Deep Dive – Climate Change Series

This is the sixth article in a series that looks at the gap between climate rhetoric and reality in terms of what municipal governments say they are doing versus what they are actually accomplishing. The series compares and contrasts policies and outcomes in Halifax, Nova Scotia, with its nearest comparable American neighbor, Portland, Maine.

If you’re wondering why care at all about these two cities, you may like to read a brief overview of the series.

Previously, Part 5 examined the implications of ignoring research that consistently shows that high-rise development maximizes emissions. Here in Part 6, we look at the research that shows how high-rise densities can be accommodated with considerably fewer stories using courtyard-type buildings. Part 7 contrasts land use reform aimed at reducing auto-dependency in Portland with the false promise of electric vehicles in Halifax.

Urban Planner as 21st-Century Prophet of Density

In the 1950s and 60s, urban planners served as the prophets of urban renewal and freeway building. They condemned large swaths of historic cities as slums to be erased, rather than recognizing them as communities worthy of strengthening through reinvestment. Today, many urban planners—particularly in Canada—serve as prophets of density. And they are convinced that high-rises are the only way to achieve higher densities. In this article, I present research that dismantles this belief.

I think it is a mistake to talk about development primarily in terms of density. But for the purpose of this article, density is the focus. This said, pursuing density for density’s sake undermines good design. Focusing on human-scale design and quality of life gives you density as a byproduct. I’ve written elsewhere about four achievements a municipality must realize to overcome the rule of perpetual urban sprawl, in both its vertical and horizontal variations. All four achievements relate to design, not density.

In early 2024, Vancouver Sun journalist Douglass Todd said that “Canadian cities have become ground zero for the battle between Asia’s high-tower approach and the go-low attitude of much of Europe.” The evidence suggests that the high-rise faction is winning. The recently executed human-scale projects of the sort I’ve seen in Dublin, OH, Portsmouth, NH, Gaithersburg, MD, Tigard, OR, and Portland, ME are rare if not non-existent in Canada.

All this has significant climate implications. As I mentioned in Part 1, Canada shares the distinction with the United States of having one of the highest per-capita rates of CO₂ production on the planet, as shown below.

As I’ve outlined in Part 5, all research to date consistently shows that high-rise development maximizes emissions. The national embrace of high-rises that Douglas Todd speaks of is a big part of why Canada is an outlier on the chart above. Based on what I’ve seen of municipal politics, planning, and development in Halifax, this embrace is only intensifying.

This is unfortunate because peer-reviewed research has shown that there is a better way to proceed.

No Merit to Claim Building Taller is More Sustainable

I’d like to introduce this research with a summary taken from one of the four studies I presented in Part 5. This study was published in Nature in 2021 and is titled Decoupling density from tallness in analysing the life cycle greenhouse gas emissions of cities.

The results of this study suggest that there is no merit to the claim that building denser and taller is more sustainable. By building dense, low-rise urban environments, the same populations can be accommodated for drastically lower carbon costs and without having to significantly increase land use.

Edinburgh Napier University research team

Below, I’ve included the background of the team that did the study. Grants from the UK’s Royal Academy of Engineering, and Edinburgh Napier University funded their work.

Francesco Pomponi—PhD, Life Cycle Assessment
Jay Arehart—PhD, Architectural Engineering
Ruth Saint—PhD, Engineering and the Built Environment
Bernardino D’Amico—PhD, Timber Engineering, Computation and Structural Design
Niaz Gharavi—PhD, Structural Engineering

Let’s look at a real-world example by comparing two of the most densely populated neighborhoods in Canada with one in Spain. Vancouver’s West End neighborhood has a density of 24,000/km² and is packed with high-rises. In Toronto, you’ll find a high-rise neighborhood called St. James Town with a residential density of 44,000/km². And in Barcelona, you’ll find the human-scale Sagrada Familia neighborhood, with a residential density of over 47,000/km². You can see each neighborhood in the sideshow below.

Vancouver-Highrise-West-End

Three corresponding videos provide a sense of each neighborhood as experienced from the sidewalk. You can “walk” the West End here, St. James here, and Sagrada Familia here.

