What the Brightline is best known for is not that it reflects the gleam of the future but the fact that it keeps hitting people. According to Federal Railroad Administration data, the Brightline has been involved in at least 185 fatalities, 148 of which were believed not to be suicides, since it began operating, in December 2017. Last year, the train hit and killed 41 people—none of whom, as best as authorities could determine, was attempting to harm themselves. By comparison, the Long Island Rail Road, the busiest commuter line in the country, hit and killed six people last year while running 947 trains a day. Brightline was running 32.
In January 2023, the National Transportation Safety Board found that the Brightline’s accident rate per million miles operated from 2018 to 2021 was more than double that of the next-highest—43.8 for the Brightline and 18.4 for the Metra commuter train in Chicago. This summer, the Miami Herald and a Florida NPR station published an investigation showing that someone is killed by the train, on average, once every 13 days.
Floridians have started calling it the “Death Train” and maintain a sense of gallows humor about it, saying that it must be “fed” regularly to keep hurricanes away. Train attendants told me that Brightline engineers and conductors sometimes darkly joke about earning a “golden ticket”—which is when the train hits someone at the right time so that the three paid days off a worker gets for emotional distress are rolled into a weekend that takes up most of the week.
...
Federal agencies have investigated the Brightline incidents and produced no firm conclusions about why they have happened so often. The company, sometimes called “Frightline” on the local news, has not been found responsible for any of the deaths. How could it be responsible for people driving around lowered gates or walking into the clearly delineated path of a train? Yet there must be some explanation for the unusual number of fatalities.
...
The Brightline runs on the route of the original Florida East Coast Railway, which was built in the late 1800s by Henry Flagler, a Standard Oil tycoon. Flagler is popularly credited with “inventing” modern Florida: His railroad allowed for the development of swampland into a series of luxury resorts dotting the coast. Everything grew up around this track—it’s the vein running through all of the oldest cities and most densely populated areas of South Florida.
...
As a result, once-familiar environments have been transformed. Take, for example, the story of Joann DePina, a 49-year-old mother of two who was killed by a Brightline train in January. DePina was walking over the tracks that cut through her neighborhood, but she was doing so on a well-worn footpath. She was technically trespassing, but there weren’t any fences or no trespassing signs, and it was a logical thing to do. DePina rented a room in a sober-living house on one side of the tracks and was crossing to get to a group meeting on the other side. She had been in recovery since 2017 and was saving money to move into her own apartment.
I walked along the tracks with her aunt Maria Furtado in May. Furtado showed me the footpath, next to the white cross she’d put up in her niece’s memory. In person, it was clear why people would walk there: The tracks split the neighborhood in half, with tightly packed houses on one side and a row of businesses on the other. To get around the tracks legally would require walking down to an intersection to cross, then walking back, adding at least 10 minutes. Taking a shortcut over the tracks looks easy enough, and it was probably easy to do so safely during the decades when freight trains were the only traffic. Hence the worn path.
...
Many train tracks are elevated to cross above roadways. Others are sunken down to cross beneath them. But the Brightline’s track intersects flatly, or “at grade,” with the roads on much of its route, including the part that runs through central Miami.
Many states have undertaken grade-crossing-elimination projects over the past half century because they make train routes dramatically safer. On the Amtrak route between Washington, D.C., and New York City, the highest-trafficked stretch of train track in the country, there are no grade crossings. The last one was eliminated in the 1980s.
There are 331 grade crossings along the Brightline route in South Florida. James Hopkins, a former Brightline conductor, cited this when explaining to me why he no longer works for the company. He mostly enjoyed his time at Brightline, he said—the company was a good employer—but he didn’t want to work on that route anymore in large part because of how often the train would hit people. At his previous job operating a freight train in the 200-mile stretch between Pittsburgh and Harrisburg, he said there were 40 to 50 grade crossings. In the 65 miles between West Palm Beach and Miami, there are 174. “It’s just real busy,” he told me. “The fatalities—this was just something I didn’t want to continue doing.”
When I visited the West Palm Beach area to look at the crossings and roads in person, I drove over the tracks dozens of times. They cut through the landscape at strange angles and in unexpected places—behind the downtown courthouse, alongside a Little League field in Delray Beach. People have been struck and killed by Brightline trains at both of these locations.
...
Brightline says that it is an advocate for closing certain crossings on its route, but that this rarely happens “without local support.” Because of all the elements at any intersection, the process of closing even one crossing can be convoluted and expensive. Sealing off the entire Brightline route or elevating the entire track would simply not be economically feasible for a private company.
Still, over a period of months, I spoke with several experts who had different opinions on many of the technical details but who all agreed that there’s no real mystery behind the Brightline deaths. “Fast trains and grade crossings are always a deadly combination,” the historian Richard White, whose 2011 book about American railroads was a Pulitzer Prize finalist, told me. He put it the most succinctly, but I did not talk with anybody who disagreed with that conclusion.
Comment box Scope: comment response, information, opinion Tone: cynical/jaded but optimistic that change remains possible Opinion: yes Sarcasm/humor: none It's nice for popular media coverage to...
Exemplary
Comment box
Scope: comment response, information, opinion
Tone: cynical/jaded but optimistic that change remains possible
Opinion: yes
Sarcasm/humor: none
It's nice for popular media coverage to correctly place the blame on the lack of investment in infrastructure. As the article notes, this is not unique to Brightline at all. They have more grade crossings than some comparable routes... the problem isn't unusual though.
Sealing off the entire Brightline route or elevating the entire track would simply not be economically feasible for a private company.
Yes it would. Being "a private company" does not make visionary projects impossible. Just not feasible for Brightline, whose lack of grade separation forces lower maximum speeds which forces longer travel times which lowers demand which lowers ticket revenue which lowers income. Their "B" bond rating is so poor because they're locked into a difficult right of way. That hurts their credit, which makes it harder to get loans to pair with already insufficient grants to improve the ROW. And so on. Same old same old! Chicken and egg!
Capital projects almost always happen with government funding anyway. Brightline West is being built because of a multi-billion dollar federal grant. Brightline Florida's grade separation would be no different. If we want to get real.....freight railroads often grade-separate urban and suburban tracks to improve throughput/reliability and reduce crashes. Flyovers are common on high-capacity lines. Freight isn't likely to grade separate 250 miles of track at a time, but they've collectively done well more than that.
We grade separated every interstate highway. Almost 50k miles of grade separation. We can do the same for the remainder of our passenger rail network. It's not too hard and it's not too expensive.
Because of all the elements at any intersection, the process of closing even one crossing can be convoluted and expensive.
There are technical challenges with grade separation in tight physical areas. It is also expensive. Could be $4m or $40m for a typical intersection, $292m+ for a uniquely nasty complicated one (rare). $40m is a lot. Can't deny it. But it's not... too much. $12 billion for the whole route? Ok, fairly big project at a DOT, but not that big for a 250-mile corridor. Not unprecedented at all. My state DOT is spending half that much to rebuild like 3 highway exits on I-95 for marginal improvements to congestion. If we're seriously worried about cost, one of these is clearly a better choice of funds than the other, but we're choosing not to do it.
Brightline says that it is an advocate for closing certain crossings on its route, but that this rarely happens “without local support.”
Emphasis on this coded speech about NIMBYism. It's difficult to make safety improvements because without fail, small groups of local residents find ways to oppose literally all infrastructure projects, no matter how beneficial, including those which would save lives. They're not centralized, it's just a theme. Sometimes this is justified! For public safety projects, usually the outcry is an overreaction. In either case, this country defers so heavily to local preferences in situations where they shouldn't (while selectively ignoring local opposition anytime they want to build another highway), like common sense grade separation, that nothing gets done. Whether or not a particular opposition position is narrowly justified is unimportant if it consistently stops anything from getting better. Sometimes we just have to accept that things change and no solution is perfect.
Local groups will always oppose closing a grade crossing because they want to maintain the convenience of crossing there, even if someone dies occasionally. Individualistically, I get it. No one likes a detour. Tragedy of the commons. But they will also rarely themselves voluntarily help pay for a grade separation because no one wants to pay higher taxes, especially those who can afford it. The only time widespread grade separation gets done consistently is when it has strong state and/or federal DOT support. It can mostly only be done with a dedicated body of technocrats pushing for it in a spreadsheet in the capital. Or by an extremely motivated set of public activists and a genuinely thoughtful sponsoring legislator. Not so common.
