Visualizing the Material Impact of Global Urbanization

Today, the global population is estimated to sit at 7.91 billion people.

By the end of 2022 or within the first months of 2023, that number is expected to officially cross the 8 billion mark. Incredibly, each new billion people has come faster than the previous—it was roughly only a decade ago that we crossed the 7 billion threshold.

How did we get here, and what has global population growth looked like historically?

In this series of six charts from Our World in Data, we’ll break down how the global population got to its current point, as well as some big picture trends behind the data.

#1: Mapping the Population Over 5,000 Years

New York, São Paulo, and Jakarta were not always bustling metropolises. In fact, for long parts of the history of civilization, it was unusual to find humans congregating in many of the present-day city locations we now think of as population centers.

The human population has always moved around, seeking out new opportunity and freedoms.

As of 3,000 BC, humans could be mainly found in Central America, the Mediterranean, the Fertile Crescent, and parts of India, Japan, and China. It’s no coincidence that that agriculture was independently discovered in many of these same places during the Neolithic Revolution.

#2: The Hockey Stick Curve

For even more context, let’s zoom way out by using a timeline that goes back to when woolly mammoths still roamed the Earth:

From this 10,000-foot view, it’s clear that human population growth started going exponential around the time of the Second Agricultural Revolution, which started in the 17th century in Britain. This is when new technologies and farming conventions took root, making it possible to grow the food supply at an unprecedented pace.

Soon these discoveries spread around the world, enabling population booms everywhere.

#3: The Time to Add 1 Billion

The data and projections in this chart are a few years old, but the concept remains the same:

It took all of human history until 1803 to reach the first billion in population. The next billion took 124 years, and the next 33 years. More recent billions have come every dozen or so.

So why then, are future billion people additions projected to take longer and longer to achieve?

#4: The Growth Rate is Shrinking

Because of demographics and falling fertility rates, the growth rate of the global population has actually been on a downward trend for some time.

As this growth rate gets closer to zero, the population curve has become less exponential like we saw in the first graphs. Population growth is leveling out, and it may even go negative at some point in the future.

#5: The Regional Breakdown

Although the rate of population growth is expected to slow down, there are still parts of the world that are adding new people fast, as you can see on this interactive regional breakdown:

Since 1973, Asia has doubled its population from 2.3 billion to 4.6 billion people.

Comparatively, over the same time frame, Europe has gone from 670 million to 748 million, equal to just an 11% increase.

#6: The Present and Future of Population Growth

Population projections by groups like the United Nations see the global population peaking at around 10.9 billion people in 2100.

That said, there isn’t a consensus around this peak.

Organizations like the Institute for Health Metrics and Evaluation (IHME) have a different perspective, and they have recently modeled that the global population will top out at 9.7 billion people by the year 2064.

As we climb to surpass the 8 billion mark in the coming months, it will be interesting to see what path humanity ends up following.

This Giant Map Shows All the Metropolitan Areas in the U.S.

The United States is the third most populous country in the world, made up of close to 20,000 cities and towns, and 333 million individuals.

Dividing these population clusters into a coherent framework of statistical areas is no small feat, and the U.S. Census Bureau’s latest map shows just how complex of a task it is.

This enormous map—which covers the entire country, including Puerto Rico—includes 392 metropolitan statistical areas and 547 micropolitan statistical areas.

For reference, here are all the current metropolitan statistical areas in the United States, organized by population:

From Metro to Micro

The wide variety of population patterns around the country can make it tricky to divide regions up into uniform units. There are two main divisions to consider when viewing this map:

1. Metropolitan Areas (metro areas) have at least one urban core area of at least 50,000 population. These are the largest population centers, sometimes encompassing many counties. In some instances, these metro areas are further subdivided into Metropolitan Divisions.
2. Micropolitan Areas are the smallest areas measured on this map (indicated by a lighter shade of green). These smaller regions, which are generally located further away from large cities, have at least one urban core area of at least 10,000 but fewer than 50,000 people.

One thing to note about all of these definitions is that the cities in these regions must have significant ties to a neighboring region—usually in the form of commuting ties. This is what warrants binding adjacent counties into a measurable area.

Another unique layer of data on this map is the shading that indicates the actual urbanized area within metro areas. In the example of Atlanta-Sandy Springs-Alpharetta, it’s easy to see how urban sprawl has expanded the urban area into a number of neighboring counties.

With this context in mind, we’ll take a closer look at three points of interest on the map that show this concept at work with varying degrees of complexity.

Level One: The Central City

The Texas Triangle offers what is perhaps the most straightforward example of metro areas.

As seen above, Houston, Austin, and San Antonio anchor their respective regions, and surrounding counties are bundled into a metro area. The surrounding counties have all been identified as having ties to the central county, and, in some cases, the urban area has spread into the neighboring county over time.

Level Two: The City Cluster

The region anchored by Salt Lake City requires more thought to divide into statistical areas.

While there are a number of population centers in the area, including Salt Lake City, Provo, and Ogden, they all have enough of an economic “magnetic pull” to warrant splitting the region into distinct statistical areas.

Of course, regions are always evolving, and occasionally these areas are updated. Salt Lake City and Ogden were previously combined into a single metro area, but were separated in 2005.

Level Three: The Megaregion

New York City is the ultimate challenge for planners looking to categorize population centers into a neat and tidy statistical box.

For one, the contiguous urban area is massive, stretching from the west side of Long Island out to the east side of New Jersey. In addition to New York City itself, the metro area includes 19 other municipalities with over 100,000 people.

Next, NYC is an unparalleled economic magnet. Measuring commuting activity is a challenge because a wide variety of people visit the city for so many different reasons. The interconnectedness of the Northeast Megaregion also adds to the complexity.

New York-Newark-Jersey City is such a big pie to carve up, that four of the country’s 11 metro divisions (as indicated by the italicized text and dotted lines) occur in this one area.

Blurring the Lines

Population patterns are constantly changing across the country, so the next version of this map may have a number of changes on it. Our “straightforward” Texas Triangle example may become tougher to divide up as the population boom continues in the region.

Here’s how the population of U.S. countries changed over the past decade:

Further complicating matters is the rapid move to remote work and distributed teams. A key element of these census divisions are commuting ties. With work increasingly not bound by geographic limitations, it remains unclear how that trend will impact this type of statistical exercise in the future.