people. Processes like this do fail sometimes, but there’s no indication that anything has failed or is likely to fail here.

4. I’m surprised this hasn’t happened before. My guess is there are people that drive on other parts of the line bus get out before the train comes. There are plenty of places around the United States where you can drive on the tracks (including right in Seattle) so that part of it probably doesn’t seem crazy (at least initially). She was probably unfamiliar with the area and when she started going up she probably thought it was an overpass. Then she freaked out (rather than backing up).

   1. Most drivers have the common sense to know that they should not be driving on a railroad track bed. But they should paint the pavement like they do a bus lane with do not enter light rail vehicles only. Or put some of those one way spikes down so that if anyone try’s to enter they have their tires flattened. Because sadly people ignore road signs all too often.

5. I wonder how much some good reflective road paint might have made it more obvious that cars should not enter.

   Though after 20 years of operation, and this being the first occurrence, perhaps not even that much money should be spent to avoid this problem in the future.

   1. I’m with you. This looks like a one-time freak accident. If it happens again this decade, then we can reconsider.

   2. The paint and prevention signage are minimal at many places on MLK. Certainly this incident doesn’t hold enough significance to warrant spending several billion to remove all the at-grade segments, but better delineation doesn’t have to be expensive. It can be done quickly at minimal cost.

      Right now there is only a lone sign in the median that is low to the ground (not to the right of the vehicle like most signs). The sign just this 🚫 with a car behind it and two lines underneath. Because the car apparently went up the southbound tracks, the driver did seem to try to stay to the left of the sign.

      SDOT seems to have bought thousands of little yellow poles to delineate traffic lanes already. Installing them at problematic locations next to the tracks looks like something that can be done cheaply and quickly. Add to that an added caution sign where the MLK median pavement ends or red paint across the median pavement around the tracks would make things much more clear.

      1. The red paint on the pavement for ten yards on either side of a cross-street in the embedded rails section is a good idea. It’s cheap and most drivers in Seattle now recognize it as “Transit Only!”

         1. Where SF Muni has paved concrete medians in the middle of Third Street, they installed both red-dyed concrete and added disruptive bumps to jolt mistaken drivers.

            https://maps.app.goo.gl/JD1KUEr89xmvDN4w7?g_st=ic

            In other cities, they usually don’t even put concrete pavement next to the tracks and have rock gravel as the top layer instead — even in more urban places like Brookline or central San Jose or Gresham.

         2. Yeah, those little bumps are better.

6. The most obvious answer is to fully grade separate the tracks in the Rainier Valley.

   ROTFLMAO with an especially big “Har-de-har-har!” at the end.

   What planet are you from, Michael? ST just agreed to spend $200 million on an at-grade station at Graham Street as a sop to the angry buzz of Seattle local officials stunned (and stung) by the truncation of the Beloved Link Extension at Seattle Center station. [i.e. without a good transfer site for 15th West/Elliott buses].

   There will be no “grade separation” of the tracks through the Rainier Valley. Especially not one undertaken to prevent a recurrence of something that happened once in twenty years. There’s no way to do it without closing either Link itself or one side of Martin Luther King Jr Boulevard for years except a bored tunnel. And ST is not going to spend seven or eight billion dollars boring a tunnel with four stations there.

   As much as people diss it, a slightly elevated bypass along Airport Way is much more likely than rebuilding the RV line. Caveat: such a line only makes sense if South King County west of I-5 gets much more dense than it is today.

   1. To be clear, there is NO way to do it [including “dig a trench”] without closing either Link itself or one side of Martin Luther King Jr Boulevard for years”. Period. End of story.

      This “billion dollar project” is vapor-ware.

      1. This “billion dollar project” is vapor-ware.

         ROTFLMAO with an especially big “Har-de-har-har!” at the end.

         What planet are you from, Tom?

         You think an engineering study from three years ago is garbage but you think a mythical line that has no study to back it up is realistic. Miles of new track and how many stations? One? Fifteen? Not to mention interlining on both ends lest we force people to make yet another transfer.

         A buried line in Rainier Valley would improve safety, reliability, headways and speed along the corridor. Is it likely? Of course not. Unless something changes, Sound Transit will continue to make stupid changes. But that wasn’t the point. If I write that “the 75 should go to Bitter Lake” or “we should get rid of the South Lake Union Streetcar” it doesn’t mean it will happen any time soon. But it could happen in the future and more importantly it should happen. Sound Transit needs to rethink all of the ST3 projects. In doing so, it could easily find that burying the trains under Rainier Valley is the most important addition for the south end. Of course it is disruptive. But guess what? So is the new downtown tunnel. That isn’t going to stop them from building it.