The current anti-intellectual administration has notably canceled grade separation projects in California as part of a vitriolic political revenge game against its perceived opponents. Administration has also indefinitely halted many Bipartisan Infrastructure Law grants, including numerous grade separations, for "reconsideration" merely because that law was championed by Joe Biden. Completely pointless.
The technical problem has been solved centuries ago. We've understood the concept of grade separation since the first time we ever built a bridge over a canal. The railroads have been doing it since the 1800s. The money exists in our economy and in our government budgets to do this. We collectively choose not to do it for a confluence of unrelated local and political reasons.
Ive said it before on tildes but I drive freight trains for a living. Before getting very far in the article I knew where this was going. Large population centers plus interactions with trains...
Ive said it before on tildes but I drive freight trains for a living. Before getting very far in the article I knew where this was going. Large population centers plus interactions with trains equals people getting hit.
Local groups will always oppose closing a grade crossing because they want to maintain the convenience of crossing there, even if someone dies occasionally. Individualistically, I get it. No one likes a detour. Tragedy of the commons. But they will also rarely themselves voluntarily help pay for a grade separation because no one wants to pay higher taxes, especially those who can afford it.
If a crossing gets closed without doing a grade seperation then often you'll have pedestrians still crossing there. You've reduced the risk... for cars.
Even some grade seperations don't eliminate risk. I was stoped once at a location with a major road underpass. Litterally as I was about to pull, a passing train warned me that a man with a bicycle was crossing over my train. People get killed doing this...
Another thing that I've seen that's less than ideal are pedestrian overpasses/bridges. People aren't going to climb that when they can just walk over the track so you're back to the same problem.
Elevating the rail is ideal for safety but you can't really just pop the tracks up and down, you need to have a long approach and descent, even longer for freight trains. As a result this makes the most sense in denser areas where you're eliminating multiple crossings in one shot.
It really is mainly an issue of money like you suggest.
I'm just glad that my terminal doesn't have too many incidents and the route I go REALLY doesnt. The only thing we regularly hit are animals. I don't like killing animals but it beats killing people. :-/
In college, our (very large) campus was bisected by an at-grade train line. It was very inconvenient if you got stuck on one side with a class starting on the other, though most faculty were...
Even some grade seperations don't eliminate risk. I was stoped once at a location with a major road underpass. Litterally as I was about to pull, a passing train warned me that a man with a bicycle was crossing over my train. People get killed doing this...
In college, our (very large) campus was bisected by an at-grade train line. It was very inconvenient if you got stuck on one side with a class starting on the other, though most faculty were understanding. Early on, the train would regularly stop on the tracks blocking the intersection. I saw a guy attempt to carry his bike and climbed between the cars. The train started suddenly, he dropped his bike on the tracks, and the train cut it in half. Very lucky that was all that happened.
There was a pedestrian overpass, but so far from the main traffic point that I never used it or saw anyone use it during my whole time as a student there.
Eventually they made some change to the train line so that the trains no longer stopped on the tracks on campus.
As I was graduating, they were taking on a major capital project to make pedestrian underpasses and some provision for the bus lines. Though surprise, surprise, the main goal of the project seemed to be making it easy for crowds from the west famous commuter lots to get to the football stadium.
IIRC in the US there is no rule or law preventing trains from just blocking a crossing all day, which us crazy. I may be wrong, I don't know US operating rules. In Canada we can only be stopped on...
IIRC in the US there is no rule or law preventing trains from just blocking a crossing all day, which us crazy. I may be wrong, I don't know US operating rules.
In Canada we can only be stopped on public crossings for 5 minutes after a car shows up. The loophole if you block a crossing accidentally is to pull at 1 mph for as long as possible until you can go again.
Illinois has such a law ILCS It's a fine, not prosecutable, but it really changed how trains were blocking the road constantly in my hometown when this was passed. Officers could just stand there...
It is unlawful for a rail carrier to permit any train, railroad car or engine to obstruct public travel at a railroad-highway grade crossing for a period in excess of 10 minutes, except where such train or railroad car is continuously moving or cannot be moved by reason of circumstances over which the rail carrier has no reasonable control.
It's a fine, not prosecutable, but it really changed how trains were blocking the road constantly in my hometown when this was passed. Officers could just stand there and ticket every ten minutes or however it escalated.
So you'd have to check state by state and even city by city to be sure. It may be the case that this law isn't enforceable anymore, due to federal rulings too, I haven't truly kept up on it and that's always a risk.
In this case, the town is literally named College Station because the university is basically the town. And they have a strong alumni network all over the state. So they probably have more...
In this case, the town is literally named College Station because the university is basically the town. And they have a strong alumni network all over the state. So they probably have more influence than normal.
I assumed they restructured the blocks (is that the right thing?) so that there was enough room outside campus for the whole train.
Ya, thats the right thing, and that makes sense. Building the system out in a way that you have enough heads up to stop in a good spot is definitely a strategy. BTW, blocks are kind of like the...
I assumed they restructured the blocks (is that the right thing?) so that there was enough room outside campus for the whole train.
Ya, thats the right thing, and that makes sense. Building the system out in a way that you have enough heads up to stop in a good spot is definitely a strategy.
BTW, blocks are kind of like the zones where the system can detect a train (I don't know how to better describe that). You can have multiple blocks between signals or you can have one signal per block (right at the spots where the blocks change). So you usually have to change the blocks to move a light.
Comment box Scope: comment response, question Tone: neutral Opinion: none Sarcasm/humor: none Thanks for your perspective. Great point about pedestrian activity. Have you anecdotally noticed any...
Comment box
Scope: comment response, question
Tone: neutral
Opinion: none
Sarcasm/humor: none
Thanks for your perspective. Great point about pedestrian activity.
Have you anecdotally noticed any wildlife crossing pattern changes in areas where we’ve built grade-separated wildlife crossings?
I’ve read some studies suggesting that they reduce crashes, but I’d love to hear from an actual train operator.
We don't have any of those where I am so I'm not sure. Familiarity seems to be a big factor though. We had a new elk herd move in and they seem to get hit way more than the other elk herd 100...
We don't have any of those where I am so I'm not sure. Familiarity seems to be a big factor though. We had a new elk herd move in and they seem to get hit way more than the other elk herd 100 miles away.
Strangely we have an area fenced off to prevent moose from getting hit. It's not the direction I normally travel though and it was in place before I hired on so I don't know about a before and...
Strangely we have an area fenced off to prevent moose from getting hit. It's not the direction I normally travel though and it was in place before I hired on so I don't know about a before and after.
I responded to @scroll_lock about wildlife crossings but I forgot to mention the wildlife fences. They do actually work and the fenced area ends on each side with a small bridge. The bridges act like a cattle guard and is a safe spot to cross (under) the tracks. Once in a while a moose gets in there and it usually ends in them getting hit.
When reading about the first death in the article, I wondered about upgrading informal crossings to signaled pedestrian crossings, and adding fencing nearby so that’s the obvious way across. With...
When reading about the first death in the article, I wondered about upgrading informal crossings to signaled pedestrian crossings, and adding fencing nearby so that’s the obvious way across. With a sign warning that it’s a high-speed train crossing.
There would still be a potential for mishaps because people will still do dumb things, but it seems like it should help? Cutting deaths in half would be a good start.
I've seen pedestrian crossings built on walking trails just without warning lights. Trains have to whistle approaching them just like a regular crossing. It would probably work by kind of...
I've seen pedestrian crossings built on walking trails just without warning lights. Trains have to whistle approaching them just like a regular crossing.
It would probably work by kind of encouraging people to CROSS instead of lingering. Also giving people a nice path instead of hiking up the grade and hopping over rails. Thats all without warning lights.
Throughout, when concrete points against Brightline are made in this article, they seem much more dubious when examined. The added photos, likewise, seem designed to push a 'scary' narrative...