So the question becomes, why is it possible for a beautiful, tranquil Spanish neighborhood filled with cafes and a strong sense of community to provide residential densities comparable to high-rise districts?

To answer this question, I’m going to summarize the key concepts behind research that has spanned over fifty years and involved people who have devoted considerable time to the study of urban form, density, energy usage, and quality of life. My summary here is little more than an introduction to a significant amount of work that I hope you can digest within a few minutes of reading.

First, let’s return to the Edinburgh Napier University research. It’s rooted in their simulation of 5000 urban environments using real-world data taken from various European cities. The simulation creates urban environments block by block using the generalized steps shown below.

The details of how the simulation is structured are beyond the scope of this article. What’s relevant is that its execution produced both high-density, human-scale cities and high-density cities composed of high-rises. And as noted in the summary above, the researchers found that human-scale (a.k.a., low-rise) urban environments could accommodate high-rise densities found in Europe at a significantly lower cost in terms of emissions.

What I’d like to do in the rest of this article is to shed light on why this is possible. This “why” has everything to do with two concerns. First, there are the municipal regulations that govern the amount of light and air that must be made available to a building. Culture influences these regulations. What’s acceptable in China may not be embraced, for example, in Germany. Our second concern is with the building type shown below. It’s called a courtyard, and it offers specific advantages in terms of residential density because of the way light and air are shared across four facades. I’ll come back to both concerns shortly.

If you have a look again at the photo of Sagrada Familia above, you’ll see that the buildings are arranged in courtyard layouts.

Three Research Teams, Five Decades of Work

Many people have played a role in explaining why it’s possible to accommodate high-rise densities at a reduced number of stories on the same site.¹ The architectural theorist and researcher Phillip Steadman is, in a sense, the lynchpin of these efforts. He was involved with this work at the outset and later collaborated with other researchers. He has also contributed a lot to public awareness of these findings.

Steadman has noted that it’s counterintuitive to claim high-rise densities can be accommodated at a reduced number of stories. Yet this reality has been known since the early 1970s, when Lionel March and Leslie Martin—two prominent figures in twentieth-century architecture—published their research findings saying as much in a well-regarded book titled Urban Space and Structures.

March, Martin, and Steadman make up what amounts to the first of three research teams that have focused on the relationship between building height, building type, and density. Below, I’ve listed the members of these teams to give you a sense of their backgrounds.

Original research team at Cambridge University’s Centre for Land Use and Built Form Studies:

Lionel March – Cambridge-educated mathematician and architectural theorist with a long, celebrated career.
Sir Leslie Martin – Ph.D. Arch. Head of Cambridge’s architecture program and designer of the UK’s most prominent post-war building, the Royal Festival Hall.
Philip Steadman—As a recent Cambridge graduate, Steadman joined the newly formed Centre for Land Use and Built Form Studies that Martin and March established in 1965. He’d go on to become an authority on urban form and energy use, working with the Dutch team below, and leading the team of researchers at University College London, also listed below.

Dutch research team:

Meta Berghauser Pont—PhD, Urbanism (Urban Morphology and Density), MArch (Architectural Theory)
Per Haupt—PhD, Urban Planning and Design

University College London research team:

Philip Steadman—MArch, Emeritus Professor of Urban and Built Form Studies, Energy Institute, Bartlett School of Energy, Environment, and Resources, University College London
Ian Hamilton—PhD, Energy and the Built Environment
Homeira Shayesteh—PhD, Architectural and Urban Studies
Stephen Evans—MSc, GIS & Land Information Management and Mapping
Graciela Moreno—MArch, Urban Design
Michael Donn—PhD, Building Sciences
Daniel Godoy-Shimizu—Research Fellow in Building Physics Modelling at University College London’s Bartlett School of Environment, Energy & Resources

The Alan Turing Connection

I wanted to share an interesting story about Alan Turing’s indirect influence on March and Martin’s work on Urban Space and Structures. Perhaps you know the name? He’s considered to be one of the most innovative thinkers of the twentieth century. If not for Turing, it’s possible that March and Martin would have never even met.

