What you may not realize is that solutions to this problem aren’t just theoretical; many are already underway. My goal is to connect the dots on a surprisingly elegant two-part strategy: first, by exploring how today’s slow, often frustrating upgrades can make the subway cooler, and second, by revealing how we can one day capture that waste heat to help warm the very buildings we live and work in.

I see two primary paths forward, which can be summarized simply: reduce the heat, and reuse the heat.

How can we reduce the heat generated?

The most direct approach is to make functional improvements that increase the efficiency and quality of service. A more reliable, modern subway network would naturally be a cooler one. When trains run smoothly, they spend less time in the heat-generating cycle of braking and accelerating.

The Regional Plan Association (RPA) put forth proposals for modernizing the system and increasing reliability in its 2018 report, Save Our Subways.1 Many of the necessary upgrades also happen to be solutions for heat reduction.

1. Upgrade Aging Infrastructure

Much of the system's ambient heat comes from outdated technology. The linear fluorescent lights found in many stations are inefficient, have short lifespans, and radiate significant heat. To address this, the MTA began a system-wide conversion to cooler, more efficient LEDs in 2024, an effort expected to be completed in 2026.2

Additionally, a significant portion of the subway’s signal technology is nearly a century old. These antique systems not only generate waste heat, but are also a source of system delays. Upgrading from the old Fixed-Block Signaling—which acts like a series of traffic lights to allow one train to occupy a ‘block’ at any given time—to modern Communications-Based Train Control (CBTC) is a game-changer. With CBTC, trains constantly communicate their exact position and speed, allowing them to run closer together safely, reducing dwell times in stations and minimizing that stop-start cycle. Less time sitting at a platform means less heat dumped where passengers are waiting.

2. Lighter Train Cars

The design of the train cars themselves plays a crucial role. Since heat loss is proportional to mass3, transitioning to lighter trains would decrease power demands and, consequently, waste heat.

The introduction of open gangway trains, where you can walk from one end to the other without passing through doors, is a win-win-win. These designs provide more space for passengers, reduce the overall weight of the train, and improve safety by making it harder to subway surf.4 So far, 20 of the MTA’s 6,910 train cars have open gangways.5

6

How can we reuse the heat that is generated?

Now, for an idea I am quite passionate about: reusing heat. If you know me personally, you likely have heard me propose this with great enthusiasm. It’s an elegant solution that turns a problem into a resource.

1. Regenerative Braking

In a traditional braking system, a train’s immense kinetic energy is converted into heat through friction, and is radiated away to heat those of us standing on the platform. Regenerative braking captures that energy instead. By essentially running the electric motors in reverse as a train slows down, the system acts as a generator, converting the energy back into usable electricity.

As the laws of thermodynamics teach us, energy can’t be created or destroyed, but it can be transformed. Regenerative braking transforms kinetic energy into electrical energy, instead of letting it become waste heat.

This isn’t a futuristic fantasy, it’s already happening! The MTA has been incorporating this capability into new train cars for the last 20 years or so. As of last year’s Capital Plan, 67% of the entire fleet is regenerative braking-capable, with most of the remaining cars scheduled for replacement in the 2025-2029 window.7

So what’s the problem? We aren’t capturing all the energy we could be.

A 2022 NYSERDA study8 found that a significant amount of potential energy is still being lost. The electricity captured by a braking train is fed back into the third rail to be used by other accelerating trains nearby. But if there are no other trains in the immediate vicinity to use that energy, it has nowhere to go. In that case, it dissipates as… you guessed it, heat.

The solution? Energy storage. By installing batteries either along the tracks or inside the train cars, the regenerated electricity can be stored instead of wasted. That stored power can then be used when it’s needed, to help accelerate the next train that comes along or even power the stations themselves.

For a glimpse of what’s possible, look at Barcelona. Their metro system uses regenerative braking to cover about 33% of their trains’ energy needs, and to power 28 subway stations.9 They even route surplus electricity to nearby charging stations for electric vehicles, creating a new revenue source.

2. Thermal Energy Capture

Beyond regenerative braking, there is another powerful way to reuse the heat generated by the subway system: capturing it directly from air and redirecting it to something we actually want to be hot.