         1. A bypass would presumably be more expensive because it’s new build, but I think Tom’s primary point is correct – elevating or burying RV Link will be massively impactful to existing service for years.

      2. I’ll drive up the tracks too if it makes Sound Transit take another look at grade separating them!

      3. I’ve told you time after time that there iris room between Airport Way and the BNSF tracks adjacent to it to take one lane from Airport Way (leaving itvthree) for an at-grade but separated very cheap line all thecway to the aircterminal. The northbound diversion would burrow under the existing tracks and BAR right next to the BAR to Airport Way underpass. The southbound merge would be elevated.

         North of the Air Trrminal the trackway would have to be elevated to just south of Albro Place, where it wouldcdescend to cleatcthe bridge between Airport Way and BNSF; thrre’s plenty of room. It would rise a bit to a semi-elevated station just north of the freeway on-ramp bridge then continue to rise enough to clear BNSF and pass over to the east side of the rail ROW squeeze through under the freeway off ramp –there em(barely( enough room — and then continhmue on north on structure above the old spur through the Water Department to Industrial Way, cross Airport Way and follow the median of Ibdustrial Way to the old UP tracks west of Sixth South which lead to the busway.

         So a Bypass NOT a fantasy, Ross, nor does it need to be a ten billion dollar tunnel through miles of industrial wasteland like Seattle Subway’s “project” you’re thinking of.

         While certainltmy not free, this option provides a Georgetown Station right in the business district instead of the middle of run-down SFH’s and is all elevated, mostly only enough to keep pedestrians off it.

         1. If the line doesn’t have at least 2-3 stations per mile, it’s not good enough.

         2. It depends on what’s there. If it’s an industrial area with few walk-up destinations, there’s little point. SODO station is based on people walking to e.g., Starbucks, the tech museum, the surplus store, the restaurant food store, or jobs in any of the buildings since it has reasonably walkable street grid for an industrial area, and bus transfers.

            In contrast, BAR station has hardly any of that. Other theoretical stations on a Georgetown segment don’t have much either. And while Georgetown is a small retail center, it’s very small and the area has little housing. So it’s not a priority place for a subway line, or for stations every third mile.

            Sound Transit had a Georgetown bypass segment in its long-range plan. It deleted it in 2014 saying it wouldn’t be needed. The primary beneficiaries of it, South King and Pierce, did not say one word to defend it. So you’re asking for something the subareas don’t want and don’t want to pay for. North King has many higher priorities than Link to Georgetown.

         3. I don’t really see the value in a bypass line here. It will never save enough time for Link to be competitive with express buses, and with a freeway alignment, it wouldn’t be that difficult or expensive to build an express bus system with reasonable transfers to Link.

            For a cost on the order of a couple hundred million, ST could build 1-2 freeway flyer stops plus a center exit connecting to SODO and Tacoma Dome.

            Honestly for not very much money ST could dramatically reduce crashes and improve the speed of Link trains on MLK. Even without a major street redesign crossing gates and full priority seems like an obvious major improvement? I mean it’s already been done on BelRed.

         4. The only way I see a bypass line is if the MLK segment gets overcrowded or if ST tries to add a second line. Even then, there would likely need to be a study to determine if it’s better to build a parallel track segment or just redesign MLK with full grade separation — or to add a new line using a rail technology that can go much faster.

            The long time period of ST3 (and expansion still underfunded for several decades even with it) really postpones planning expansions of the system in meaningful ways for at least a decade if not two.

   2. It’s not just for this one situation, it’s for all the strayings, collisions, and near misses in the MLK segment that occur every month. When the alternatives (surface, elevated, underground) were being compared, the cost in human life and passengers’ time and ST’s financial liabilities of a surface alignment should have been included in the cost comparisons, but they weren’t; they were ignored and valued at zero. This artificially lowballed a surface alignment’s costs compared to grade-separated. If MLK had been grade-separated from the beginning, all those dozens of collisions and service interruptions over the years could have been avoided.

      Recently there has been some interest in a retained-cut (open trench) alternative like on 112th in Bellevue. That wasn’t considered during the initial segment, but it might be a lower-cost solution now compared to elevated or tunnel, and might be more pedestrian-friendly than overpasses for the cross streets, where pedestrians would have to go up and down to reach Link or cross the street.

      1. The problem isn’t that they didn’t take into account surface collisions.

         The problem is they thought the collision rate would be similar to E Burnside MAX in Portland. The street design of ML King lends itself to a lot more accidents.