Throughout, when concrete points against Brightline are made in this article, they seem much more dubious when examined. The added photos, likewise, seem designed to push a 'scary' narrative rather than honestly support the points. The overall conclusion – that level crossings and higher-speed trains are a bad combination – is reasonable, but the narrative sometimes ends up reading more as a hit piece against Brightline specifically. In some sense, it feels like exactly the sort of hypocritical NIMBYism that so often appears in the US: alternatives to cars should be lauded, except when there is any attempt to actually build one.
Consider the comparisons. The article argues that Brightline has particularly horrendous statistics for Florida. It points out that Brightline has almost seven times the number of fatalities as Amtrak, which "operates through many of the same urban areas as well as some additional ones", and merely "serves fewer passengers." Yet that number seems much less surprising when considering that it appears Amtrak's service in Florida consists of one daily train and one thrice-weekly train, while Brightline appears to run ten round trips per day now, and may have run more previously (Wikipedia lists 24 daily round trips in 2023). When comparing to SunRail, "another commuter train that operates around Orlando", the article uses the rate-per-million-miles that would have made the Amtrak comparison much less compelling to the narrative, but doesn't mention that SunRail, according to Wikipedia, has an average speed of 30 mph.
Or consider the argument that the track cuts "through the landscape at strange angles and in unexpected places". The accompanying photographs enhance this by using strange camera angles and crops, rather than showing the places the article describes. If they were to show a satellite image of one of those "unexpected" places they give as an example, the little league field in Delray Beach, you might be confused: it is a straight segment of track, parallel to a road, and appears to be fenced on the little league side. In what way is this "unexpected". Was it an example of a "strange angle"? All the nearby level crossings, while I might argue they should not be so near other intersections, are almost perpendicular.
Or the examples? A pedestrian walking on a spur path over an active train track, argued to be the 'logical' choice? A driver who pulled around a stopped car and closed gate, was high, and was directly told not to by a passenger, but is argued to be an example of a dangerous environment regardless of all those details? A couple who turned off their car and sat on an active train track with an active signal? One might suggest that an encouragement to not care about crossing tracks, whether on foot or by car, comes from an encouragement of car culture: "they're probably old tracks that aren't used much" is a common suggestion I've heard when seeing tracks.
Yes, level crossings are dangerous, especially at higher speeds. Yes, more fencing would be an improvement. But looking at the route, where the most dangerous segments seem to be along a previously-existing route, it seems clear that no amount of expenditure alone would allow for a major safety improvement, as any attempts would face an enormous NIMBY backlash. There's a certain argument that the encouragement of dense, especially residential development right next to historical rail routes was perhaps linked to a cynical attempt to make any future use of the right of ways difficult and unsafe, ensuring cars remained the only viable transport mechanism in the area. Adding fencing in many places seems like it would be an inconvenience to nearby houses. Elevating the rail in many places seems like it would significantly affect nearby houses, and potentially need land purchases, voluntary or otherwise. None of this is going to be acceptable to local property owners.
What is needed here are regulatory and cultural changes to make it so that safety improvements can actually be made; unfortunately, this is something where spending money on lobbying and marketing is probably better than spending it on trying to build infrastructure projects that will end up blocked.
Those seem like good points about the statistics. I was with you until the end, where you seem to be arguing that nothing worthwhile can be done. This seems defeatist? What makes you so sure?
Those seem like good points about the statistics. I was with you until the end, where you seem to be arguing that nothing worthwhile can be done. This seems defeatist? What makes you so sure?
I think some editing to the comment ended up not being saved: the point was that regulations and culture need to change in order to allow safety to improve without being blocked by NIMBYism and...
I think some editing to the comment ended up not being saved: the point was that regulations and culture need to change in order to allow safety to improve without being blocked by NIMBYism and the cultural primacy of cars, and that spending money to that end is probably the better option than wasting it on planning the necessary safety improvements, fighting to put them in, and then losing.
I wonder if there is some relatively low-cost way to add a signal to an informal pedestrian crossing, such as the first example in the article, and add fencing next to it. Sure, grade separation...
I wonder if there is some relatively low-cost way to add a signal to an informal pedestrian crossing, such as the first example in the article, and add fencing next to it. Sure, grade separation is best, but if the money isn’t there, something could still be done.
Comment box Scope: comment response, information, opinion Tone: neutral Opinion: yes Sarcasm/humor: none For sure physically possible. I live near a railroad abutting a riverside pedestrian trail....
Comment box
Scope: comment response, information, opinion
Tone: neutral
Opinion: yes
Sarcasm/humor: none
For sure physically possible. I live near a railroad abutting a riverside pedestrian trail. There are multiple at-grade pedestrian/bike crossings on this ACTIVE freight line in an urban area. The railroad is entirely fenced except for these ~3 crossing points. The intersections have crosswalk lights which are supposed to display a “Stop” walk signal when a train is approaching. In theory the intersections could be further upgraded to raise automatic barriers to physically block access (like for a rail/road crossing, but maybe more effective with sufficient pedestrian fencing). The walk signals could also be better placed for visibility, and an auditory signal could be added—this all might help, won’t hurt, and it’s not too expensive. It’s mostly a moot point because trains only occasionally run here during daytime.
The crossings are important for the trail and eliminating them outright would incentivize dangerous behavior as @mild_takes mentions. People would use a pedestrian bridge if forced, but there are space constraints and an ADA-compliant ramp may not fit; an elevator would be expensive and liable to break. Signalization and utility work is always complex, and the city would need to work with the railroad to do anything. It’s annoying to do - and it’s no one’s priority - so the city has just left it as-is. Imperfect without trenching the railroad. Statistically unlikely to lead to any fatalities (just not high enough traffic/speeds), but it could happen. If someone dies, the local activists will raise hell and maybe get a ped bridge.
No solution is perfect. These stop-gap solutions like signals are vulnerable to human error. People are irrational: they ignore direct warnings, they climb over fences, and they make mistakes (trip and fall, misjudge distance, etc…..). Sight/hearing impaired people might not comprehend the signals. Drunk people do drunk stuff. Teenagers intentionally evade fencing to go surfing. Movable barriers electronically or mechanically fail. People even walk long distances on grade-separated or fencing tracks from unfenced areas (for whatever reason) and then get stuck.
Grade separation is still the best solution though. The money exists in our tax base and economy. It’s a societal, philosophical decision—do we value Vision Zero more than equally expensive highway widening projects?
I’m not sure where you’re going with this “the money exists” stuff. Sure, there is money out there somewhere, in theory, and the Fed could create more money. But real projects operate under budget...
I’m not sure where you’re going with this “the money exists” stuff. Sure, there is money out there somewhere, in theory, and the Fed could create more money. But real projects operate under budget constraints.
Efficiency matters, particularly for expensive infrastructure projects. For the cost of one grade separation, what could you do to improve safety along the entire rest of the line?
From a broad historical perspective, today’s prosperity is built on yesterday’s efficiency improvements.
Comment box Scope: comment response, information, opinion Tone: resolute, optimistic Opinion: yes Sarcasm/humor: none Projects have budget constraints. We choose what those projects are, and their...
Exemplary
Comment box
Scope: comment response, information, opinion
Tone: resolute, optimistic
Opinion: yes
Sarcasm/humor: none
I’m not sure where you’re going with this “the money exists” stuff. Sure, there is money out there somewhere, in theory, and the Fed could create more money. But real projects operate under budget constraints.
Projects have budget constraints. We choose what those projects are, and their scope. It's not "in theory" - it's "in reality" and it doesn't involve wrecking the economy. (Other countries do this better than the USA, it's not trailblazing) Similar idea as why we fail to feed everyone in the country despite producing more than enough food to do so: an issue of distribution, not a lack of abundance. But this might be simpler to fix... no need to upend capitalism, just build some bridges.
State and federal DOTs in the USA are allocated enough money with current tax structures to complete transformational, visionary capital transportation projects, especially for passenger rail. MPOs and DOTs choose, repeatedly, to allocate an oversized portion of that funding to inefficient roadway projects that at best maintain the status quo and usually fail to deliver on promises to resolve congestion or holistically improve safety. They justify this car-centric spending allocation with a combination of institutional inertia, pseudoscientific/circular traffic modeling projections, and fear of political backlash. Even "progressive" MPOs like in Seattle and Philadelphia lock themselves into self-fulfilling prophecies about car centricity. Planners individually come up with lots of great ideas, but institutional knowledge in these places favors the dominant cultural milieu, which gives high roadway use the benefit of the doubt by default. (Source: I talk to these kinds of people a lot, also I just spent a lot of time reading my MPO's new long-range plan. It's okay, but it still makes assumptions that incentivize overindexing on car infrastructure. The DOT will ignore most of it anyway to build wider highways.)
