Brazil - Retrospective

It's time for another retrospective as final works enter the pipeline for Project 7 - Brazil. Technically the project isn't over, but all of the research is, so it's about time that I put aside some time to put my thoughts into words.

First, before I jump in, let's go over some basic facts about Brazilian public transit and Brazil in general.

1. Internationally, hearing Brazil and Transit makes one think about BRT. However, while modern BRT originated in Curitiba, Curitiba's system was actually converted from a prior bus priority system, and when it comes to overall quality, Colombia is actually a better role model for BRT. I'll go into more detail later in this post, but just keep in mind that while there are a lot of BRT systems in Brazil, not all are created equal.

2. Brazil lost most of its intercity railways. This is an unfortunate fact. What cities have done with the rights of way varies greatly, with many having been converted to either suburban metro systems, bus rapid transit, or VLT. VLT is one of those things that I'll discuss in greater depth later in this post because its definition is very broad and nearly every system in existence is poorly designed.

3. Brazilian public transit integration is notoriously poor, and like elsewhere in Latin America, it is often operated by private companies. The solution in Brazil, like elsewhere in Latin America, has been to build massive integration terminals where one can switch between buses in the same network but operated by different companies.

4. Brazilian Portuguese is pronounced very differently from that of Portugal, and the spelling reforms make street names and station names inconsistent.

With that, I'll explain the research phase for this project and some reasons why it was much more challenging than expecting.

I began my research on Brazil by looking at the top 150 cities by population and grouping them in a document by metropolitan area. Yes, 150 cities. The reason for this was Criciúma, a mid-sized city, having a surprisingly high quality BRT line, and Sobral having a commuter rail (VLT - again, will discuss this in greater depth later) line. It immediately became obvious that I'd have to split up my research by region. However, afterwards I needed to split things up further in order to not lose my sanity, as the Southeast contains both Rio and São Paulo.

Once I had identified cities, determining which cities had higher order transit began. Now, all of the cities with rail are documented on Urbanrail.net and/or Wikipedia. The Wikipedia Template did most of the work for me here. But then came the BRT systems. It's generally understood what constitutes a high quality system, but most systems in Brazil are not of this sort. And so begins our first detour.

I have five terms listed above. Perhaps surprisingly, nobody in Brazil seems to call Bus Rapid Transit "Ônibus de trânsito rápido", instead preferring the BRT and BRS labels despite them being... English (or using Corredor de Ônibus, which tends to indicate a lower quality). Yes, Runcorn predates Curitiba, but the term BRT was invented after those systems were created. As for what the other terms mean, BRT tends to represent actual BRT, with enclosed stations, off-board fare payment, and all of the amenities it brings. But then we get to BRS. This term is the equivalent of BRT-Lite and Arterial BRT on this website, though as a branding, most systems actually just use the generic "corredor de ônibus" instead of calling their systems BRS. Rio's BRS network is actually a bunch of curbside bus lanes with the Faixa Azul treatment. The Faixa Azul treatment is the relatively new 'hot' way to separate a bus lane from a general purpose lane (AKA the cheap and inexpensive way), whereby a blue line is painted to delineate the lanes. Blocks on the roadway and other physical separation is not cheap, and this is an inexpensive solution. You see this treatment in cities like Manaus, Rio, Sorocaba, etc. And here is where the 'inexpensive' treatment becomes more pressing.

First, let's discuss historic corridors. Bus corridors are not new in Brazil and have been around for a long time. BRTData.org credits the creation of Lima's first busway to 1972 (it has since been converted to full BRT). I'm not exactly sure if this was the inspiration for the Brazilian cities, but many cities, such as Curitiba, São Paulo, and Porto Alegre, began rolling out dedicated bus lanes throughout their cities in the 1970s and 1980s. These were just dedicated bus lanes with bus stops. They are what I consider to be 'BRT-Lite' in that they have the dedicated roadway space but not much else. Some were separated from general traffic, while others were not. Curitiba was the first city to elevate these to something new and impressive - Bus Rapid Transit. Off-board fare payment, fare integration with free transfers within the stations, and level boarding with high floor buses and deployable ramps immediately became a hit... except that most cities didn't get the memo about the complete package being important, and were prepared to cut out anything they considered to be unimportant or expensive.

