Note: This is the second in a series. Earlier installments are listed at the end.
In considering whether the 198 corridor could be redesigned using Olmstedian principles, the easternmost end would seem to be the easiest. After all, the first half mile or so, from the 33 split to Parkside Avenue, was once Humboldt Parkway. So for the first half mile, simply eliminate the expressway, and return the parkway. All done planning. Good riddance, Robert Moses! Welcome back, Olmsted!
If only it were that simple. Some have called for the same approach for the section of the 33 that also used to be part of Humboldt Parkway. But the community there, as represented by the Restore Our Community Coalition, knows that notion is a non-starter. Their proposal is not to eliminate the expressway, but deck it over.
And their section of the former parkway, if anything, is easier to address than than the 198 section. There is no confusing intersection, like at Main Street, and no roundabout-turned-at-grade-intersection like at Parkside Avenue. No, their plumbing problem is more of a straight pipe replacement: swap out one piece of horrible traffic pipe and replace it with a neighborhood-friendly one of equal capacity. But the Humboldt Parkway section of the 198 is complicated by two old, unreliable, unpredictable, leaky traffic faucets that frequently clog and are challenging for even regulars to use. All while serving as a critical distribution bottleneck for a relentless traffic volume from the 33.
A pressure point in Buffalo’s traffic circulation
For better or for worse, the Kensington is one of the region’s busiest and most vital highways, and that won’t be changing anytime soon. Especially significant is that, in the area of the 198, the 33 makes its closest approach to Main Street anywhere outside of downtown. For the upper West Side of Buffalo, and all of North Buffalo, this is the point where you connect to the 33. If you are connecting to the east of the city, or to the airport, or to the Thruway, you generally use the 33, and you access it there. Of course, that illustrates a major problem of urban expressways: they contain and confine traffic, concentrating it and funneling it to single points. At those points, then, there is an incredible, high-pressure burden on the local street grid. In cities, once-lovely streets and intersections become expressway on and off ramps. Traffic infrastructure designed to deliver the equivalent of glasses of water functions badly or breaks down entirely when a fire hose is plugged into it.
Main Street at the 198 is a textbook example of this. Using another analogy, the 198 becomes the hand that extends the 33’s reach. Connecting streets along the 198 corridor like Parkside, Delaware, Elmwood, and Grant become the hand’s fingers reaching into the West Side, North Buffalo, and even Kenmore street grids. But before it can get there, all this traffic must pass, like sand in an hourglass, by Main Street. And except for what is siphoned off there, the Parkside Avenue intersection sees most of it, too. Based on 2015 traffic data, which may predate the lower speed limit on the 198, 70,000 cars per day (Average Daily Traffic) travel the expressway between the 33 and Main Street. Between Main and Parkside, it’s about 15,000 less, and west of Parkside, it’s 15,000 less again.
Clearly, the former Humboldt Parkway section of the 198 is a huge pressure point in Buffalo’s traffic circulatory system, and whatever we do with it should guard against causing any life-threatening clots. But it’s already prone to them, largely because the intersections there are complex, confusing, and often difficult to navigate. Traffic is prone to backups not only on the 198, but also into the neighborhoods. The node at Main Street is made especially challenging and confusing by the way Kensington Avenue plugs in. As Buffalo Rising wrote earlier this year,
This long discussed intersection has been the source of angst for years. The intersection is a logjam of confusing streets and lights, creating an unsafe environment for cyclists and pedestrians, not to mention drivers. Often times, there is no rhyme or reason when it comes to cars trying to race through segments of the intersection in order to catch a yellow light. That, or vehicles dart from one direction to another in hopes to not get caught in traffic holdups.
Also, DOT reports identify what should be no surprise to anyone: the Parkside intersection is also a major cause of delays, backups, and accidents where the 198 meets the city street at-grade with a stop light.
And as crazy as those nodes are for motorists, they can be a nightmare for cyclists and pedestrians.
In rethinking the 198, it is essential to rationalize these intersections and improve traffic flow there. Is it possible to do that, while also improving the landscape, and the urbanism of the surroundings?