Roadway projects induce more roadway use. Roads are some of most self-destructive and least inefficient forms of transportation we have ever devised. No other form of ground transportation deteriorates at the rate of a highway, an inefficiency inherent to the mode. Most vehicles have single occupants despite weighing a ton or more. For the number of people a highway transports, its vehicles exert a lot of force onto its asphalt. (The "weight of people transported" to "weight of vehicle" ratio is bad.) That loosens asphalt particles, which enables water filtration, which creates cracks and potholes that freezing weather exacerbates. Paving roads with more durable materials raises costs and has various engineering problems. Every attempt to make highway vehicles more efficient leads to 1) a bus or 2) a train.
Major resurfacings have billion-dollar price tags that hardly anyone questions. Multi-billion dollar widening and "traffic optimization" projects repeatedly fail to resolve urban and suburban congestion. Complex car-centric signalization projects make marginal improvements to design capacity while worsening pedestrian safety and disincentivizing walking. More and larger roadways have higher maintenance costs, which creates the appearance of a "tight budget." Roadway maintenance projects are generally prioritized ("we can't maintain our existing roads, so we definitely can't build your train line") even though a mode shift toward rail would be more cost-efficient (and safer) in the long term.
All choices! We continue to overbuild roads because it's a cultural preference, not because it's efficient, effective, or good for society.
The mileage of urban/suburban passenger rail services in the USA is quite low. We grade separated 50,000 miles of interstate highway crossings (it's part of the design standard from... 1957!!!). The Northeast Corridor is like 457 miles. Brightline Florida is around 250 miles. Even every single Amtrak route including freight lines is ~21,000 miles (mostly rural with few crossings and fewer fatalities per crossing than urban/suburban). We have done much harder, much more expensive things in the past. We can do this if we prioritize it!
Efficiency matters, particularly for expensive infrastructure projects. For the cost of one grade separation, what could you do to improve safety along the entire rest of the line?
Not much. We have done most of the easy stuff, like signaling, train speed limits, and basic gates. Trespassing + grade crossing collisions are collectively the source of ~95% of passenger rail fatalities in the USA: look at (Trespasser Fatalities + Highway-Rail Incident Fatalities) / Total Fatalities.
Most VRU deaths (people outside vehicles) on railroads are caused by trespassing, which mostly begins by a pedestrian accessing a railroad from an at-grade crossing. (People rarely access railroads from random places, instead they come from another road/path which intersects with it.)
Most passenger deaths are from derailments of some variety, which are mostly caused by cars on the track (aka highway-rail incidents). Cars also overwhelmingly access railroad tracks from grade crossings.
Debris, signaling issues, poorly maintained tracks, failing rolling stock, and other misc causes can cause fatalities, but this is a small proportion of deaths. It's like 2 dozen a year (out of ~800-900). Worth fixing, just not the priority!
Fencing reduces pedestrian deaths. Financially, fencing is peanuts. You could chain-link fence the 500 miles (250 x 2) of Brightline Florida's route for ~$50 million, assuming an expensive choice of fence. But you can't fence grade crossings themselves. Grade crossings are the places where people access railroads, so to stop pedestrian trespassing, you have to grade separate anyway. Fencing also does not stop any train-car collisions.
4-way mechanical gates at grade crossings can also help. (Gates that cover both lanes, so you can't drive around them.) These are worth investing in! They don't stop pedestrian trespassing though, and they don't stop cars from driving onto the tracks, they just make it psychologically less appealing (cars can often still drive around, off-road, or drive through them: they're flimsy to prevent people from getting trapped). They also mechanically fail easily, which reduces their effectiveness. (It would be possible to engineer stronger and more encompassing barriers at intersections, like huge gates that rise from the ground and fully block an intersection, which combined with fencing could stop some car collisions, but not in the case of driving onto the tracks at an "allowed" time and just staying there; and pedestrians could still trespass for the same reason.)
Platform screen doors make it harder to accidentally fall on the tracks at a station. This is not a statistically big safety issue, more of a social fear (it's overreported in media). It mostly keeps debris off tracks, which reduces electrical fires for third rail systems (not applicable to Amtrak or Brightline, which use catenary/conventional fuel). However, PSDs require high-level platforms, which cost about as much as a grade separation to build. (We need to make all station platforms high-level anyway for alighting efficiency reasons, so building PSDs on low-level platforms and then destroying them to rebuild the platform is a waste of money.) PSDs do not reduce pedestrian trespassing between stations though, only at them.
Upgrading signaling systems can allow higher safe speeds and conceivably reduce collisions slightly. Boston recently invested in signal system improvements to eliminate most slow zones on one of its subways. This is more of an efficiency improvement than safety per se.
Grade separations serve the dual purpose of improving operational efficiency and safety:
Federal guidelines limit non-grade-separated segments of train lines to 79 mph even if curve design and rolling stock are safely capable of higher speeds. At-grade crossings mean that even if we expensively straightened tight curves to allow higher design speeds, it would remain illegal and unsafe to operate at HSR speeds on many routes without also doing grade separation.
Grade separations reduce the frequency of non-fatal delays. This improves on-time reliability, which significantly improves commuter ridership. This is especially important on shared track, such as when Amtrak leases a track/operating rights to a local railroad like SEPTA or vice versa.
A time-efficient passenger rail network reduces automobile use. Trains are just safer than cars. By reducing the amount of VMT (Vehicle Miles Traveled) per person, we reduce collisions in general. Spending money on active transportation (walk/bike) and Bus Rapid Transit also reduces automobile VMT locally/regionally. This is actually quite cheap and quite efficient, with many independent benefits, but indirect for this purpose.
Sometimes it might be possible to avoid having to grade-separate tricky intersections by just building a new section of right of way. Unfortunately, American planners are infamously poor at efficiently spending money on transit/rail projects. They make inefficient designs using inefficient methodologies, and have relatively little interest in emulating global best practices. Sometimes this is the right solution, I just wouldn't count on it.
But grade crossings are well-understood as a component of road infrastructure and don't suffer as much from that problem as a bloated train station concourse design does. American DOTs can sure build road overpasses!
The best use of funds on a timeline is probably to grade-separate all intersections near population centers first, in order of proximity to the largest population centers. It's more efficient to do the urban/suburban crossings first and the rural ones later, because rural crossings statistically have less traffic and therefore cause fewer fatalities. Some exceptions apply. But this could halve the price tag of grade-separating "everything," or more.
I'll take it as given that some budgets for maintaining existing transportation infrastructure are very large and that the money could plausibly be spent differently. But this seems like too...
I'll take it as given that some budgets for maintaining existing transportation infrastructure are very large and that the money could plausibly be spent differently. But this seems like too high-level an argument? It would be more convincing to talk specifically about how the Florida Department of Transportation in particular manages its transportation budget and how it should be spent on different projects.
That would require doing homework, though. I don't want to give anyone a project; I'm just saying maybe we should be less confident that we know the answers based on general principles.
You say "we have done most of the easy stuff" and maybe that's usually true, but from the article, it didn't sound like it was true for this particular railroad line in south Florida? A convincing argument would have to be more specific to that particular line, explaining what it's really like and how the article got it wrong. (pallas made a good start on this.)
I don't see the relevance of some other things you're bringing up. Sure, there are almost 50,000 miles of grade-separated Interstate highways in the US. But that's money already spent. How do you get from there to budgeting for new projects? It seems like similar logic to saying that the New York City subway system exists, therefore we have lots of money.
Money already spent isn't a source of funds. Existing infrastructure is a starting point for future improvements. It can be an advantage or a disadvantage, depending on how it was built and how well or poorly it aligns with new needs.
Comment box Scope: comment response, information, opinion Tone: neutral, same kind of weird cynical/optimistic thing going on Opinion: yes Sarcasm/humor: a little bit making fun of government...