Turing is the father of computer science and played a critical (but not singular) role in breaking the German enigma codes during World War Two. Had those codes not been broken, it’s less clear when or if an Allied victory would have been secured.²

I noted above that Lionel March’s career was a fusion of mathematics and architecture. In fact, he was a math prodigy. But March also grew up as the son of a building superintendent, which wasn’t a background that typically led you to Cambridge.

At age 17, March conceived a theory that attracted the attention of mathematicians in the UK, who forwarded his 16-page paper to Alan Turing. This led to Turning writing a letter to March praising his theoretical work. Later, Turing wrote a letter of recommendation to support March’s application to Cambridge.

Once at Cambridge, March developed a complementary interest in architecture. This led him to work with Leslie Martin on those counterintuitive findings we’re focused on in this article.

As mentioned, Steadman worked with March and Martin early in his career. In fact, a year before Urban Space and Structures went to press in 1972, Steadman co-authored a book with March titled The Geometry of Environment.

With the researchers introduced, let’s return to the two key factors that make it possible for high-rise densities to be accommodated at a reduced number of stories on the same site. They are, again, (1) regulations governing light and air, and (2) the use of courtyard buildings.

The Importance of Light and Air

Most municipalities have regulations in place to ensure buildings receive sufficient amounts of light and air. When regulations are non-existent, you can wind up with a place like Kowloon, a small, 6.4-acre, walled city under Chinese jurisdiction that sat, oddly enough, inside British-controlled Hong Kong. It’s a worst-case scenario that I’m using to make a point.

The place has an interesting history that I’m going to skip here, except to say many residents involved themselves in drug trafficking and other forms of crime. Living conditions were dark, damp, and grim.

The skyline you see in the photo below no longer exists. Hong Kong authorities demolished Kowloon with Beijing’s consent in 1993. But by the late 1980s, it had grown to a city of 35,000.

The bird’s-eye view of a model of Kowloon below illustrates why so little light and air permeated the buildings. Before its demolition, Kowloon was the densest city in the world at 1,300,000/km². So yes, if we are simply aiming to maximize population density, then constructing wall-to-wall high-rises is the way to go. But of course, that is not how most cities operate.

All this is to say that getting sufficient light and air to buildings is important. It follows then that most cities with high-rises have regulations that keep them separated from one another.³ As we’ll see in the next section, when researchers looked at high-rise buildings in the EU and the UK, they found that most could have been designed as courtyard-type buildings at much lower heights and still provide identical floor space. Why? Because the need to separate buildings largely disappears with human-scale development.

None of this suggests that you can create a city that looks like Vienna and have its density exceed the most packed district in an Asian city filled with towers. For example, Barcelona’s Sagrada Familia neighborhood does not reach the level of density found in Hong Kong’s Kwun Ton district (i.e.,47,000/ km² versus 57,000/km²⁾. What this particular comparison tells us, however, is that you can build at the human scale and get a density that is close to the most intense high-rise districts in the world.

The Significance of Courtyard-Type Buildings

In their pioneering work on urban form and density, March and Martin factored in assumptions regarding minimum distances between high-rise buildings. They were the first to show that, when factoring in these minimums, it was possible to achieve comparable densities on a site with significantly fewer floors using courtyard-type buildings.

Below are images of the two building types they defined that are relevant here.⁴ On the left, we have a pavilion (i.e., a high-rise), and on the right, we have the courtyard shown previously. Notably, the two buildings represent equivalent densities.

Steadman’s team illustrates this principle with a real-world example in the figure below. On the left, you see a design for a site in London created by the architecture firm Foster + Partners. On the right is a courtyard-type building providing the same square footage on the same site.

This project, known as “250 City Road,” was approved by then-mayor Boris Johnson with high-rises intact before Steadman’s team created their conceptual alternative.

So, how is it possible that a structure built at a considerably lower height can provide the same floor space as a much taller building?

The reason is this. A courtyard can be built at a considerably lower height than a pavilion, occupy a greater percentage of the building site, and still provide the necessary light and air.

