Time: We all have a sense of it, an innate feel for it. We see it and use it every day. If you’re like me, the first thing you do in the morning is check the time on your phone to see if you need to get out of bed or if you can close your eyes and catch a few more z’s.

Once you’re up and moving, time dictates when and what you eat, when you leave and come back to your house, and countless other activities.

Yet, if you’re like me (before I took a job at NIST), you might not have thought too much about where the time on your phone comes from. We tend to take time for granted.

During the past year, I’ve been working on a website about atomic clocks, one of NIST’s claims to fame and contributions to the world. This project has given me a lot of, ahem, time to think about time. And what I’ve realized is that pretty much everything I thought I knew about time, I didn’t really know.

Time, it turns out, is a far more profound, subtle, unstable—and fascinating—concept than I had ever appreciated.

Time is a human construct

Perhaps the most unsettling insight I’ve had while writing about clocks is that time, as we know it, is a human construct.

That isn’t to say that time has no meaning independent of our timepieces. There is clearly a physical and psychological reality to time. The universe expands. We remember the past, not the future. Rooms spontaneously become messier but never cleaner. Time appears in many of the fundamental equations that govern our universe.

But the time that our clocks display and the rate at which our clocks tick are determined by human decisions. Neither of these things reflects a fundamental truth about the universe.

Let’s try to sort that out a bit.

Read more: How do we know what time it is?

The time on our phones, computers, and smartwatches is produced by a global network of devices called atomic clocks. Some clocks—a wall clock, for example, or a quartz wristwatch—may not interact directly with this network. But chances are that when you need to set such a clock, you turn to your phone or smartwatch for the time. So, ultimately, just about every modern-day clock is referenced to atomic time. Whether we realize it or not, we are all immersed in an invisible web of time signals.

Harold Lyons (right) shows off the world’s first atomic clock to NIST director Edward Condon (left) and U.S. Department of Commerce Secretary Charles Sawyer. Condon is holding a model of the ammonia molecule that the clock was based on. The wall clock at the top served mainly to communicate that the device below was indeed a clock. Credit: AP

But where do these time signals come from? When I tell people I’m writing about the atomic clocks for NIST, they often respond with something like, “Oh yeah, NIST runs the atomic clock, doesn’t it?”

If only it were so simple. No single organization—not NIST, not the U.S. Naval Observatory (despite actually operating something it calls the “master clock”), not even the International Bureau of Weights and Measures—has a perfect atomic clock to rule all other clocks.

In fact, the whole notion doesn’t make sense once you think about it. Let’s say someone claims to have the perfect clock. How would we test it to ensure its ticks are spot-on? We’d have to compare it to some natural, completely unvarying frequency. But where would we find such a frequency?

For millennia, people used the rotation of the Earth to mark time, and our planet served as a sort of master clock. But as mechanical and, later, electronic clocks got better and better, it became clear that our planet is not a stable clock at all. The Earth’s rotation rate wobbles due to several factors, including the effects of tides, complex dynamics that govern the rotation of its core, and even the melting of polar ice due to climate change. (More on that later.)

NIST-F4 is a primary frequency standard that is currently under evaluation. This fountain is an upgrade of the NIST-F1 apparatus. Once the evaluation is completed, NIST-F4 will contribute to UTC and calibrate the absolute frequency of next-generation optical atomic clocks. Credit: NIST

Over time, human-made clocks became more stable than the planetary clock we had relied on for thousands of years. Humans also started inventing technologies such as intercontinental air travel and GPS, and those global systems required a highly accurate, universally agreed-upon way of telling time.

So, during the past century, human technology has supplanted the planet, moon, and stars as the source of time.

And since there is no perfect clock, people instead developed a system that involves running many very good clocks in labs around the world and comparing their frequencies at a central location: the International Bureau of Weights and Measures (BIPM), technically located in international territory outside Paris.

Everyone agrees to correct their local time to match BIPM’s coordinated universal time, or UTC. For example, NIST produces the time UTC (NIST), which rarely deviates from UTC by more than a few billionths of a second. This system allows us all, no matter where we are in the world, to share an understanding of what time it is.

But this global integration comes at a cost: If coordinated universal time were to falter, even for a moment, we would no longer know exactly what time it is. The costs to the world economy would be enormous. The need to keep global time as steady as possible is why NIST and other measurement (or metrology) labs have alerts on their atomic clocks, and scientists on call at all times in case a clock starts acting strangely.

As I’ve learned about atomic clocks, I’ve gained a far greater appreciation for everything that happens behind the scenes to make accurate time reliably appear on my phone every morning. Atomic timekeeping is an unceasing global symphony. BIPM is the conductor, and each country is a musician striving to play in perfect harmony with all the others.

From astronomy to atoms

This modern way of keeping time depends on pure, unchanging frequencies “hidden” inside atoms. If you’re interested in how this works, I encourage you to check out our new atomic clock website.

Read more: How do atomic clocks work?

In short, we determine the second using the cesium atom. Microwaves at a particular frequency cause an electron inside the cesium atom to make a jump between two quantum energy states.

The clocks that NIST and other metrology labs use to produce official time are essentially fancy devices built to measure the microwave frequency that’s most likely to trigger this quantum jump.

While the cesium resonant frequency is created by nature, the way we translate that frequency into ticks of time is not. People, not nature, have decided how long a second should be. The modern definition of the second goes back only to 1967, shortly after the cesium clock was invented.