Comment box
Scope: comment response, information, opinion
Tone: neutral, same kind of weird cynical/optimistic thing going on
Opinion: yes
Sarcasm/humor: a little bit making fun of government agencies
Well, you make good comments, I'm saying:
that this project isn't relatively over-expensive on a micro level compared to benefits from a safety or transportation efficiency perspective. It's expensive but it's in the same class as other big projects FDOT does regularly.
we can do way more TOD and VZ infrastructure on a macro level from a financial perspective. I see why you think this is irrelevant. I am saying this to emphasize that we designed our whole world, and we can re-design it. It's fully possible. The sky is the limit. It's not too hard or too difficult in general. (Building the interstates was harder & more expensive at a time when we were poorer. so this shouldn't FEEL so hard.) I just think it's important to be optimistic. Our culture is often the opposite.
i also talk too much.
It would be more convincing to talk specifically about how the Florida Department of Transportation in particular manages its transportation budget and how it should be spent on different projects.
I can say with full confidence that the federal DOT and every state DOT allocate excessive funding to highway projects at the expense of passenger rail. In absolute dollars it's something like a 7:1 ratio nationally, and probably more like 14:1 in Florida (from memory, don't cite that). Dollars don't translate to impact, but this tells you at a high level what their priorities are.
A few DOTs do better, like California (because of CAHSR, SF, LA) and New York (because of NYC), but most of the land area in those states continues to receive car-centric capital investment. (Although New York has been removing a few highways, like in Rochester.) You don't actually need to look at the finances to see this: just look at ridership splits by mode: if Florida allocated funds to efficiently move people, it would already have at least one high-speed rail line connecting all its major metros, but it doesn't. (FDOT is only interested in efficiently moving vehicles, which is financially inefficient.) Florida rail ridership is accordingly low.
Some MPOs and RPOs (lol) are more progressive about planning than others, but it hasn't filtered through to the DOTs yet. They're bastions of hardcore traffic pseudoscience. They're extremely insular and are not interested in "activism." MPOs are also scared of that word, but they at least try. (Think of traffic engineers as people who like to solve math problems. In their spare time, they optimize Factorio production lines. They have a different education than urban planners and their priorities are typically these math problems, not societal change per se. Some exceptions, not a lot.)
I have not worked with Florida MPOs personally but I have not heard of them trying visionary things like Denver or Pudget Sound MPOs.
I don't see the relevance of some other things you're bringing up. Sure, there are almost 50,000 miles of grade-separated Interstate highways in the US. But that's money already spent. How do you get from there to budgeting for new projects? It seems like similar logic to saying that the New York City subway system exists, therefore we have lots of money.
Eliminating and narrowing highways would free up DOT budgets to focus on more efficient transportation infrastructure. So would choosing not to build more highways. Highways have an opportunity cost.
We do have lots of money. Our economy has grown since 1957 and so has our tax base. Quick maths: federal taxes alone collected in 1957 were about $83 billion and about $5.2 trillion in nominal dollars. Per-capita, that's about $487/person in 1957 and $15k/person in 2025. Adjusted for inflation, that's about $5000/person in 1957, so we have 3x the federal tax revenue as then. At the most theoretical level, we managed a staggeringly large, inefficient infrastructure project with 1/3 of the federal budget as present. (Note interstates were built over years.) Obviously budget priorities today include some systems that did not exist in 1957, but it's not like we have less money in absolute terms. And the choice to raise or lower taxes to cover any delta is still a societal decision that we have chosen to neglect.
Big-picture, these are projects on the time frame of a 25-year long-range plan. In the short term, of course Brightline has budget constraints with whatever grant they can obtain for a grade separation. But for how big these population centers are, it's still just not that expensive.
My earlier figure of $12b for the whole route is probably an underestimate. Let's say $20b from Miami to Orlando. FDOT capital budget in 2025 was $15 billion, ~5% of which was for railroads. Practically, it would take at least 5 years to grade-separate Brightline's whole route. If the state government paid for 100% of it, then over 5 years that would be $4b/yr. If they did top-priority separations in the first 5 years and low-priority separations in the following 5 years, then that's $2b/yr for 10 years.
Of course the state DOT will not pay for 100% of the cost. Brightline could finance maybe 1/3 of it. The federal government could chip in a substantial amount, maybe another third. So that means over 5 years the cost to the DOT is closer to $1.3b/yr or $650m for 10 years.
That is a big project. It's just not absurdly big. For context, the NYC Gateway Program is about $16b. Big project for sure! But only a portion of the budget. Meanwhile, FDOT are spending $500 million to add new lanes to I-4, $340 million to widen a small section of I-275, etc., adding up to a supermajority of that $15b budget. (There are more examples of big boondoggles, and the small ones are collectively expensive too.) If FDOT just decided to incentivize a modal shift to rail, it would cost the same amount or less upfront, and have lower maintenance costs. Transit-Oriented Development pre-empts wasteful highway spending. Alternatively, this could be easily paid for with new revenue using a small "cap and trade" tax (like California), small sales tax, small gas tax, or even (self-defeatingly) a small train ticket tax. The kind of person who comes to Tildes will complain about regressive taxes, so OK, a small income tax or property tax or corporate tax or whatever needs to happen. Fine with me. Florida citizens choose their tax system. Any politician could push for a ballot measure or legislature vote to make this happen. If they don't, it's because they choose not to: they ignore the safety benefits, they ignore the maintenance efficiency benefits, they ignore the broader economic development benefits. (They know what Brightline is, it's not obscure.)
You say "we have done most of the easy stuff" and maybe that's usually true, but from the article, it didn't sound like it was true for this particular railroad line in south Florida?
A grant for $24,934,138 announced in August 2022 to the Florida Department of Transportation for improvements to 330 highway-railroad crossings, along 195 miles of corridor, including fencing, crossing delineators, crisis support signage and other intrusion prevention mitigations.
195 =/= 250, but they have likely focused on urban and suburban segments, or this complements existing protections.
The problem is that these solutions just don't help much. People get onto tracks at grade crossings because they use roads to get to grade crossings. They're legally allowed to cross there. Then they illegally stay in the railroad ROW, and then they get hit by a train.
That announcement about the grants is a good find, but it sounds like the money hasn't been spent yet. (Announced in 2022, but apparently there was a "grant backlog.") It sounds like you're...
That announcement about the grants is a good find, but it sounds like the money hasn't been spent yet. (Announced in 2022, but apparently there was a "grant backlog.")
It sounds like you're skeptical that it will do very much, but that's a prediction.
I'm skeptical that it's easier to do things now than in 1957. Yes, the country was poorer, but it was also emptier. A lot of the difficulty of building new infrastructure is that there is a lot of expensive property in the way and it needs to be routed around. There were also no environmental impact statements back then. A richer country also has higher wages.
Here's a graph on the cost of highway construction. It only goes back to 2003, but in that time, the cost has gone up 3x.
blog post quantifying the death rate and comparing it to cars, also scaling it in how much harms are to dollar amounts https://www.jefftk.com/p/brightline-is-actually-pretty-dangerous tl;dr: 20x...
https://archive.is/WV0ud
From the article:
...
...
...
...
...
Comment box
It's nice for popular media coverage to correctly place the blame on the lack of investment in infrastructure. As the article notes, this is not unique to Brightline at all. They have more grade crossings than some comparable routes... the problem isn't unusual though.
Yes it would. Being "a private company" does not make visionary projects impossible. Just not feasible for Brightline, whose lack of grade separation forces lower maximum speeds which forces longer travel times which lowers demand which lowers ticket revenue which lowers income. Their "B" bond rating is so poor because they're locked into a difficult right of way. That hurts their credit, which makes it harder to get loans to pair with already insufficient grants to improve the ROW. And so on. Same old same old! Chicken and egg!
Capital projects almost always happen with government funding anyway. Brightline West is being built because of a multi-billion dollar federal grant. Brightline Florida's grade separation would be no different. If we want to get real.....freight railroads often grade-separate urban and suburban tracks to improve throughput/reliability and reduce crashes. Flyovers are common on high-capacity lines. Freight isn't likely to grade separate 250 miles of track at a time, but they've collectively done well more than that.
We grade separated every interstate highway. Almost 50k miles of grade separation. We can do the same for the remainder of our passenger rail network. It's not too hard and it's not too expensive.