The foundational technology for this idea already exists right here in New York. The Second Avenue subway lies too deep underground to be ventilated by simple sidewalk grates. Instead, it uses a hydronic (water-based) system to cool its stations.10 The hot station air is blown over a cooling coil filled with cold water. The heat from the air is absorbed into the water, cooling the station. This now-warmed water is pumped to a building on the surface, where a cooling tower releases this captured heat into the atmosphere.

The MTA is already doing the hard work of capturing vast amounts of thermal energy. The crucial next step is to see that captured heat as a valuable commodity instead of waste. A team of students did just that, proposing that instead of venting all that energy into the sky, the hot water from the MTA’s system could be diverted to a nearby building to provide its hot water. It’s a brilliant application of a circular economy—one system’s wasted output is another system’s valuable input.

This type of large-scale solution is most feasible for new, deep-bore stations. For the vast majority of the system's century-old stations, a more modular technology called radiant cooling panels offer a promising path forward. Think of radiant cooling as creating a cool surface that pulls heat away from its surroundings, much like a cold window on a winter day. The thin panels have a closed loop of chilled water inside them, and would be installed on platform walls or ceilings. This creates a double cooling effect: they directly absorb radiant heat from passengers, providing immediate comfort, and they gently cool the surrounding air. Tests for a pilot program on the London Tube found that these panels could lower ambient temperatures by a remarkable 18-27 °F.11

Let’s scale up this vision and connect the pieces. Imagine not just one building, but an entire neighborhood connected to a shared network of underground pipes—a geogrid or thermal energy network.12 Such a system could allow buildings to efficiently share thermal energy. An office that needs A/C in the summer could dump its waste heat into the shared loop, which a neighboring apartment building could then draw from to heat its water.

This is where the concepts converge beautifully. The subway system, as a massive and consistent source of waste heat, could become the anchor of these neighborhood-scale thermal networks. The radiant cooling panels in the stations could even be powered by the geogrid itself. Instead of each station needing its own chiller unit, they could simply tap into the neighborhood’s shared loop for their supply of cold water. This creates a truly symbiotic relationship: the city helps cool the subway, and the subway helps heat the city.

Imagine a future where the waste heat from a braking train in Brooklyn helps warm the classrooms of a nearby public school. Or where the constant heat generated by a data center in Manhattan provides hot water for a low-income housing development. This is a system that creates a more just and resilient city—lowering energy costs for residents, creating new green-collar jobs, and transforming a large piece of public infrastructure into an active engine for decarbonization. It’s a tangible step toward building a more humane system, one where the invisible arteries of our city work in concert to help everyone flourish.

Can we put numbers on this?

Coming next time, I’ll attempt to back up these ideas with some rough calculations of cost and energy savings. Stay tuned!

How hot are they, really?

I was surprised to find a lack of recent, publicly available data on subway platform temperatures2. While historical measurements from the Regional Planning Association do exist, they aren’t very extensive. This data gap sent me on a short but sweaty quest… As any person with a thermometer would do, I decided to take my own measurements. I focused on morning and evening, since the RPA measurements were mostly at midday, and was especially curious to see if the temperature varied much with cooler outdoor temperatures.

The table below compiles my readings with some that I found online345. I acknowledge that this isn’t the most rigorous sample size, but I do think it illustrates that there is a problem.

Why are they hot?

When I searched for a quantitative analysis of subway heat sources, I was again met with limited success. While many qualitative descriptions exist, I could find only one study (from 2003!) that produced a mathematical model for heat transfer in subway systems6. The energy balance equations in the paper are a great starting point, but they don’t directly apply to our modern system. With that in mind, here is a breakdown of what makes our subways so hot.

According to the study, the top contributor of heat (85%) is the braking mechanism. The friction to stop the train converts its kinetic energy to thermal energy. If you would, picture a subway system that operates perfectly—no delays, trains always on time. In this (hard-to-imagine) scenario, trains would only need to brake when they stop at a platform. But, as you know, this does not describe our system. When lines become congested due to delays, trains are forced into a cycle of stopping and starting. This means more braking, which means more heat.