The direct result of these differing approaches to bus corridors was that Brazilian bus rapid transit style service roughly split into three general categories. First, actual BRT. Expensive to build, hard to maintain, but with a level and quality of service that was truly good (though still dependent on continued investment and maintenance). There are a lot of these systems, and they continue to be built. Most use high floor vehicles, but some newer systems use low floor vehicles. It's worth mentioning that Brazil seems to not be very good at low floor BRT level boarding when compared to countries like the US (where it is unfortunately done without actual bus lanes - AKA the part of bus priority that's critically important), but that's a separate issue.

Next, high floor BRT with lots of cost cutting. Turns out that nearly all of these systems actually qualify as full BRT according to the standard due to level boarding and a general tendency to prevent turns across the busway or outright prevent intersections altogether. The first things to be cut were off-board fare payment, good ROW separation, and any kind of quality of life amenities (e.g. real time arrival information). Criciúma has one of the best setups in this category due to actually using physical separators for its ROW, while others such as those in Santo André and Manaus suffer from very poor maintenance (Santo André is considering full closure while Manaus let its system rot to the point where roofs were falling onto the station platforms prior to the 2014 world cup forcing them to fix things). Porto Alegre's maintenance is so poor that over the course of a few decades, the sun and vandalism destroyed station name signs on its stations - incredibly bad for wayfinding.

As for the final category, it's low floor bus corridors. These mostly resemble the original ones from before the BRT era, though raised platforms (not level boarding though) and physical separation are present in quite a few. These nearly universally lack level boarding and offboard fare payment, tend to not have station names, and tend to have stations made of very poor materials or no stations at all. Anápolis's system hasn't even opened yet (Correction: It opened in 2020) and it's already rusting. These corridors are numerous, continue to spring up, and are almost never BRT-grade. They also tend to be open busways. That being said, these corridors do offer a lot of value by allowing regular city buses to bypass traffic and so are important to build despite their flaws (though investing in stations that are built to last would definitely help!). It is worth mentioning that most new EMTU corridors in the São Paulo and Campinas metropolitan areas fall into this category, with the ROW quality of new lines being a downgrade from their original ABD BRT network.

As a final note, BRT and BRS can coexist. Recife and Campinas are good examples, local buses and BRT both use the main north-south corridor in Recife, with BRT serving as a high quality express. Campinas is a story for when I discuss the cities I researched in detail.

Now that we've covered the delight that is BRT in Brazil, let's discuss some of the casualties of the research phase before moving onto VLT, just so that I have everything covered here for transparency. Note that pretty much all research was conducted before actually beginning work on creating pages on the MRA. These are Maringá's Mega BRT Morangueira - Kakogawa and Cascavel's Av. Tancredo Neves, Av. Brasil, and Av. Barão do Rio Branco. No level boarding, high floor buses serving curb level stops, and low quality stops in general. I learned about Maringá from BRTData and was shocked that a city in the same state as Curitiba could have such a low quality brand new bus corridor. Cascavel on the other hand has nice stations without names, no level boarding, and... physically separated (as far as the ITDP scoring committee is concerned) lanes with bypasses at stations. If it weren't for the fact that it's using high floor buses with curb height stations and stations don't have names on them, I'd probably be considering retroactively granting it a page on the MRA (they ban turns across the busway, so BRT Basics would be 8/8/0/7/0 - above the minimum score of 20).

Finally, São Luís was an interesting case. They have high floor buses but I couldn't find anywhere with high floor platforms. A mystery that may come back to bite me later...

Now let's jump into the VLT mess before full city by city retrospectives.

"Light Vehicle on Rails" is a very vague term. It also happens to therefore mean literally anything. Just like 'Light Rail' is a loaded term in the US and Canada, VLT is a loaded term in Brazil. It roughly decomposes into three categories - Light Rail Vehicles, Diesel Light Rail, and Gadgetbahns.