There are technical challenges with grade separation in tight physical areas. It is also expensive. Could be $4m or $40m for a typical intersection, $292m+ for a uniquely nasty complicated one (rare). $40m is a lot. Can't deny it. But it's not... too much. $12 billion for the whole route? Ok, fairly big project at a DOT, but not that big for a 250-mile corridor. Not unprecedented at all. My state DOT is spending half that much to rebuild like 3 highway exits on I-95 for marginal improvements to congestion. If we're seriously worried about cost, one of these is clearly a better choice of funds than the other, but we're choosing not to do it.
Emphasis on this coded speech about NIMBYism. It's difficult to make safety improvements because without fail, small groups of local residents find ways to oppose literally all infrastructure projects, no matter how beneficial, including those which would save lives. They're not centralized, it's just a theme. Sometimes this is justified! For public safety projects, usually the outcry is an overreaction. In either case, this country defers so heavily to local preferences in situations where they shouldn't (while selectively ignoring local opposition anytime they want to build another highway), like common sense grade separation, that nothing gets done. Whether or not a particular opposition position is narrowly justified is unimportant if it consistently stops anything from getting better. Sometimes we just have to accept that things change and no solution is perfect.
Local groups will always oppose closing a grade crossing because they want to maintain the convenience of crossing there, even if someone dies occasionally. Individualistically, I get it. No one likes a detour. Tragedy of the commons. But they will also rarely themselves voluntarily help pay for a grade separation because no one wants to pay higher taxes, especially those who can afford it. The only time widespread grade separation gets done consistently is when it has strong state and/or federal DOT support. It can mostly only be done with a dedicated body of technocrats pushing for it in a spreadsheet in the capital. Or by an extremely motivated set of public activists and a genuinely thoughtful sponsoring legislator. Not so common.
The current anti-intellectual administration has notably canceled grade separation projects in California as part of a vitriolic political revenge game against its perceived opponents. Administration has also indefinitely halted many Bipartisan Infrastructure Law grants, including numerous grade separations, for "reconsideration" merely because that law was championed by Joe Biden. Completely pointless.
The technical problem has been solved centuries ago. We've understood the concept of grade separation since the first time we ever built a bridge over a canal. The railroads have been doing it since the 1800s. The money exists in our economy and in our government budgets to do this. We collectively choose not to do it for a confluence of unrelated local and political reasons.
Ive said it before on tildes but I drive freight trains for a living. Before getting very far in the article I knew where this was going. Large population centers plus interactions with trains equals people getting hit.
If a crossing gets closed without doing a grade seperation then often you'll have pedestrians still crossing there. You've reduced the risk... for cars.
Even some grade seperations don't eliminate risk. I was stoped once at a location with a major road underpass. Litterally as I was about to pull, a passing train warned me that a man with a bicycle was crossing over my train. People get killed doing this...
Another thing that I've seen that's less than ideal are pedestrian overpasses/bridges. People aren't going to climb that when they can just walk over the track so you're back to the same problem.
Elevating the rail is ideal for safety but you can't really just pop the tracks up and down, you need to have a long approach and descent, even longer for freight trains. As a result this makes the most sense in denser areas where you're eliminating multiple crossings in one shot.
It really is mainly an issue of money like you suggest.
I'm just glad that my terminal doesn't have too many incidents and the route I go REALLY doesnt. The only thing we regularly hit are animals. I don't like killing animals but it beats killing people. :-/
In college, our (very large) campus was bisected by an at-grade train line. It was very inconvenient if you got stuck on one side with a class starting on the other, though most faculty were understanding. Early on, the train would regularly stop on the tracks blocking the intersection. I saw a guy attempt to carry his bike and climbed between the cars. The train started suddenly, he dropped his bike on the tracks, and the train cut it in half. Very lucky that was all that happened.
There was a pedestrian overpass, but so far from the main traffic point that I never used it or saw anyone use it during my whole time as a student there.
Eventually they made some change to the train line so that the trains no longer stopped on the tracks on campus.
As I was graduating, they were taking on a major capital project to make pedestrian underpasses and some provision for the bus lines. Though surprise, surprise, the main goal of the project seemed to be making it easy for crowds from the west famous commuter lots to get to the football stadium.
IIRC in the US there is no rule or law preventing trains from just blocking a crossing all day, which us crazy. I may be wrong, I don't know US operating rules.
In Canada we can only be stopped on public crossings for 5 minutes after a car shows up. The loophole if you block a crossing accidentally is to pull at 1 mph for as long as possible until you can go again.
Illinois has such a law
ILCS
It's a fine, not prosecutable, but it really changed how trains were blocking the road constantly in my hometown when this was passed. Officers could just stand there and ticket every ten minutes or however it escalated.
So you'd have to check state by state and even city by city to be sure. It may be the case that this law isn't enforceable anymore, due to federal rulings too, I haven't truly kept up on it and that's always a risk.
In this case, the town is literally named College Station because the university is basically the town. And they have a strong alumni network all over the state. So they probably have more influence than normal.
I assumed they restructured the blocks (is that the right thing?) so that there was enough room outside campus for the whole train.
Ya, thats the right thing, and that makes sense. Building the system out in a way that you have enough heads up to stop in a good spot is definitely a strategy.
BTW, blocks are kind of like the zones where the system can detect a train (I don't know how to better describe that). You can have multiple blocks between signals or you can have one signal per block (right at the spots where the blocks change). So you usually have to change the blocks to move a light.
Comment box
Thanks for your perspective. Great point about pedestrian activity.
Have you anecdotally noticed any wildlife crossing pattern changes in areas where we’ve built grade-separated wildlife crossings?
I’ve read some studies suggesting that they reduce crashes, but I’d love to hear from an actual train operator.
We don't have any of those where I am so I'm not sure. Familiarity seems to be a big factor though. We had a new elk herd move in and they seem to get hit way more than the other elk herd 100 miles away.
Does fencing the railroad off help in that case?
I think that presents issues for wildlife movement.
Strangely we have an area fenced off to prevent moose from getting hit. It's not the direction I normally travel though and it was in place before I hired on so I don't know about a before and after.
I responded to @scroll_lock about wildlife crossings but I forgot to mention the wildlife fences. They do actually work and the fenced area ends on each side with a small bridge. The bridges act like a cattle guard and is a safe spot to cross (under) the tracks. Once in a while a moose gets in there and it usually ends in them getting hit.
Fences stop average people from wandering in but if it's cutting off an existing route people travel then they end up geting cut.
When reading about the first death in the article, I wondered about upgrading informal crossings to signaled pedestrian crossings, and adding fencing nearby so that’s the obvious way across. With a sign warning that it’s a high-speed train crossing.
There would still be a potential for mishaps because people will still do dumb things, but it seems like it should help? Cutting deaths in half would be a good start.
I've seen pedestrian crossings built on walking trails just without warning lights. Trains have to whistle approaching them just like a regular crossing.
It would probably work by kind of encouraging people to CROSS instead of lingering. Also giving people a nice path instead of hiking up the grade and hopping over rails. Thats all without warning lights.
Throughout, when concrete points against Brightline are made in this article, they seem much more dubious when examined. The added photos, likewise, seem designed to push a 'scary' narrative rather than honestly support the points. The overall conclusion – that level crossings and higher-speed trains are a bad combination – is reasonable, but the narrative sometimes ends up reading more as a hit piece against Brightline specifically. In some sense, it feels like exactly the sort of hypocritical NIMBYism that so often appears in the US: alternatives to cars should be lauded, except when there is any attempt to actually build one.
Consider the comparisons. The article argues that Brightline has particularly horrendous statistics for Florida. It points out that Brightline has almost seven times the number of fatalities as Amtrak, which "operates through many of the same urban areas as well as some additional ones", and merely "serves fewer passengers." Yet that number seems much less surprising when considering that it appears Amtrak's service in Florida consists of one daily train and one thrice-weekly train, while Brightline appears to run ten round trips per day now, and may have run more previously (Wikipedia lists 24 daily round trips in 2023). When comparing to SunRail, "another commuter train that operates around Orlando", the article uses the rate-per-million-miles that would have made the Amtrak comparison much less compelling to the narrative, but doesn't mention that SunRail, according to Wikipedia, has an average speed of 30 mph.