   3. “To be clear, there is NO way to do it [including “dig a trench”] without closing either Link itself or one side of Martin Luther King Jr Boulevard for years”

      That doesn’t mean it can’t happen, or that it might not be better than having these ongoing accidents. The ultimate problem is that it wasn’t grade-separated in the first place. When you have a flaw like that, you have to retrofit it, even if it’s costly in terms of money or disruption. Again, that’s why you need a transit agency that puts transit best practices first and doesn’t built substandard alignments.

      “This “billion dollar project” is vapor-ware.”

      There is no “the project” to grade-separate MLK. Just preliminary concepts and some interest in addressing the issue. There’s no concrete project proposal yet, much less a price tag, or politicians committing to it, or a funding source. In the interim ST is looking at some incremental approaches, like more pedestrian gates or better signage or warning signals or such.

      1. That’s a lot of ST-pleading, there, Mike. And, to be very clear, a retained cut/trench would NOT “improve the pedestrian environment” in any way at all. Riders would still have to cross one-half of Martin Luther King Jr Blvd to reach the stations, unless the trench were put under one-half of the existing street and the existing railbed given to cars. Then roughly half the riders would have to cross the entire street.

         The most useful thing they could do for pedestrian safety is to lower the speed limit on King Boulevard and enforce it with automatic radar and cameras. Then, as a sop to the whiners, only put half as many points on the car owner’s license. If you want to avoid getting points from your feckless offsping, don’t lend them the car. They can take the train; it’s right there.

         1. “frequent automatic radar and cameras”. Like every other block.

         2. And, to be very clear, a retained cut/trench would NOT “improve the pedestrian environment” in any way at all.

            Nonsense. The train tracks take up a lot of space. People get hit by them. If the train is buried it gives the city space to wide the sidewalks, add more trees, move the car lanes closer together so it doesn’t take so long to cross the street. It would definitely improve the pedestrian environment.

         3. Well yes, if the train is actually “buried”, but not if it’s running in an open trench. There would have to be at least as much space between the driving lanes and the trench as that required between the driving lanes and the trackway and probably a bit more to avoid cars falling into the trench.

            Indeed, it would require Jersey Barriers along the entire trenched portion. Bollards and chains mark the separation now.

         4. “Bollards and chains mark the separation now.”

            Actually this is not true for most (say over 80%) of the MLK section with Link tracks. There is almost no delineation north of Cloverdale. Even the trackway curb is slight enough to be easily mountable by a vehicle. There is no red concrete or paint. There is minimal signage. There are very few if any bollards.

            This is the situation which contributes to driver confusion, like how a driver can steer a car can end up on the Mt Baker platforms.

            Go walk or ride or drive the corridor and you will see how little separation there is along much of the corridor.

            There is a simple chain linking short metal poles in the median near Oregon St. That is also an MLK major intersection that’s shown to be one of the safest per accident history.

         5. ” trackway curb is slight enough to be easily mountable by a vehicle.” It’s that intentional for emergency vehicles to be able to cross over the tracks?

         6. *Isn’t

   4. Hence the “long term” phrasing in the article and discussion of improved at grade crossings in the short term. I don’t think Michael implied or intended to suggest that we should break ground tomorrow on such a project. I’m not particularly happy about the dismissive and pessimistic attitude that you are putting forth when interacting with articles, to be honest.

      In any event, in the interest of productive discourse, I shall continue.

      Grade separation is much more likely when you consider the reliability and accident rate across the entire segment. Many people have been killed along this segment because they made illegal U turns or didn’t look when crossing. While ST is not at fault for these incidents, it makes sense to grade separate the segment because:
      a) a bypass doesn’t solve the problem
      b) you could automate the 1 line if you grade separated it (including rebuilding Stadium) because it is separate from the rest of the system after DSTT2.
      – no you can’t just automate half of the system and have drivers on the other half, that makes no sense and would be operationally a nightmare.

      c) it’s the equitable thing to do. Bypassing the RV without fixing the safety issue would not only be bad for transit, it would neglect to invest in the RV to solve a real problem. The Duwamish bypass would be nice for people in South King and Tacoma, but it solves a different problem than safety in the RV. By prioritizing grade separation in the valley, you can improve travel times in the south in addition to making the RV safer.
      d) attitudes to changing road infrastructure have changed since you likely last were on the ground here asking people about them. IIRC you live in Texas and have done so for a long time. Things are vastly different here in Seattle. There is a sizeable proportion of people who do have an appetite to make legitimately transformative changes to the street infrastructure to prioritize transit or active/non motorized transportation. I have no doubts people would be frustrated with MLK construction, but the city could even build it back at 4 lanes if they took the side and still have a bike/walking path along the businesses on the east or west side. Saying it’s never going to happen is both unimaginitive and not necessarily recognizing the reality of the situation here.