First, Light Rail Vehicles. Brazil calls Trams Bondes, as in the last surviving one in Rio. VLT in the light rail case refers to both high floor and low floor light rail vehicles. Now, you may ask, high floor light rail vehicles in Brazil? Where? The answer: It's gone now, because the ROW used was incredibly inefficient, and after service stopped all of the infrastructure was stolen, concluding with the rolling stock being destroyed in an accident. Now the ROW is being reused for the Corredor Perimetral BRT in Campinas. As for low floor, we have the modern tramway in Rio, the railway converted to light rail in Santos and São Vicente, and... the mostly built but abandoned project in Cuiabá which is depressing enough that it's best to ignore it. Most modern uses of VLT nowadays refer to electric light rail vehicles, though not exclusively, because of the next category.

Diesel Light Rail. AKA, we don't want to abandon this loco-hauled railway/we want to reactivate this disused railway as cheaply as possible. Or not, in the case of Teresina, where they literally built a single track elevated viaduct along the river to carry their VLT to the edge of downtown. While it may be tempting to consider these light rail in the American sense, the reality is that they are mostly commuter railways operated with smaller diesel powered rolling stock, with awful headways. They're similar to the railbuses in Argentina, in that sense.

Finally, Gadgetbahns. VLT in these cases typically refer to monorails, and in particular that of Salvador. We'll... discuss this later. Thankfully monorail is the only viable gadgetbahn in common use in Brazil (cough Aéromovel cough), so the term is not usually misused here.

Overall, VLT is a term that can refer to completely different technologies, rolling stock, and service patterns.

Anyways, it's time for the city retrospectives. As usual, these are less about the maps and more about discussing interesting aspects of each city that didn't make it or weren't relevant for their MRA pages.

I'm glad I chose Criciúma as my first city for Brazil. One BRT line, well defined, with a single real service pattern. Too bad this simplicity would disappear immediately.

Curitiba, the city of BRT. It truly opened my eyes - mainly because the corridors had only a few services, but the services and the corridors didn't actually line up. In the west, CIC Norte was the end of the busway, but the BRT service didn't reach that far. In the south, a bus corridor continued with regular stops quite a distance, but I left it off the page and map (2.7 km to Terminal Central de São José Dos Pinhais). And across the city, on areas without BRT corridors, integrated tube stations were present, challenging my initial thought that they would only be on actual corridors. The approach taken has a benefit of allowing free transfers between lines without large terminals, and spreads benefits of the integrated transit network to parts of the city not on a BRT corridor. Finally, the Linha Verde. I get that the current southwestern part is useful, but... who thought a diagonal line that doesn't interchange with the east corridor was a good idea? Either way, construction on this project is taking forever. Hopefully it gets finished.

Porto Alegre was where the 'is this BRT' hit me in Brazil. Suburban metro with a gadgetbahn connector to the airport, sure. But so many bus corridors with shoddy often nameless stations but otherwise good infrastructure? And Av. Sertório, where I couldn't tell the station names on the actual stations because they had all faded or been vandalized? In the end my work was a mess and I put aside the fact that some of the corridors might make BRT Basics. In retrospect, some of the corridors definitely met the requirements now that I know about placed bumps on the road qualifying as physical separation, but they are very much on the low end of BRT. If the BRT upgrade plans from a decade ago ever go through, I'll gladly give them the treatment they will then deserve. But only if they put actual names on the stations.

To this day, I have no idea if the EPTG BRT corridor ever began service (Google implies it did, which is why I added it to the MRA), but it's definitely not in good shape. As for the Expresso DF Sul, it has been pointed out by multiple organizations that it is a prime example of 'BRT to nowhere' with its decaying stations located in the middle of fields. As for extensions, the husk of what looks to be planned stations on EPIA remains...

On a more cheerful note, I initially thought the EPDB and EPPN had BRT lines. Turns out that they're median gas stations that happen to look like BRT stations from Google Satellite View. Sigh.