Or consider the argument that the track cuts "through the landscape at strange angles and in unexpected places". The accompanying photographs enhance this by using strange camera angles and crops, rather than showing the places the article describes. If they were to show a satellite image of one of those "unexpected" places they give as an example, the little league field in Delray Beach, you might be confused: it is a straight segment of track, parallel to a road, and appears to be fenced on the little league side. In what way is this "unexpected". Was it an example of a "strange angle"? All the nearby level crossings, while I might argue they should not be so near other intersections, are almost perpendicular.
Or the examples? A pedestrian walking on a spur path over an active train track, argued to be the 'logical' choice? A driver who pulled around a stopped car and closed gate, was high, and was directly told not to by a passenger, but is argued to be an example of a dangerous environment regardless of all those details? A couple who turned off their car and sat on an active train track with an active signal? One might suggest that an encouragement to not care about crossing tracks, whether on foot or by car, comes from an encouragement of car culture: "they're probably old tracks that aren't used much" is a common suggestion I've heard when seeing tracks.
Yes, level crossings are dangerous, especially at higher speeds. Yes, more fencing would be an improvement. But looking at the route, where the most dangerous segments seem to be along a previously-existing route, it seems clear that no amount of expenditure alone would allow for a major safety improvement, as any attempts would face an enormous NIMBY backlash. There's a certain argument that the encouragement of dense, especially residential development right next to historical rail routes was perhaps linked to a cynical attempt to make any future use of the right of ways difficult and unsafe, ensuring cars remained the only viable transport mechanism in the area. Adding fencing in many places seems like it would be an inconvenience to nearby houses. Elevating the rail in many places seems like it would significantly affect nearby houses, and potentially need land purchases, voluntary or otherwise. None of this is going to be acceptable to local property owners.
What is needed here are regulatory and cultural changes to make it so that safety improvements can actually be made; unfortunately, this is something where spending money on lobbying and marketing is probably better than spending it on trying to build infrastructure projects that will end up blocked.
Those seem like good points about the statistics. I was with you until the end, where you seem to be arguing that nothing worthwhile can be done. This seems defeatist? What makes you so sure?
I think some editing to the comment ended up not being saved: the point was that regulations and culture need to change in order to allow safety to improve without being blocked by NIMBYism and the cultural primacy of cars, and that spending money to that end is probably the better option than wasting it on planning the necessary safety improvements, fighting to put them in, and then losing.
I wonder if there is some relatively low-cost way to add a signal to an informal pedestrian crossing, such as the first example in the article, and add fencing next to it. Sure, grade separation is best, but if the money isn’t there, something could still be done.
Comment box
For sure physically possible. I live near a railroad abutting a riverside pedestrian trail. There are multiple at-grade pedestrian/bike crossings on this ACTIVE freight line in an urban area. The railroad is entirely fenced except for these ~3 crossing points. The intersections have crosswalk lights which are supposed to display a “Stop” walk signal when a train is approaching. In theory the intersections could be further upgraded to raise automatic barriers to physically block access (like for a rail/road crossing, but maybe more effective with sufficient pedestrian fencing). The walk signals could also be better placed for visibility, and an auditory signal could be added—this all might help, won’t hurt, and it’s not too expensive. It’s mostly a moot point because trains only occasionally run here during daytime.
The crossings are important for the trail and eliminating them outright would incentivize dangerous behavior as @mild_takes mentions. People would use a pedestrian bridge if forced, but there are space constraints and an ADA-compliant ramp may not fit; an elevator would be expensive and liable to break. Signalization and utility work is always complex, and the city would need to work with the railroad to do anything. It’s annoying to do - and it’s no one’s priority - so the city has just left it as-is. Imperfect without trenching the railroad. Statistically unlikely to lead to any fatalities (just not high enough traffic/speeds), but it could happen. If someone dies, the local activists will raise hell and maybe get a ped bridge.
No solution is perfect. These stop-gap solutions like signals are vulnerable to human error. People are irrational: they ignore direct warnings, they climb over fences, and they make mistakes (trip and fall, misjudge distance, etc…..). Sight/hearing impaired people might not comprehend the signals. Drunk people do drunk stuff. Teenagers intentionally evade fencing to go surfing. Movable barriers electronically or mechanically fail. People even walk long distances on grade-separated or fencing tracks from unfenced areas (for whatever reason) and then get stuck.
Grade separation is still the best solution though. The money exists in our tax base and economy. It’s a societal, philosophical decision—do we value Vision Zero more than equally expensive highway widening projects?
I’m not sure where you’re going with this “the money exists” stuff. Sure, there is money out there somewhere, in theory, and the Fed could create more money. But real projects operate under budget constraints.
Efficiency matters, particularly for expensive infrastructure projects. For the cost of one grade separation, what could you do to improve safety along the entire rest of the line?
From a broad historical perspective, today’s prosperity is built on yesterday’s efficiency improvements.
Comment box
Projects have budget constraints. We choose what those projects are, and their scope. It's not "in theory" - it's "in reality" and it doesn't involve wrecking the economy. (Other countries do this better than the USA, it's not trailblazing) Similar idea as why we fail to feed everyone in the country despite producing more than enough food to do so: an issue of distribution, not a lack of abundance. But this might be simpler to fix... no need to upend capitalism, just build some bridges.
State and federal DOTs in the USA are allocated enough money with current tax structures to complete transformational, visionary capital transportation projects, especially for passenger rail. MPOs and DOTs choose, repeatedly, to allocate an oversized portion of that funding to inefficient roadway projects that at best maintain the status quo and usually fail to deliver on promises to resolve congestion or holistically improve safety. They justify this car-centric spending allocation with a combination of institutional inertia, pseudoscientific/circular traffic modeling projections, and fear of political backlash. Even "progressive" MPOs like in Seattle and Philadelphia lock themselves into self-fulfilling prophecies about car centricity. Planners individually come up with lots of great ideas, but institutional knowledge in these places favors the dominant cultural milieu, which gives high roadway use the benefit of the doubt by default. (Source: I talk to these kinds of people a lot, also I just spent a lot of time reading my MPO's new long-range plan. It's okay, but it still makes assumptions that incentivize overindexing on car infrastructure. The DOT will ignore most of it anyway to build wider highways.)
Roadway projects induce more roadway use. Roads are some of most self-destructive and least inefficient forms of transportation we have ever devised. No other form of ground transportation deteriorates at the rate of a highway, an inefficiency inherent to the mode. Most vehicles have single occupants despite weighing a ton or more. For the number of people a highway transports, its vehicles exert a lot of force onto its asphalt. (The "weight of people transported" to "weight of vehicle" ratio is bad.) That loosens asphalt particles, which enables water filtration, which creates cracks and potholes that freezing weather exacerbates. Paving roads with more durable materials raises costs and has various engineering problems. Every attempt to make highway vehicles more efficient leads to 1) a bus or 2) a train.
Major resurfacings have billion-dollar price tags that hardly anyone questions. Multi-billion dollar widening and "traffic optimization" projects repeatedly fail to resolve urban and suburban congestion. Complex car-centric signalization projects make marginal improvements to design capacity while worsening pedestrian safety and disincentivizing walking. More and larger roadways have higher maintenance costs, which creates the appearance of a "tight budget." Roadway maintenance projects are generally prioritized ("we can't maintain our existing roads, so we definitely can't build your train line") even though a mode shift toward rail would be more cost-efficient (and safer) in the long term.
All choices! We continue to overbuild roads because it's a cultural preference, not because it's efficient, effective, or good for society.
The mileage of urban/suburban passenger rail services in the USA is quite low. We grade separated 50,000 miles of interstate highway crossings (it's part of the design standard from... 1957!!!). The Northeast Corridor is like 457 miles. Brightline Florida is around 250 miles. Even every single Amtrak route including freight lines is ~21,000 miles (mostly rural with few crossings and fewer fatalities per crossing than urban/suburban). We have done much harder, much more expensive things in the past. We can do this if we prioritize it!