      1. All good points, and all completely ignoring the financial implausibility or disruption of doing any of them. Especially the disruption of functionally closing South Link for at least five years. VERY few people are going to ride to Rainier Beach, transfer to a bus to Mt. Baker — a bus stuck in the congestion caused by the ongoing construction in the middle of the road which is certainly going to take a lane out of each side for a construction buffer at least in the active segment — and then change back to the train. They might do it for a couple of weeks, but not for a five year period (if it can be done it in five years).

         So that effectively pushes the “inexpensive” trenched alternative envisioned in the article off the table. I will admit that if the trench were in dug in one-half of King Boulevard with the train still occupying the middle it would be possible. But that requires believing that for five years drivers would accept King Boulevard being a two-lane road with some three lane sections.

         If you want to avoid taking lives unnecessarily by a tram line down the middle of a major arterial — something that you can find all over the world; Seattle is not a special case — lower speed limits, limit left turns across the trackway and completely ban U-turns. Then enforce it all with frequent cameras and points on peoples’ licenses. Don’t want your kid to point your license? Don’t lend him the car; he can take the train. It’s right there.

         And, yes, you can automate just half the system. It obviously doesn’t save as much money as automating the entire system, but eliminating the cost of drivers between Lynnwood and Symphony for Line 1 and 2 (they have to board early in order to be ready to take control at the end of automation and alight after closing up when entering it) is worth millions of dollars per year in labor costs.

         In fact, Line 2 could probably board and alight even farther along at Judkins Park, increasing the savings on it. MI is probably too close to SoBel to make the transition. And Line 1 could stay automated to Mt. Baker if Lower Royal Brougham were closed from just east of the tracks to Fourth South and made a part of the busway west of Link. Operators could board and alight at SoDo.

         1. Especially the disruption of functionally closing South Link for at least five years. VERY few people are going to ride to Rainier Beach, transfer to a bus to Mt. Baker — a bus stuck in the congestion caused by the ongoing construction in the middle of the road which is certainly going to take a lane out of each side for a construction buffer at least in the active segment — and then change back to the train.

            You are confused. The five year estimate was for the elevated option. There is no timeline for the trench option. You are also confused about the bus bridge. It would not be on the entire line. It would be between stations (basically segment by segment). They also mention single tracking as a possibility. There are several options and it doesn’t go into detail on each one because we aren’t there yet. This is an early estimate but the study is more advanced than a lot of things we just committed to building.

         2. Ross, is someone actually proposing to dig a twenty foot deep trench next to an active LR track? That’s what would be required for a “single-tracking” method.

            And to do that repetitively over a long enough period to cover four miles? Wow that’s bold. Remember, you can’t use the segments in the trench between the starting point and the next cross-over “beyond” the trench. There’s no way to cross-over between adjacent tracks separated by twenty vertical feet.

            So the single-track portion would keep getting longer and longer.

            And then — and this is the really hard part — the other half of the twenty-foot deep trench would have to be dug while trains run through the completed first half right next to the excavation!!!

            Oopsie if the front-loader spills some dirt on the active track below or a passing train or even worse hits the new OCS a few feet away.

            I guess the good news would be that once the first direction trench was completed the segment-by-segment process on the other, trickier side would shorten the single-tracked operation a segment at a time since once a segment had both tracks in the trench ordinary cross-overs would work.

            From a system operations perspective, doing both directions of the trench a segment at a time (a MUCH easier and less catastrophe-fraught method) and bus bridging the segment under construction would make Federal Way Link into the “South Side Link Extension” and orphan all the rolling stock in the South Maintenance Facility. That’s probably not a fatal flaw, but something to plan around.

            And, finally, how do you build the platforms for trench stations in the single-tracking plan? They’d have to be side platforms, because you can’t put half of a center platform underneath the remaining active surface track of the segment on which you’re working. So that means that the trench would have to widen for the length of the station.

            If you put the new platforms directly beneath an existing platform, that would work, but the excavation would be quite tricky. There would not be any buffer between the outside edge of the trench and the roadway — the existing stations come quite close to the driving lanes — so it’s likely that the adjacent traffic lane would have to be closed, if only to park the truck receiving the spoils.

            It seems to me that the idea of single-tracking is simply not plausible. Maybe the side-by-side segment-at-a-time method would work, but the resulting bus bridge would still be a ridership-killer. And remember, that bus bridge would require that through passengers descend into or climb out of a trench station at one end of the bus bridge after the first section was completed.

            Another wrinkle is that building the station that replaces the station at the end of a segment would require closing the “next” surface station for the entire time of a segments excavation, because stationscare larger than the trench and take a long time to build, even spartan ones.