Let’s see how this works.

Rather than think of the courtyard as a single building, consider it as being four connected buildings sharing the same interior space, as shown below on the right. That shared open space provides light and air to four separate building facades. This efficient sharing of space frees up land for an expanded footprint of a building having fewer stories than the pavilion it replaces.

There is a fair bit of analysis and math behind this general idea in which courtyards occupying a larger percentage of a parcel can provide high-rise densities. If you’re interested in some of this background, you can read a paper by Lionel March titled Mathematics and Architecture Since the 1960s. You can also have a look at a paper Martin wrote titled The Grid as a Generator.

Providing Empirical Proof

Theory is one thing. Empirical proof is another. The Dutch researchers named above, Meta Pont and Per Haupt, analyzed high-rises in metropolitan Berlin, London, Barcelona, Stockholm, and Amsterdam and confirmed that what March and Martin theorized held in real life. Philip Steadman and his team at University College London did similar work with additional high-rises in the UK and generated consistent results. In most cases, courtyards built with considerably fewer stories provided equivalent floor space on the same site.

Closing out this article, I’d like to say just a few words to clarify how these efforts fit into a broader focus for each team that goes well beyond the scope of this article.

Principal Focus of the Dutch Team

In a general sense, Pont and Haupt are concerned with the relationship between density and urban form. People conventionally define density in a couple of different ways, such as the number of dwelling units per hectare, population density, and land use intensity. None of these measures, however, tells you much, if anything, about the form of development.

Recognizing this gap, Pont and Haupt developed a methodology that relates density to urban form. In 2005, they presented this methodology in a paper titled The Spacemate: Density and the typomorphology of the urban fabric. Later, in 2020, they expanded on these ideas in a book (freely available), titled Spacematrix: Space, Density, and Urban Form. For our purposes, I’m going to focus briefly on a narrow, but central concept relating to building height.

Pont and Haupt use an expanded definition of density, including the three attributes listed below.

Floor Space Index (FSI)—The ratio of the total floor space to the overall size of the parcel on which the building sits. In the US and Canada, this same ratio is called a Floor Area Ratio (FAR).
Ground Space Index (GSI)—Also referred to as ground coverage, this is the percentage of the parcel occupied by the building.
Building Height (L)—The number of stories (or “layers”) in a building.

The relationship between these three attributes is:

Building Height (L) = Floor Space Index (FSI)/ Ground Space Index (GSI).

Looking at the formula above, we see that to reduce a building’s height (L) and maintain the same amount of square footage on a parcel (i.e., FSI remains unchanged), we need to increase the percentage of the parcel occupied by the building (i.e., GSI must increase). So the question becomes, can we increase the amount of land taken up by the building to reduce its height and still provide adequate amounts of light and air? And, as I said, in most of the real-world cases that Point and Haupt examined, the answer was “Yes.”

Principal Focus of the of University College London (UCL) Energy Institute

Philip Steadman and his fellow researchers at the UCL Energy Institute study the relationship between building form and energy usage. In Part 5, I talk about one of their studies, which found, like all related studies done to date, that high-rises consume significantly more energy per square foot to operate than human-scale buildings. One study from 2018, called Energy Use and Height in Office Buildings, found that buildings having 10 or more stories generated 67% more CO₂ emissions relative buildings having 5 or fewer stories.

Since 2009, the Institute has been developing an extensive 3D model of the building stock in regions across the UK. They describe this work in a journal article titled Building stock energy modelling in the UK. Below, Steadman (with striped tie) appears with other members of the Institute, along with Channel 4 presenter Kevin McCloud, who has featured their work in a piece called The Great Climate Fight (available online to UK residents only).

Steadman Tying it All Together

In 2012, Steadman published a paper showing how March and Martin’s theoretical work supports the empirical findings of Pont and Haupt. It’s titled Density and Built Form: Integrating ‘Spacemate’ with the Work of Martin and March. For this article, I only want to make you aware that this paper exists. In bringing together the work of the English and Dutch researchers, Steadman provides the information that allows us to grasp, more deeply, why courtyard-type buildings are good at accommodating higher residential densities.