Read more: A brief history of atomic time

Louis Essen at the U.K.’s National Physical Laboratory, who built the first practical cesium clock, needed to reference his new atom-based ticks of time to the old planet-based ones. So he and a colleague teamed up with William Markowitz, an astronomer at the U.S. Naval Observatory. Over nearly three years, Markowitz used a special camera to measure the astronomical second—equivalent to 1/31,536,000 of the year—as well as he could. Essen measured the number of microwave oscillations that elapsed in that amount of time. A few years later, their value became the basis of a new official definition of the second.

To me, this measurement represents one of the most profound events in thousands of years of timekeeping history: the handoff of time from astronomy to atoms. It affects nearly every person on Earth every day and will continue to do so for the foreseeable future—even if the second is redefined in terms of an atom other than cesium.

Yet neither Essen nor Markowitz (nor, for that matter, the creator of the original atomic clock, Harold Lyons) is a well-known figure. None of them won a Nobel Prize. I find this a bit mystifying.

NIST-7, the last of NIST’s cesium beam clocks. Credit: NIST

At the same time, the fact that the length of the international second was essentially determined by two scientists in the 1950s has certain implications. For one thing, atomic clocks have greatly improved since the 1950s. I suspect that if the Essen-Markowitz experiment were redone today, it would produce a slightly different second—though nowhere near different enough for any of us to notice.

Additionally, it inevitably led to a second that was not well matched to Earth’s rotation rate—which has slowed considerably in the intervening decades. This has resulted in 27 leap seconds being added to coordinated universal time since it was formally adopted in 1972. While the leap second has kept atomic midnight hitched to astronomical midnight, timekeepers view it as an awkward fix that has overstayed its welcome. This is the case, in part, because no one can predict more than a few months out when the next one will be needed. Tech companies also hate leap seconds because they mess with internet timekeeping systems and can cause outages.

The leap second is about to become more awkward because, probably due to changes in Earth’s core, the rotation rate of the Earth’s crust (which happens to be the layer of Earth that we live on) has sped up in recent years. This may create the need for a negative leap second in the next few years—something that has never happened.

To make matters even more complicated, several recent papers have suggested that so much water has melted from the poles due to global warming and accumulated around the equator that it has slightly slowed the rate at which the spinning of the Earth’s crust is accelerating. While often portrayed in the media as a disruption to timekeeping, this unexpected slowing may give scientists a few more years to grapple with the disruption of a negative leap second.

While my life (and hopefully your life) won’t be too severely upended by whatever happens with the leap second, I find the fact that no one knows when—or even if—the next one will occur to be a bit unsettling. Atomic clocks have revealed to us that the spinning planet, which once must have seemed a bedrock source of stability and predictability, is fickle and capricious—just like us!

An opportunity for awe

Here at NIST, we believe that more accurate measurements deliver benefits to humanity. They often do, but I think there is another facet we need to keep in mind.

Timekeeping using the stars and the Earth’s rotation is easy for people to understand and practice. Societies around the world and throughout history developed timekeeping systems based on astronomical events. These systems worked well when people did not generally travel or communicate over long distances. Our distant ancestors probably never imagined there would be a need for another way to tell time.

Atomic clocks, by contrast, make time an abstraction and place it in the hands of experts on whom the rest of us depend. It’s difficult for anyone without a deep background in quantum physics to fully understand how atomic clocks work. (Believe me—I’ve been working at it for more than a year.)

If leap seconds are abandoned, as countries have in principle agreed to do, the divorce between time and astronomy will become absolute. This will represent another profound and likely irreversible threshold in human history. Those of us alive today can at least take comfort that the drift between the atoms and the planet will probably not become noticeable in our lifetimes. But for future generations, it certainly will.

As we’ve gained accuracy and precision, we’ve lost something in terms of intuitiveness—an ability to instinctively understand time based on our sensory experience.

For perpetually time-challenged people like me, the increasing dominance of atomic time has also had a more practical consequence. When I was a child, if I showed up late to a class or appointment (as I often did), I could plausibly say that my watch was a few minutes slow. Nowadays, no one would believe me because our timepieces are all synced to within a few milliseconds.

Atomic clocks, of course, already existed when I was a child. What has changed is our ability to efficiently disseminate atomic time via GPS and the internet, to the point that none of us can escape its reach.

I think we should acknowledge and honor these losses, while at the same time celebrating the benefits of more accurate time measurement.

In that vein, I want to close by arguing that something has also been gained beyond the practical benefits of GPS, email time stamps, and so on. That something is an opportunity for awe.

The recognition of atomic frequencies and the technological realization of atomic timekeeping is a fascinating story of scientific discovery, intuition, and invention that spans a century and a half—and counting. It started with flashes of insight from some of history’s most brilliant minds. It progressed through decades of painstaking experiments.

And it has been pursued by passionate scientists with an almost fanatical devotion to pushing clock accuracy to yet another decimal point. Even those of us who will never know the thrill of making a precision measurement can appreciate this story and the benefits it brings us.

I believe the way time is measured and delivered to people all over the world is one of the most beautiful and inspiring examples of international technological collaboration in service of humanity. It’s comparable to how astronauts from various nations have long worked together peaceably and productively in space, even as their respective countries have engaged in bitter feuds on Earth.

At the end of the day, time binds and unites all of us. Whoever you are or wherever you’re from, we are all equal players in the unceasing drama of time.

Published Nov. 4, 2024, in the NIST Taking Measure blog.