            So the bus bridge would always be for two segments, not just the one whose trackway was being replaced, except the final segment if it is built south to north.

            The bit between Rainier Beach and the end of the elevated north of BAR will require a bus bridge all the way to TIBS unless BAR has been opened.

            And yes, it would almost certainly take five years to build this thing doing both tracks per segment, essentially one year per segment. Any shorter estimate would be flim-flam or require a full bus bridge between Mt Baker and BAR/TIBS with simultaneous construction over the entire length of the project.. There’s a lot of dirt to remove and haul away.

         3. Ross, is someone actually proposing to dig a twenty foot deep trench next to an active LR track? That’s what would be required for a “single-tracking” method.

            Good Lord, Tom — just read the report. I’m tired of constantly quoting from it. If you think the engineers made a mistake, explain why. But you are basically just shooting from the hip and ignoring everything that was written.

         4. Fair request, Ross. I have read the entire Trench portion of the report now and (thankfully, given the clear risks) find no mention of a single-tracking option. I don’t know where you got that. Maybe somebody else proposed it?

            They do include the need for Jersey Barriers.

            They don’t consider any of the operational difficulties of orphaning the south portion of Line 1. I guess that was out of scope or perhaps outside their expertise. And nowhere I see is there an estimate of the time needed to complete this huge excavation. I think my five year estimate is conservative, simply because of the enormous volume of earth to be removed and the stations. It’s half again as much spoil as twin bored tunnels would require.

            They do say that during construction the project would bulge into the adjacent traffic lanes, probably on both sides, so my observation that the bus bridge would be caught in construction congestion is implicitly correct, since the roadway will be significantly narrowed. In some stretches that means single-laning in each direction.

         5. I have read the entire Trench portion of the report now and (thankfully, given the clear risks) find no mention of a single-tracking option. I don’t know where you got that.

            Page 16, under the CONSTRUCTION IMPACTS AND MITIGATIONS section (where you expect to see it):

            Construction phasing alternatives for maintaining a single track in service during construction could be warranted. This would include studying constructing the trench in halves, or as a whole with a shifted alignment, however substantial challenges would be anticipated.

            This is the way this is supposed to work. It is not a detailed study. But it is detailed enough to know it is plausible. At some point they would have to make decisions that involve a trade-off between disruption versus cost, just as they will eventually have to do for the new tunnel. But to just dismiss the thing as impossible is similar to the Montlake-vent-shaft myth that persisted on this blog until a Sound Transit representative said it was untrue. Or the idea that a line from Westlake to Ballard is impossible because there is nowhere to put the trains (until ST confirmed that they could store them in Interbay). That myth came up the other day (in a different conversation). These things can be done and until we get detailed information, we don’t know how disruptive they will be. But at this point, the initial engineering looks promising, simply because the cost is not huge. If they estimated it would cost $5 billion (which is less than West Seattle Link) then I would say there is no hope. But that clearly isn’t the case.

         6. Construction phasing alternatives for maintaining a single track in service during construction could be warranted. This would include studying constructing the trench in halves, or as a whole with a shifted alignment, however, substantial challenges would be anticipated.

            So, yes, you are correct. I did not see that reference, just the stuff in the previous paragraph.

            I still think it would be unbelievably difficult to dig that second half with the first half completed. There would have to be some sort of pretty strong temporary wall to keep debris from the excavation on the second side from falling into the in use tracks. But that just postpones the mess until the time to demise the intermediate wall. Obviously, the wall could be built to be permanent and the entire structure a foot or a bit more wider. There would have to be pass-through doors for use once both sides were opened.

            So I guess it might be possible.

            Really, the most practical way to do a trench is to dig it in one of the roadways while keeping the existing trackway running as normal. I guess that goes under the rubric of “as a whole with a shifted alignment.”Then convert the tramway to a two-lane roadway directly adjacent to the surviving half. That would certainly be the inexpensive way to do it because it could be done all at once, but it would very likely run into a buzz saw of public outrage about closing down that much car capacity during construction.

            There are sanitary sewers on both sides, but the City Untilities map shows them close to the curb.

            So far as the money, why would it be less than $5 billion? It requires digging a four mile long trench twenty or a bit more feet deep and at least that wide, building a double track railroad in it with overhead power distribution, and FOUR underground stations (assuming Graham survives). If it would cost $200 million to widen and flatten the trackway at Graham for one-thirtieth (admittedly, a very rough estimate) of the length of a trench, it’ll cost $5 billion.