In this paper, Steadman notes, “The general public and large parts of the architectural and planning professions still believe…that in order to raise densities, it is necessary to build higher. He emphasizes that “this belief needs to be seriously questioned and carefully qualified.”

Why Prophets of Density in Halifax Ignore the Research

The research presented both in this article and Part 5 raises a question. Why would prophets of density ignore the information that researchers make available to the public? Part of the answer lies with Upton Sinclair, who famously said, “It is difficult to get a man to understand something when his salary depends on his not understanding it.”

Municipal planning is inherently hierarchical and highly political. In the absence of municipal leadership that can break the rule of perpetual urban sprawl, inertia plays out in all the wrong ways. To maintain a steady paycheck, many municipal planners will keep their heads down and buy into bad ideas to keep the right people happy.

This is the situation in Halifax, a medium-sized metro area that, as of July 2024, had the largest per capita number of high-rises under construction in all of North America. Halifax residents, including a former councilor, have all gone before municipal council to express concerns regarding high-rises and emissions. In return, municipal leadership and my former planning colleagues alternatively dismiss or ignore their entreaties for reasons relating to the culture I described in Part 2.⁵

As described in Part 3, there is no relationship between Halifax’s climate action plan (HalifACT) and its development practices.⁶ This raises questions of a professional, moral, and ethical dimension worth exploring further.

Does Portland Continue to Become a Better Version of Itself?

Portland offers reason for cautious optimism regarding minimizing high-rise-related emissions. As I described in Part 2, Maine has a strong communitarian culture. Twenty-seven different community organizations took part in creating Portland’s master plan. This plan is understood to be a “statement of community values and a framework to advance those values.” A key objective of this plan is to maintain the city’s “authentic character.” It is inherently a vision that values human-scale development.

Although Portland’s climate action plan doesn’t acknowledge the fact that high-rises maximize CO² emissions, the city hasn’t yet embraced “Asia’s high-tower approach” as seen in Halifax. Over a twenty-six-year period—between 1999 and 2025—only two have been built in Portland, one 18 stories tall and the other 10 stories tall.⁷ Notably, in July 2024, Halifax had thirty-three high-rises under construction. Portland had none.

Portland’s vision, policies, and practices have produced results. Over the past twenty years, the city’s national reputation as a desirable place to live, offering a high quality of life, has only grown. In 2025, Cambridge-based Apartment Advisor considered Portland to be the third-best city in the country for recent college graduates. A 2023 NY Times analysis of census data shows that college graduates are sticking around Portland in larger numbers.

There’s always legitimate room for criticism, but what Portland has done has largely been working. Collective efforts have made Portland a better version of itself. The question becomes, will the city continue to build on its strengths, or will it change direction and embrace the “high-tower approach” to urbanism that’s ubiquitous in Asia and Canada?

In November 2025, Portland City Council significantly increased height limits in the historic downtown. This vote reflects a conviction that high-rises are the only way to achieve higher residential densities. The decision reflects precisely the false belief that Steadman, Pont, Haupt, March, Martin, and other researchers have worked to correct.

In a non-existent ideal world, Portland would reverse course on this decision before landowners sell parcels at inflated prices due to increased height limits. There are four good reasons for doing this.

First, as the research makes clear, human-scale heights can produce densities that rival or exceed those found in high-rise districts. It’s about getting the design right, and Portland knows how to do this.

City staff and private sector design professionals have established a record of delivering quality, human-scale projects. Portland already has a form-based code that is shaping development in one part of the city. It’s a matter of executing one or more public-facing design charrettes that produce either a form-based code fostering as-of-right development or looser design standards underpinning a discretionary approval process.

The result of either process would include creating courtyard-style development in suitable locations that abide by the community vision laid out in Portland’s master plan. This kind of development not only produces a superior public realm. It’s also more suitable for cold-weather cities as high-rises deprive pedestrians of sunlight and produce unwanted wind in the winter months.

The second reason for reversing course on the height increases is that Portland’s own history shows us that high-rises are not needed to support a larger population.