I think so.
This Place Matters
But first, so as not to be like those single-minded engineers who designed the 198, let’s look at the urban context. This is also in keeping with the calls by the Scajaquada Corridor Coalition‘s Design Chair, Lou Haremski, for more urban planning context (in the radio interview transcripted here) in any reconfiguration of the 198.
The muddles at Main Street and Parkside Avenue are not just a problem for people trying to get from point A to point B. At the 198, Main Street has almost entirely lost its sense of place, despite being a significant node in the city for many reasons. Being there feels like stumbling across a pocket of suburbia in the city, complete with surface parking lots in two directions, and crossing overpasses over expressways. Like much suburbia, it is in an environment where drivers are focused on getting where they are going, and through a confusing set of turns, lanes, lights, and ramps more than they are looking out for pedestrians – let alone appreciating their surroundings.
At Main Street, to the north of the 198 are two critical institutions founded by Buffalo’s Catholic community: Sister’s Hospital and St. Mary’s School for the Deaf. To the south are two more critical institutions founded by Buffalo’s Catholic community: Canisius College and Medaille College. Yet the 198, like a Mason-Dixon line, seems to keep them entirely separated, north from south.
Kitty-corner from each other at this node are two of Buffalo’s greatest and justifiably proudest neighborhoods: Parkside and Hamlin Park. Yet rather than tie them together, this node creates a perpetual no-man’s land between them, like an unresolved overseas war. Sisters Hospital and Lyndon B. Johnson tower similarly stare at each other, disconnected, across the automotive wasteland.
Even MetroRail is affected. To best serve the colleges, the stations were put south of the morass, on either side of Main Street with a passage under Main Street. But because of the morass, the hospital and school just to the north aren’t nearly as well served, and therefore don’t fully benefit from being on the only subway line in Upstate New York.
The same is also the case for the Parkside neighborhood and Delaware Park itself. Parkside’s other MetroRail access point, Amherst Street station, at the northernmost edge of the neighborhood, is much easier to access than the Humboldt Hospital station at the southernmost edge of the neighborhood. That station is also the closest to Delaware Park, but who would know that? MetroRail says nothing about the nearby park, there is no visual link between the two places. An in keeping with the modernist philosophy of ignoring context, the design of the station has nothing remotely parklike about it. It looks more like a modernist suburban bank branch.
So in rethinking the 198 in this area, we need to add “recreating a sense of place” and “knitting back together the torn urban fabric” to our to-do list. Can even a genius like Olmsted help us with all that?
As it turns out, Olmsted left us with all the ingredients we need.
Olmsted introduced the circle into Buffalo’s landscape, which transformed intersections throughout the city into beloved places. And I described in my last piece, he and Vaux introduced the traffic underpass in their earliest parks. But for this section of the 198, we don’t even need to take a page from history, because the pieces of the solution are already there. Although the surface streets and intersections are problematic at Main Street, the underpass there at least helps keep through traffic moving along. Agassiz Circle could borrow the underpass idea to help alleviate its traffic bottleneck. And at the same time, Main Street could borrow the circle from Agassiz, to help rationalize surface traffic and recreate a sense of place lost over six decades ago. To create a sense of place, a circle must have a name, and who better to honor than the sisters of Sisters Hospital?
How would this work? The surface streets that, essentially, serve as on- and off-ramps from and to Main Street would, instead, plug into the traffic circle or roundabout. Because the circle is the best type of intersection for traffic flow, it would help keep traffic moving on the city streets above, and from creating backups onto the 198, the 33, and in the neighborhood. All the while through traffic on the 198 corridor would continue to pass under, unimpeded, as it does today.
A key here is that traffic circles, in addition to providing a parklike node, and creating a sense of place, also help to rationalizing intersections. Circles don’t care if three roads plug in, or four, or five – as long as there is room. As we know from Buffalo’s circles, they also nicely accommodate roads that plug in at odd angles – as Kensington Avenue does at Main. In this scenario, Kensington Avenue would no longer connect directly to Main Street, but would originate at the circle. The traffic counts on that end stub of Kensington show it to be, by far, the least-used chunk of Kensington Avenue from its origin in Buffalo to its end in Amherst. Traffic circles and roundabouts are also the best intersection design for traffic flow, a key issue here.