Heat loss from the train cars themselves contribute the next highest amount of heat (13% according to the study). Inside the cars, heat is generated by body heat and lighting. I would theorize that this share may be much higher today, as the air conditioning vents hot air directly onto the platform. This makes me want to build my own model for heat transfer in NYC subways specifically…

On the platform, the sources of heat are similar to those inside the train—lighting and the body heat from millions of commuters. These are compounded by the heat generated by signaling technology and displays.

There aren’t many places for this heat to go. There are some ventilation shafts in the tunnels, but there is a lack of effective airflow in most stations themselves. So the heat builds up in the tunnels, warming up the surrounding thermal mass (rock and earth). This creates another cycle where the tunnels and platforms become heat reservoirs, trapping the thermal energy and making it even hotter. The heat has nowhere to escape, and so, it heats you.

What can we do about it?

I’ll explore this question in more detail in future writing, but I do have a few ideas based on my research thus far.

The first approach is to reduce the amount of heat generated in the first place. This would require prioritizing functional improvements to the system as a whole. A more efficient, on-time subway network would naturally generate less heat, as trains would spend less time in the vicious cycle of stopping and starting. By investing in modern signaling technology and operational improvements, we can create a system that is not only more reliable for riders but is also less taxing on its own infrastructure.

The second approach, a prospect I find extremely exciting, is to leverage the heat that is being generated. Regenerative braking technology can be used to convert the kinetic energy of a braking train into electrical energy, capturing it in a battery to power the trains or other parts of the subway system. Additionally, thermal networks can be built to capture the excess heat and distribute it to heat water and spaces aboveground. Imagine a system where the heat from your commute helps to warm a nearby apartment building! This is the kind of thinking that can transform a liability into a resource, and a problem into a solution.

That is the purpose of this section, "Building in Public."

It will serve as a living record of my professional journey, but my hope is for it to become far more. I am proud of the work I've been entrusted with, and I believe good stewardship means sharing the lessons learned from that work. For me, every project has been an attempt to engage with the very questions that drive this publication.

My aim is that by openly recounting these stories, they might spark an idea, ignite a conversation, or offer guidance to others wrestling with similar problems.

Subscribe now

“You take delight not in a city’s seven or seventy wonders, but in the answer it gives to a question of yours. Or the question it asks you, forcing you to answer, like Thebes through the mouth of the Sphinx.”

Imagine a dialogue between Marco Polo and the Chinese Emperor Kublai Khan, in which Polo vividly recounts the cities he discovered on his travels. This is the setting of Italo Calvino’s Invisible Cities, the immersive book that inspired the title of this publication.

Eventually, he reveals a secret (spoilers!): “Every time I describe a city, I am saying something about Venice.” Though Polo recounts 55 different cities, they are, in reality, all fragmented reflections of his hometown.

This idea reveals a profound truth about our cities—and more broadly, our institutions—they are invisible in their totality. Consider New York City. Even if you dedicated a lifetime to walking down every street, riding every subway, and working within every branch of government, you still wouldn’t have seen everything. And here's where it gets even more fascinating: human experience is beautifully peculiar. Two people could follow the exact same path, yet their individual perspectives would paint vastly different realities for each of them.

“Cities, like dreams, are made of desires and fears, even if the thread of their discourse is secret, their rules are absurd, their perspectives deceitful, and everything conceals something else.”

This publication will explore how the tangible elements of our cities and governments—policies, infrastructure, and more—shape, and are shaped by the intangible: desires and fears, quality of life, a sense of justice, and our collective capacity to flourish. We will move beyond the visible to the invisible layers of human experience that define them.

Our perception of cities and governments is undeniably subjective. However, I will strive to objectively write about what is, and why it is, in all of its inherent subjectivity. Only after seeking understanding will we (perhaps) move on to what ought to be.1 For it is vital to understand something before attempting to change it.2

I hope these sometimes interconnected, often winding threads of writing contribute to the weaving of a more complete tapestry of answers about our invisible cities.3 And that through this shared exploration, we may better understand the questions we ask, and the questions that are asked of us.