            There’s no cheap way to move five hundred thousand tons of soil (an equally rough estimate from Gemini). Remember how WSDOT crushed the Battery Street Tunnel decking and just left the rubble in the hole? They did it to avoid hauling it off, and it was less than a mile.

            Now, yes, I understand that ST is pretty good at hauling away spoils from all their tunneling, or at least, their contractors are. But it’s still a huge amount of soil to dig up, haul away, and dispose of.

      2. Also, I don’t live in Texas; I did for eight years in the eighties and early nineties. Before that I lived in Seattle for ten years and worked for Metro answering calls before it was all automated. We had a joke: “We’re the people who tell you where to go and where to get off!”

         I live in Vancouver (“Vancouver Not BC, Washington not DC”) now and have relatives and friends in Hansville and Seattle I visit regularly. I don’t know the details of the bus system today like I once did, but I do know the street system in Seattle intimately, including the topography on which it lies.

         And I know the state tax system in which the Leg says loudly and clearly to Seattle: “You don’t have any taxing authority WE don’t give you. And we’re going to be certain that you Libtards in Seattle can’t spend your taxes without ‘adult supervision’ by the Sububurbs all around you.”

         Don’t like the sarcasm? You probably have “connections”; get me banned. It’s all for naught these days. The Bored [sic] can’t be bothered to listen anyway.

         And an addendum to the previous. Holgate would have to be overpassed as well as Lander for Line 1 automation to extend to Mt. Baker. Lander is already budgeted in the grade level version of Line 3. (What a stupid thing; two parallel double-track railroads right next to each other for sixteen LR trains per hour, total. Oy-vey!)

         1. Tom, nobody is trying to get you banned. I simply find your sarcasm annoying, unproductive and stifling of genuine discussion.

   5. I think you have a valid point that the incident is not a mandate for a full grade separation, Tom. It’s mentioned in the article unfortunately. Sure a grade separation is an ultimate solution in the long-term but this needs to be addressed quickly. Discussing the long-term is untimely.

      I see prevention of similar incidents as best addressed with low-cost actions that could be completed in a month.

      I also see this as an SDOT task rather than an ST task. If someone misses a sudden turn in the road and rams into a house because of poor signage, it’s the fault of the road and not the house owner. This to me needs to lands on SDOT’s to-do list ASAP to add better signage, paint and things like that.

   6. I entirely agree. Why waste money to rebuild Rainier Valley and all the stations? It’d be incredibly expensive and disruptive, and provide no new service.

      Rainier Valley isn’t *that* slow… But it adds in the scheme of things for riders TIBS and South. A bypass line would be way cheaper and deliver on regional connectivity goals.

      Same amount of money, but actually serving new stations and new transit goals. It also will double the transit speed in S King.

      1. And doesn’t nothing to solve the safety or reliability issues of the valley. A bypass solves a completely different issue. Theyvare not like replacements

      2. Why waste money to rebuild Rainier Valley and all the stations? It’d be incredibly expensive and disruptive, and provide no new service.

         1) Faster.
         2) More frequent.
         3) More reliable.
         4) Safer.
         5) Improves MLK Way.

         A bypass line would be way cheaper and deliver on regional connectivity goals.

         A bypass line would provide very little new service as well. You’ve got maybe one station (Georgetown). That’s it.

         No one knows if building a bypass would be cheaper. It would not be faster, more frequent, more reliable or safer for riders in Rainier Valley or Beacon Hill. It certainly wouldn’t improve MLK Way. It would not improve the connection from Rainier Valley to the south, which likely accounts for a significant portion of the ridership there (based on rider directional data — https://seattletransitblog.com/2025/08/25/ridership-patterns-for-link-1-line/).

         You also have an awkward situation with regards to interlining. It is both a branch and a reverse branch. This is a timing nightmare. Imagine the bypass saves five minutes. Trains are running every ten minutes. A train goes south through SoDo, on the way to Rainier Valley. Another train goes by the same spot, five minutes later, heading towards Georgetown. They arrive at TIBS at the same time. Oops.

         Even if you time it correctly you are spending a fortune running trains to an area that gets relatively poor ridership per mile (and per service hour). This greatly increases maintenance cost (which is related to both). You have to buy more trains so that you can run twice as often, south of SoDo. You could turn back some of the trains after SeaTac but now you are forcing a transfer for those traveling between Rainier Valley and Highline College, one of the few actual trip pairs that are substantially faster and reasonably popular with Federal Way Link. Just what we need, another untimed transfer.

         Of course it would be crazy to do that right now. We already have a branch, with half the trains going to the East Side and half going south. They run trains every 8 minutes on each branch during peak and every 10 minutes midday. This is OK, if not ideal. But if you branched the second line again (with a bypass) then you are running trains every 16 minutes to Rainier Valley during peak and every 20 minutes midday. This is unacceptable.