Not much. We have done most of the easy stuff, like signaling, train speed limits, and basic gates. Trespassing + grade crossing collisions are collectively the source of ~95% of passenger rail fatalities in the USA: look at
(Trespasser Fatalities + Highway-Rail Incident Fatalities) / Total Fatalities.Grade separations serve the dual purpose of improving operational efficiency and safety:
Sometimes it might be possible to avoid having to grade-separate tricky intersections by just building a new section of right of way. Unfortunately, American planners are infamously poor at efficiently spending money on transit/rail projects. They make inefficient designs using inefficient methodologies, and have relatively little interest in emulating global best practices. Sometimes this is the right solution, I just wouldn't count on it.
But grade crossings are well-understood as a component of road infrastructure and don't suffer as much from that problem as a bloated train station concourse design does. American DOTs can sure build road overpasses!
The best use of funds on a timeline is probably to grade-separate all intersections near population centers first, in order of proximity to the largest population centers. It's more efficient to do the urban/suburban crossings first and the rural ones later, because rural crossings statistically have less traffic and therefore cause fewer fatalities. Some exceptions apply. But this could halve the price tag of grade-separating "everything," or more.
I'll take it as given that some budgets for maintaining existing transportation infrastructure are very large and that the money could plausibly be spent differently. But this seems like too high-level an argument? It would be more convincing to talk specifically about how the Florida Department of Transportation in particular manages its transportation budget and how it should be spent on different projects.
That would require doing homework, though. I don't want to give anyone a project; I'm just saying maybe we should be less confident that we know the answers based on general principles.
You say "we have done most of the easy stuff" and maybe that's usually true, but from the article, it didn't sound like it was true for this particular railroad line in south Florida? A convincing argument would have to be more specific to that particular line, explaining what it's really like and how the article got it wrong. (pallas made a good start on this.)
I don't see the relevance of some other things you're bringing up. Sure, there are almost 50,000 miles of grade-separated Interstate highways in the US. But that's money already spent. How do you get from there to budgeting for new projects? It seems like similar logic to saying that the New York City subway system exists, therefore we have lots of money.
Money already spent isn't a source of funds. Existing infrastructure is a starting point for future improvements. It can be an advantage or a disadvantage, depending on how it was built and how well or poorly it aligns with new needs.
Comment box
Well, you make good comments, I'm saying:
that this project isn't relatively over-expensive on a micro level compared to benefits from a safety or transportation efficiency perspective. It's expensive but it's in the same class as other big projects FDOT does regularly.
we can do way more TOD and VZ infrastructure on a macro level from a financial perspective. I see why you think this is irrelevant. I am saying this to emphasize that we designed our whole world, and we can re-design it. It's fully possible. The sky is the limit. It's not too hard or too difficult in general. (Building the interstates was harder & more expensive at a time when we were poorer. so this shouldn't FEEL so hard.) I just think it's important to be optimistic. Our culture is often the opposite.
i also talk too much.
I can say with full confidence that the federal DOT and every state DOT allocate excessive funding to highway projects at the expense of passenger rail. In absolute dollars it's something like a 7:1 ratio nationally, and probably more like 14:1 in Florida (from memory, don't cite that). Dollars don't translate to impact, but this tells you at a high level what their priorities are.
A few DOTs do better, like California (because of CAHSR, SF, LA) and New York (because of NYC), but most of the land area in those states continues to receive car-centric capital investment. (Although New York has been removing a few highways, like in Rochester.) You don't actually need to look at the finances to see this: just look at ridership splits by mode: if Florida allocated funds to efficiently move people, it would already have at least one high-speed rail line connecting all its major metros, but it doesn't. (FDOT is only interested in efficiently moving vehicles, which is financially inefficient.) Florida rail ridership is accordingly low.
Some MPOs and RPOs (lol) are more progressive about planning than others, but it hasn't filtered through to the DOTs yet. They're bastions of hardcore traffic pseudoscience. They're extremely insular and are not interested in "activism." MPOs are also scared of that word, but they at least try. (Think of traffic engineers as people who like to solve math problems. In their spare time, they optimize Factorio production lines. They have a different education than urban planners and their priorities are typically these math problems, not societal change per se. Some exceptions, not a lot.)
I have not worked with Florida MPOs personally but I have not heard of them trying visionary things like Denver or Pudget Sound MPOs.
Eliminating and narrowing highways would free up DOT budgets to focus on more efficient transportation infrastructure. So would choosing not to build more highways. Highways have an opportunity cost.
We do have lots of money. Our economy has grown since 1957 and so has our tax base. Quick maths: federal taxes alone collected in 1957 were about $83 billion and about $5.2 trillion in nominal dollars. Per-capita, that's about $487/person in 1957 and $15k/person in 2025. Adjusted for inflation, that's about $5000/person in 1957, so we have 3x the federal tax revenue as then. At the most theoretical level, we managed a staggeringly large, inefficient infrastructure project with 1/3 of the federal budget as present. (Note interstates were built over years.) Obviously budget priorities today include some systems that did not exist in 1957, but it's not like we have less money in absolute terms. And the choice to raise or lower taxes to cover any delta is still a societal decision that we have chosen to neglect.
Big-picture, these are projects on the time frame of a 25-year long-range plan. In the short term, of course Brightline has budget constraints with whatever grant they can obtain for a grade separation. But for how big these population centers are, it's still just not that expensive.
My earlier figure of $12b for the whole route is probably an underestimate. Let's say $20b from Miami to Orlando. FDOT capital budget in 2025 was $15 billion, ~5% of which was for railroads. Practically, it would take at least 5 years to grade-separate Brightline's whole route. If the state government paid for 100% of it, then over 5 years that would be $4b/yr. If they did top-priority separations in the first 5 years and low-priority separations in the following 5 years, then that's $2b/yr for 10 years.
Of course the state DOT will not pay for 100% of the cost. Brightline could finance maybe 1/3 of it. The federal government could chip in a substantial amount, maybe another third. So that means over 5 years the cost to the DOT is closer to $1.3b/yr or $650m for 10 years.
That is a big project. It's just not absurdly big. For context, the NYC Gateway Program is about $16b. Big project for sure! But only a portion of the budget. Meanwhile, FDOT are spending $500 million to add new lanes to I-4, $340 million to widen a small section of I-275, etc., adding up to a supermajority of that $15b budget. (There are more examples of big boondoggles, and the small ones are collectively expensive too.) If FDOT just decided to incentivize a modal shift to rail, it would cost the same amount or less upfront, and have lower maintenance costs. Transit-Oriented Development pre-empts wasteful highway spending. Alternatively, this could be easily paid for with new revenue using a small "cap and trade" tax (like California), small sales tax, small gas tax, or even (self-defeatingly) a small train ticket tax. The kind of person who comes to Tildes will complain about regressive taxes, so OK, a small income tax or property tax or corporate tax or whatever needs to happen. Fine with me. Florida citizens choose their tax system. Any politician could push for a ballot measure or legislature vote to make this happen. If they don't, it's because they choose not to: they ignore the safety benefits, they ignore the maintenance efficiency benefits, they ignore the broader economic development benefits. (They know what Brightline is, it's not obscure.)
They have done most of this, or allocated funding for it:
195 =/= 250, but they have likely focused on urban and suburban segments, or this complements existing protections.
The problem is that these solutions just don't help much. People get onto tracks at grade crossings because they use roads to get to grade crossings. They're legally allowed to cross there. Then they illegally stay in the railroad ROW, and then they get hit by a train.
That announcement about the grants is a good find, but it sounds like the money hasn't been spent yet. (Announced in 2022, but apparently there was a "grant backlog.")
It sounds like you're skeptical that it will do very much, but that's a prediction.
I'm skeptical that it's easier to do things now than in 1957. Yes, the country was poorer, but it was also emptier. A lot of the difficulty of building new infrastructure is that there is a lot of expensive property in the way and it needs to be routed around. There were also no environmental impact statements back then. A richer country also has higher wages.
Here's a graph on the cost of highway construction. It only goes back to 2003, but in that time, the cost has gone up 3x.
blog post quantifying the death rate and comparing it to cars, also scaling it in how much harms are to dollar amounts
https://www.jefftk.com/p/brightline-is-actually-pretty-dangerous
tl;dr: 20x more deadly per passenger-mile, $2.5B per year in deaths at DoT value of human life