The circle-and-underpass combo also has built-in flexibility. The sunken portion of roadway could be extended east, and decked over, to visually restore an additional section of Humboldt Parkway. If traffic volumes drop dramatically, perhaps due to changes in transportation technology, the underpass could be narrowed or eventually even eliminated, with all traffic returned to surface level. Or, a lane in each direction of the underpass could be made transit-only, for a bus-rapid-transit system that would closely link Buffalo State and Medaille Colleges and all the culturals along the Scajaquada Corridor, and Delaware Park, with MetroRail.
This combination isn’t something new or unique. For one, traffic underpasses exist in many cities here and abroad. Many of these were created early in the automotive era, in the 1930s, 40s, and 50s, as a way of relieving traffic congestion at key intersections, or restoring sanity to key streets or districts by routing through-traffic underneath. One of the most well-known is the Battery Park Underpass in New York City. One of the most infamous is the Alma Tunnel in Paris, where Princess Diana was killed.
Similarly, roundabouts or traffic circles atop underpasses are also nothing new. For a half-century, the city of Bristol, England has had one, in the Old Market district. The Temple Way passes underneath. There is a newer one in a suburban district of Beirut, Lebanon, and one under construction in New Zealand. Twin roundabouts with underpasses are planned for a science park in Holland. And, although expressway-centric, the FHWA also recognizes the roundabout-with-underpass combination as an “alternative interchange configuration,” citing an example from a suburb of Albany.
Keeping in mind that this is one conceptual idea, and no one is gassing up the bulldozers, this proposal would involve needing to take down or relocate a few houses. Many of the houses in that area are owned by one of the two colleges, who would presumably be willing to part with them for just compensation. The few houses that are privately owned are on one quadrant of the circle, and it might be possible to reap benefits from three-quarters of a circle while waiting for those houses to become available.
The circle would also involve taking down a single building at Sisters Hospital. In terms of design, it is the least appealing building on their campus. In fact, it presents a bunker-like appearance, which is somewhat understandable for a laboratory building, yet it detracts from a sense of place and community. With compensation would come the opportunity for a nicer building, with underground parking, directly fronting the circle. It would create a sense of engagement with the community, rather than a sense of being in a bunker or behind a fence and parking-lot moat. It could also be plugged directly into MetroRail, creating a coatless connection that would boost intermodal transportation options for those visiting or working at the hospital, reducing the need for parking there.
So as not to be environmentally wasteful, the existing building could perhaps even be moved a short distance, and serve as the nucleus around which a more friendly building could be built.
And speaking of the environment, this proposal would also create the opportunity to put linked MetroRail access at all four quadrants, to serve all the key neighborhoods and institutions around the circle. This is something that should be done at more MetroRail nodes, which typically have access at only one quadrant of their intersection. This is among issues discussed in the current Transit-Oriented Development study currently underway. Note that to bring MetroRail access to all four quadrants of the circle would require pedestrian passage under the sunken roadway. That may sound extreme, but MetroRail at that location already passes under the sunken roadway. Whether or not that might facilitate adding the passageway, it makes clear that it’s not impossible.
From the air, Agassiz appears to retain much of its form as a circle. But on the ground is another matter. A heavily traveled urban expressway comes up to the surface there, creating a trial-by-hazard for any pedestrian or cyclist brave enough to try to cross it. Based on accident statistics, it also creates plenty of chaos for motor vehicle traffic, as well. Turning this Olmsted-designed circle into an intersection with a high-speed expressway in turn turned Parkside Avenue into an expressway on-ramp, creating stomach-turning effects on one of Buffalo’s great residential neighborhoods.