         It is possible with the second tunnel of course. We could be running trains every 4 to 5 minutes to the Seattle Center. But we are still forcing transfers for people headed to the airport from the north end. Good luck timing the trip on the “express train”.

         It really only improves things for those commuting from the southern suburbs, an ever shrinking portion of transit riders. You dramatically increase maintenance costs after spending who knows how much building this line. It ignores one of the fundamental advantages of rail — you spend a lot of capital but you save a lot on operating expenses. There are a lot of people in Northgate or Lake City Way that would be better off with express buses to downtown but forcing them onto trains saves a lot of money (which and then be used to improve the system in other ways). But this would *increase* operating expenses with only a minimal increase in ridership (with all due respect to Georgetown riders). In contrast, burying the line in Rainier Valley would save a huge amount of money in operating expenses.

         I’m not saying we will ever be able to afford it, or that ST will have the good sense to favor it over other, dubious projects. But it would provide a lot more value for a lot more people than a bypass.

7. Very surprised that driver could go this far.

   1. A perk of driving a SUV I suppose

      1. I still don’t think that thing was high enough off the tracks to not have been significantly scraping. She has to have had a very special level of oblivion, impairment, and/or denial to have made it that far…

8. Now that’s what I call car brain

9. Did they need to close the NB tracks? Why? Can’t trains in both directions serve the NB platform? Of course, security would have to inform people boarding where to go.

   1. Did you see how they extracted the car? Much safer to pause all operations while extracting the car than to risk it swinging and hitting a train filled with riders.

   2. They had to shut off the overhead wires. It’s a good idea any time you’ve got an oversize object such as a crane at work.

   3. I wonder what was happening at the control center when she first entered the elevated portion. Since she was not fully high-centering the vehicle, no doubt she was shunting the signal circuit.

      1. Your question made me curious. I checked Link’s online scanner, set the date to June 2. One of the first calls that came in to the Link Control Center at 6:09:19 was from a Link Operator notifying the LCC of a car on the tracks at the station. It didn’t seem like the Controller was aware of this.

         If anyone wants me to post the link to the radio call, let me know.

         1. I’ve seen a rollback (flatbed) truck trigger crossing gates, while the car it was removing from the tracks hadn’t.

            It must require a substantial metal connection probably not available in a vehicle’s frame.

10. wouldn’t be a problem if it had been elevated from day one in MLK. And with Graham Street station only doubling down on the bad decisions Link is basically crippled for the foreseeable future south of CID. The bypass of Rainier Valley via marginal way is the only feasible fix

11. When the line was being designed half the people in the Ranier valley wanted a tunnel, the other half wanted at-grade.
    They pretty much unanimously rejected an elevated line.

12. Tourists, what are you gonna do?

    1. Can’t trust anyone over… 70[is the new 30?] !

13. This is very rare (only happened once so far), so we should probably do nothing at all about it.

    If it ever happens again, put some traffic spikes at the street entrance. Problem solved.

    And if we can’t even build Ballard Link, we certainly can’t grade-separate MLK Link. If collisions on MLK are really such a big problem, stop allowing vehicles to cross MLK. There are more people going by train than by car on MLK now, so the trains are more important. If it’s absolutely essential, add some underpasses or bridges at a couple places for vehicles to get to the other side (paid for out of Seattle’s road budget). Or in put in retractable gates and/or spikes.

    1. Everyone discusses grade separation by changing the location of the rail line.
       I’ve wondered how much it would take to reconfigure the intersecting streets to be underpasses (, and possibly closing minor crossing streets).
       I remember S 180th St/SW 43rd St being converted to an underpass to go under the rail lines for (at the time) around $40 million.

       Reconfigure left turn access on MLKing Way to be only via the underpasses.

       If one were to apply the inflation rate, we could speculate each intersection would cost maybe $60 million?

       Just spitballing here, since I haven’t heard anyone of any official capacity entertaining that approach.

       1. Jim, an underpass might work at Othello and at Alaska, but for Graham, Orcas, Kenyon, Cloverdale and Henderson the topography to the west favors an overpass, strongly on Orcas and Graham. On Orcas there’s a hill to the east as well.

          The problem for either an overpass or an underpass is that the cross streets are so narrow that the crossing street would be cut-off from King Boulevard unless ramps between the Boulevard and the over pass were also built. That would be really expensive and turn the Boulevard into a mini-freeway.