Thankfully, the biggest drama with Goiânia was the new north-south corridor and... not knowing how many stations there are, actually. I'll find out when it actually opens. The opening of which has now been pushed back to 2023.

Uberaba had absolutely no drama at all. Except the service pattern of Route 100, which uses Terminal Oeste as an intermediate station.

The most interesting part about the Uberlândia work was that they have a BRS corridor as well going to the north. Will be interesting to see if it gets integrated into the BRT network.

Belo Horizonte's BRT system is MOVE, which is a pretty good system featured by the ITDP. Lesser known is the MOVE corridor in the northeast, which has no dedicated lanes at all, separate single directional stations on opposite sides of the MG-010 highway, and a tendency to be forgotten. It's not featured on my maps or the page on the MRA, but if dedicated lanes are provided, they may eventually find a place on the MRA.

I'll admit, Rio took forever because the complexity was daunting. But in the end the real decision was how to classify the commuter lines, and how to handle the Deodoro-Santa Cruz through operation that began recently. The map was more interesting than the city research, with the exception of the forever-stalled TransBrasil and wondering what they were thinking when they designed TransOlímpica

As for Niterói, I actually discovered TransOceânica by complete accident. But unfortunately, until they resuscitate Metro Line 3, the east side of the bay will remain a lower priority for the region as a whole.

An excellent use of a former rail corridor, the Santos VLT is the best currently operating in the country. Its planned extensions... well, we'll see how things go.

Sorocaba was interesting because I heard about it by randomly seeing the BRT inauguration on a YouTube channel months before BRTData added it. It was also where I first encountered the Faixa Azul in action and learned what it represented. It's a good system but really needs to be extended into the city center.

When doing the research for Campinas, problems kept on cropping up. First, BRTData said that it was operational, but in reality the BRS services using the BRT ROW were operational, but the BRT services were not. This was when I first learned about BRS and the distinction between it and BRT. Next, there is a median stopping BRT-Lite loop around the downtown. This BRS surprised me even though I knew about it - mainly because of its extent. In addition, I knew about the Sumaré-Americana EMTU corridor thanks to BRTData. However, I did NOT know it was part of a massive network of BRT-Lite corridors. This added a lot to the research, and I discovered the corridor by accident when seeing a station on an expressway and then following it... to Sumaré. But overall, shock aside, I have everything ready to go for when the Campinas BRT actually opens.

I'll admit, São Paulo was an experience. Brazil's foremost metropolitan area is huge, with lots of bus priority corridors. I debated whether or not to actually feature these on the MRA, but in the end I went and... did it. It was a massive amount of work, but the result is a full view of how the network actually functions and how São Paulo, like with Argentina and Chile's BRT networks, elevated the reliability and speed of buses without branding them as BRT (albeit with varying quality levels depending on where you are). Way more than Porto Alegre, that's for sure!

The surprise came when I was researching the ABC BRT alignment. Initially, I had thought that the stations I saw were for the ABD project. But then I did actual research on the alignment and discovered that São Bernardo do Campo had built its own separate BRT-Lite network with multiple corridors. It was a shock to me because I hadn't expected subnetworks to form within satellite cities already served by the EMTU, and there was no documentation of this at all in Wikipedia. In my panic as I added the first routes, I discovered Santo André's BRT corridor, and then a corridor in Guarulhos (which had no station names). One of the biggest retroactive post-publishing updates I've made to a MRA page in between initial page publishing and map publishing! But it was very useful and a learning experience since the Santo André corridor was particularly interesting.

Belém was a little challenging to research. Two São Brás stations was fine, but two Bosque stations was not. The northern portion was also missing station names on Google Maps, and I had some trouble finding information. The extension along BR-316 also wasn't managed by the city, so it confused me as well until I found documents for it. Will be interesting to see how the network evolves.

I never thought I'd have to drop an active network from scope, but here we are. BRT Feira began trial service with so few vehicles that the service may as well not exist. I never would have considered this possibility. It still hasn't left trial service, as far as I know. 4 buses operating a two corridor BRT network seven months after operations began is simply inexcusable.