Have a look at the map below. It was made shortly before 1920 when Portland’s population stood at 69,272. The vast majority of residents lived on a heavily urbanized peninsula. You can see that it’s covered with a dense network of roads. You can also see that building footprints look a lot like courtyard-type layouts. That’s not coincidental. Now, have a look at the neighborhoods on the mainland (e.g., Deering and Woodfords) and you’ll notice they’re more sparsely populated. With this in mind, I’m estimating that about 75% of Portland residents in 1920 lived on the urbanized peninsula. This gives us a figure of 51,954.

Now let’s jump to the present day. According to the latest census in 2020, Portland was home to 68,409 people. In the table below, you’ll find 23 census “block groups” on the peninsula that were home to 22,982 people. What this means is that back in 1920, an additional 28,972 people lived on the peninsula (i.e., 51,954 minus 22,982) without there being a single high-rise. Even if my earlier estimate of 75% should really be more like 65%, that still means about 22,044 more people lived there in the past than today—again, without high-rises.

The third reason to not increase building heights relates to affordability. Many people share a goal of increasing the availability of workforce housing, so wage-earning Portlanders who work in the city can afford to live in the city. As shown below, the higher you build, the more expensive the cost on a per square foot basis for reasons I’ll speak to in a separate article.

Given this, it makes little sense to talk about affordable housing and high-rises in the same breath. A solid affordable housing policy benefits from having community-driven urban design standards in place that produce human-scale development throughout a municipality.

The fourth and final reason Portland should forgo high-rises is the emissions relating to both embodied and operational carbon. As I’ve described in Part 4 and Part 5, building demolition and high-rise construction are antithetical to any credible municipal program aiming to reduce emissions.

Portland probably won’t reverse their decision, but they could. Alternatively, Portland taxpayers could then send its municipal politicians and planners to Halifax to study the place firsthand. The question is whether a Canadian city that aspires to be more like Dubai holds any appeal.

Researchers like Steadman, Martin, March, Pont, and Haupt are all telling us we have choices regarding what we build. Shanghai and Barcelona illustrate how different the outcomes can be.

Although the Edinburgh Napier University simulations would include the footprints of real-world courtyard type buildings their study is not focused on how and why individual courtyard type buildings can accommodate high-rise densities. ↩︎
In 2014, Hollywood devoted a movie to Turing called The Imitation Game. Although I cannot vouch for it’s historical accuracy, this wonderful six-minute clip does Turing justice. ↩︎
There are also structural reasons for maintaining minimum distances between high-rises which relate to their foundations which is relevant, but not a subject I’m going address in this article. ↩︎
March and Martin define three key building types in their work, the court, pavilion, and streets. Courts correspond my use of the term courtyard. Pavilions are synonymous with high-rises (a.k.a., towers) Streets is the building type that’s not relevant to this article. They are akin to attached housing that runs parallels to a street. I find the term street a bit misleading, and am not sure why they didn’t use a word like townhouse, or attached house, but nobody was asking me for my opinion in 1972. ↩︎
I’d like to be proven wrong here, but cultural dynamics are deeply entrenched as I describe in Part 2. To illustrate this point, I can refer you to a public hearing held in September 2021 in which the pathologies of municipal politicians and planners are on full display. In this particular hearing, council voted in support of moving three high-rises forward in the approval process. Only one councilor, Patty Cuttell, served the public interest and voted in opposition to the high-rises’ construction. ↩︎
The disconnect between any climate action plan development practices is a problem given that 61% of emissions tie back to municipal decision-making in terms of what and where things get built. ↩︎
The 10-story InterMed Building (2008) is 135 feet tall. The 18 story Casco Building (2023) stands 204.5 feet tall. ↩︎

Up Next: Part 7 contrasts land use reform aimed at reducing auto-dependency in Portland with the false promise of electric vehicles in Halifax. Previously, Part 5 examined the implications of ignoring research that consistently shows that high-rise development maximizes emissions.

Accommodating High-Rise Densities with Courtyard-Type Buildings