          1. Take 1 lane from MLKing for the ramp, the other lane can be the through lane and local access.

          2. Not sure there need to be ramps. Have right hand turns only and add roundabouts at the next intersection to allow drivers wanted to head left to to U turn.

             For example, for Orcas, the overpass is a single lane (each way). Install a single-lane roundabout at 37th and 33rd.

             The signalized intersections likely need to remain for the Truck Routes, which if I’m reading the SDOT map correctly would be Othello and Alaska.
             https://www.seattle.gov/documents/Departments/SDOT/FreightProgram/MajorTruckStreets.pdf
             https://experience.arcgis.com/experience/01fc8dac51d4456b8e1640a5b394de39

          3. Have right hand turns only and add roundabouts at the next intersection to allow drivers wanted to head left to to U turn.

             Yeah, exactly. It is worth considering the roadway in relation to other streets. It is not like Bothell Way, which has unique functionality for most of the way. There are alternatives here, even before you improve the ability to make a U-Turn. For example, consider this trip by car. A left turn on Alaska is the obvious choice. But if we don’t allow left turns then this is an option: . This is for short trips. For longer trips there are bound to be alternatives that cost you very little. For example this trip here. The fastest option is to head up MLK and turn left on Orcas. But taking a left onto Cloverdale — a couple blocks from MLK — is just about as fast.

             The key is just accept the fact that drivers will have to go out of there way sometimes. Here are a couple examples from the Netherlands. (I was actually trying to find a different underpass but that is the idea.)

             This approach also reduces the number of underpasses you need. You really only need a relative handful of crossings. Yes, this does run the risk of higher speeds on MLK. Traffic would flow better there. But that is an issue with streets that have lots of crossings now (like nearby Rainier) and dealing with that is a different issue.

          4. Graham, Orcas, Kenyon, Cloverdale and Henderson the topography to the west favors an overpass

             Sure, but you could also make an underpass work in most of these cases. My guess is the locals would strongly favor an underpass versus an overpass. For example:

             Graham — It is steep to the west but it levels out before it gets to MLK. For about 100 yards it is flat. That’s cutting it close but I think it could work.

             Orcas — Similar but it starts gaining altitude more slowly before it really picks up around 32nd. I think it would be easier in general. I think you start at the ramps at 33rd. It is a little tricky because the street isn’t that wide. You would probably have to take out some trees just to allow local access (and right turns).

             Then there is pedestrian access. Theoretically it is OK to have beg buttons timed with the trains. The study looked at pedestrian access to the stations (via tunnels and overpasses) and it isn’t that expensive. But I could also just see using beg buttons, based on the timing of the trains. Everything should be based on the trains, basically. Once there is a gap with the trains, the light turns to walk and all the cars stop. That might mean a long wait but there should be gaps even if the trains are faster and come more often. Having the option of taking an over/underpass as well is ideal. It would be like various places on Aurora (where you can wait for the light or cross over).

          5. “I could also just see using beg buttons, based on the timing of the trains.”

             Isn’t that what we have now? And people run the light because they don’t want to wait for the next train. And it’s still two tracks to cross so a long distance.

       2. They never studies the ideas of automotive under/overpasses. The focus was on pedestrian safety (people getting hit by the train) which is why they studied pedestrian overpasses/tunnels. Those are fairly cheap but are often ignored and don’t have as much benefit as grade-separation. Basically the problem is the cars and you are right — it might be cheaper and easier to deal with crossing traffic instead of moving the tracks. Another study would have to be done.

       3. Redesigning MLK. Is a fun thing to envision yet expensive thing to do.

          However, I think that the more timely, needed solution is simply for SDOT to be assigned to develop new striping and signage and delineations to the street. Things can be done this month — as opposed to developing a project that would take years to make a reality.

          Right now, there is scant visual recognition that I train could be coming down the tracks until the crossing bells go off.

          If SDOT and ST simply put the effort in to better delineate the tracks to be as prominent as a protected bicycle lane, I’m convinced that MLK accidents would drop significantly.

          SDOT has created all these various design treatments for protected bicycle lanes to discourage cars from infiltrating them. Yet the effort isn’t replicated in MLK for train tracks. That’s what is needed first. .

    2. And if we can’t even build Ballard Link, we certainly can’t grade-separate MLK Link.

       Making MLK grade-separated would be a substitute for the other plans. It is like Ballard-to-UW. There is no way we are building everything we now have planned and Ballard-to-UW as well. But if we restart the planning process for Link expansion it is quite likely Ballard-to-UW appears on top. MLK grade-separation is similar. Again, what is striking is the price tag. It costs about 10 billion for a tunnel. Forget about it. But the initial estimate is 1 billion for a trench. This idea deserves further study.

Comments are closed.