In today's edition of "let's name the BRT after two destinations it doesn't go to", we have Fortaleza. Corredor Antônio Bezerra-Papicu does not reach either Antônio Bezerra nor Papicu, and also has single directional stations exclusively, which is interesting. Corredor Messejana-Centro similarly does not actually reach Messejana.

I really do want Fortaleza's network to succeed, but that will depend on the first leg of the Linha Leste opening.

Salvador has four interesting things. First, funiculars that are not meant for tourists and are an integral part of their transit network. In Valparaíso's case, I excluded them, but for Salvador I did not. Second, a BRS to BRT conversion project. It will take some work but the Corredor Lapa-Iguatemi will do much good for the city. Third, the BYD SkyRail project. Service patterns aside, it is a shame that they began by demolishing part of the suburban network instead of, well, starting with building the part of the line that isn't replacing the suburban rail line and then doing the suburban portion later once some benefit had been delivered. The end result of this is destruction of an important service with no replacement. And finally, this was when I finally realized that "Linha " is Brazilian for an urban roadway expansion project and that this is the real reason why Curitiba's Linha Verde has such a weird alignment. Turns out that Salvador's Corredor Azul and Corredor Vermelha don't have BRT at all yet.

There are a few things that stood out about Recife. First, the BRT network. Both lines have their quirks, with the Corredor Leste-Oeste having some missing infill stations, as well as a run down set of one way stations on the mixed traffic portion in the west. The Corredor Norte-Sul naturally makes a U-shape in downtown (sigh) and acts as an express for local BRS services on its northernmost stretches, which is interesting (and possible due to stops having pull outs for local buses). The network also has a lot of mixed traffic operation, unfortunately, and some abandoned stations.

Then there's Av. Sul, my mortal nemesis. BRTData called it BRT, and that caused a lot of thinking. After all, it doesn't even have stations, yet managed to be considered BRT? Was it a copy paste mistake that it was Bronze? Either way, as of the writing of the post, I invoked the 3 km minimum rule to downgrade it on the MRA, so that's something I don't have to worry about anymore.

Oh, and about the VLT lines having 51 and 102 minute headways? They should have a chat with the people who developed the MAX timetables in Calgary.

Teresina scared me when I first explored its BRT network. It didn't make sense to me. Of course BRTData showing the parts of the corridors that had no infrastructure didn't help. The idea of full BRT stations for open busway operation also didn't click with me at the time (I've now seen enough of it that it doesn't phase me at all). But now I can fully appreciate how disjointed the actual BRT infrastructure is, and hope that they manage to continue their very recent infrastructure buildout to create a world class transit network. Oh, but the VLT might require a LOT of work if they want to continue calling it a Metro.

We close with Manaus. Initially, I had dropped it from scope since every source said it was closed... until I saw a Diário do Transporte post that talked about a new order of high floor buses. That got me to return and... lo and behold. In the past few years, they've significantly expanded their network (albeit with some stations not seeing service yet) and have done some serious renovations. And it got me to do a formal MRA Scores review - not only for it but also for Porto Alegre and Santo André's high floor corridors. It was during that process that I learned about the bump dividers being considered physical separators and from there, Manaus became one of the few cities to be removed from MRA scope only to return. Pending Campinas, Anápolis, São José de Campos (in trial operation now), or Feira de Santana (in eternal trial operation), it will be the last Brazilian city for the Metro Route Atlas for a while.

Brazil's been a wild ride. I've watched and read enough material to get some proficiency in understanding Brazilian Portuguese, and I've learned a lot about how you can do bus priority. It's been a great project and while I likely won't be doing any massive projects (as far as city count is concerned) until I take on the Indian Subcontinent or anywhere in Europe, I'll make sure to internalize all of the techniques I've learned for research, page building, and mapping over the course of this long project.

And so, shortly, the Metro Route Atlas will have completed work for every currently operational higher order transit network in all of the Americas.