Over the last fifteen years of study of the 198, repeated pleas by the community to restore Agassiz Circle have fallen on deaf ears with the DOT, citing traffic volumes. And indeed, traffic volumes are very high at the intersection, due to factors mentioned above. Parkside Avenue serves as a key link between Buffalo’s second-most-vibrant commercial district, Hertel Avenue, and the expressway network.
An approach like Sisters Circle, with through traffic passing unimpeded underneath, could remove enough of the surface traffic volume, and calm the rest, to make recreating the circle a viable option.
The DOT should be open to the underpass option, not just because it would improve traffic flow, but also because it borrows an existing solution from just down the road – and engineers tend to like already proven solutions. Previous studies of the corridor have identified the Parkside at-grade intersection as one of the most significant impediments to level of service on the 198. So let’s eliminate it.
In digging out the underpass, the cut would have to include ramps for traffic going between the sunken roadway and the circle.
No one would see this change up close as much as the Burbank neighborhood, which fronts the southwestern quadrant of Agassiz Circle. While having traffic resume using the circle again would bring some traffic closer to their homes, it could be designed in such as way as to soften the impact, by keeping their address street, and the entrance to Medaille, separate from the traffic in the circle. An old transportation diagram, pictured here, shows how that could work.
But most of all, Burbank and Medaille would gain tremendously in quality of life by being able, most likely for the first time in the memory of most, to take a stroll into the park without thinking about whether their life insurance is paid up. With the 198 sunken below the circle, and perhaps for some distance west, as well, there could even be an arched pedestrian bridge west of the circle for them to directly access the park, avoiding the circle altogether.
Humboldt Parkway redux
To create these sister circles, the entire 198 between them would be sunken. Most of it already is. That would create an ideal opportunity to reconnect the streets on either side of Humboldt Parkway that have been divided by the 198. That could be done with either pedestrian bridges, or, even better, decking it like the ROCC plan for the 33 and restoring the parkway on the surface. This would create the best possible gateway into Delaware Park from Main Street – after all, Olmsted intended the parkways as park links – and also complement ROCC’s efforts to restore a section of the parkway closer to MLK Park. If these two sections of Humboldt Parkway near the parks on either end were restored, the push would then be on to restore it entirely.
They built this horror show in pieces, so we’ll have to fix it in pieces, too.
Note that if you deck it over, it’s still possible to access the sunken roadway below. This image from plans for restoring the Humboldt Parkway near the science museum show the kind of slots in the deck-over for ramps that would be used in such a case.
Is this Olmstedian?
Neither Olmsted nor his longtime partner, Vaux, were available to be interviewed for this article. We don’t know what solutions they might have suggested to this design problem. So we have to view this problem through the lens of their writing, their body of work, and, especially, what they originally created here.
This proposal would re-establish an Olmsted-designed circle as a functioning circle, although modified with an underpass. It would also create a new circle, at a node of the city that has changed drastically since Humboldt Parkway was originally laid out, in both urban form and traffic. The sister circles would work together to return a landscaped approach to the park, knit torn urban fabric back together, manage the traffic that we can’t simply wish away, and perhaps even help us ratchet it down over time.
The circles, like real sisters, would borrow from and share with each other.
Perhaps they could even be joined by a third sister one day, by inspiring a circle for nearby Main and Delavan, a place crying out for one. A circle there (perhaps “Griffin Circle,” named in honor of Canisius College), coupled with Sisters Circle, would bookend the “institutional” section of Main Street, occupied not with houses or storefronts, but by three institutions: Canisius and Medaille Colleges, and Forest Lawn Cemetery.
This then, is really the essence of landscape architecture, a profession essentially invented by Olmsted and Vaux. Sure it involves plants, and Olmsted knew plenty about those. But it’s much more about designing our environment. And that’s where the DOT went wrong in creating the 198: they took a beautifully designed environment, and turned it into a road design problem. This proposal does the opposite: it takes a road design problem, and uses it as the basis for recreating a beautiful environment. The two should not be mutually exclusive.
Next: Taking the Low Road.
Olmstedian Scajaquada series so far: