Akshay DineshA day feels simple because the Sun rises, crosses the sky, and sets. A second feels simple because clocks count it evenly. The trouble is that those two kinds of time are not quite the same. Earth does not rotate with perfect smoothness, while atomic clocks are built to keep seconds with astonishing regularity. Leap seconds were invented to keep those two worlds close together, adding an extra second to Coordinated Universal Time when Earth time and atomic time drift too far apart.
For decades, that occasional extra second seemed like a tidy compromise. Astronomers could keep civil time tied to the turning planet, while scientists, governments, broadcasters, and ordinary clocks could use a shared global standard. But the modern world now depends on time in places most people never see: satellite navigation, financial records, power systems, phone networks, database logs, and internet servers. In those systems, a single unusual second can be more than a curiosity. It can be a technical problem.
Two kinds of time are being kept at once
The ordinary idea of a day comes from Earth’s rotation. If the planet made one perfectly steady turn every 24 hours, timekeeping would be easier. But Earth’s rotation changes slightly because of tides, the motion of the atmosphere and oceans, shifts inside the planet, and the long-term tug between Earth and the Moon. These changes are small, usually measured in milliseconds, but precision timekeeping notices them.
Atomic time starts from a different principle. Instead of watching the sky, it uses the steady behavior of atoms. The official SI second is defined using the microwave frequency associated with cesium-133 atoms. That kind of measurement is so regular that atomic clocks can keep time far more evenly than Earth’s rotation can. International Atomic Time, or TAI, follows that atomic rhythm without trying to bend itself around the planet’s uneven spin.
Coordinated Universal Time, or UTC, sits between those systems. It ticks at the same rate as atomic time, but it has historically been adjusted by whole seconds so it stays close to UT1, the time scale based on Earth’s actual rotation. NIST describes the purpose of leap seconds as keeping UTC within 0.9 second of UT1. That is why a leap second is not like daylight saving time or a time-zone change. It is a correction between the smooth clock of atoms and the slightly irregular clock of Earth.
What actually happens during a leap second
When a positive leap second is inserted, UTC does something clocks almost never do: it shows 23:59:60. The sequence at the end of the UTC day becomes 23:59:59, then 23:59:60, then 00:00:00. Leap seconds are normally scheduled at the end of June or December, after the International Earth Rotation and Reference Systems Service decides whether the correction is needed.
The first leap second was added on June 30, 1972. Since then, all leap seconds have been positive, meaning they added one second rather than removing one. The most recent one was added at the end of December 2016. The U.S. Naval Observatory’s leap-second notice for 2026 says no leap second was introduced at the end of June 2026, which continues the long pause since 2016.
That long pause can make leap seconds sound unimportant, but it actually shows the deeper issue. Earth’s rotation has not behaved in a simple, predictable pattern. For much of the leap-second era, Earth was generally running a little slow compared with atomic time, so extra seconds were added. More recently, Earth has sometimes rotated a little faster than expected. That raises the possibility of a negative leap second someday, where one second would be skipped instead of repeated. The official timekeeping community has noted that such a step has never been used, and many systems were never designed with it in mind.
Why one second can disturb digital systems
For a person, one second is barely enough time to blink and notice a clock. For computers, one second can contain millions or billions of operations. Many systems also assume that time moves forward in a clean sequence. Logs are sorted by timestamp. Transactions are ordered by time. Network messages expire after precise intervals. Satellites and receivers use timing to calculate distance. If a clock repeats a second, skips a second, or handles the adjustment differently from another clock, the mismatch can ripple outward.
Different organizations have dealt with leap seconds in different ways. Some systems insert the extra second directly. Others use a technique often called a leap smear, spreading the adjustment over a longer period so the clock never shows 23:59:60. That can make a system run smoothly by itself, but it also means not every system agrees on the exact time during the smear window. For ordinary browsing, that may not matter. For high-precision networks, scientific instruments, trading systems, and satellite navigation, inconsistent time can be a real engineering concern.
The International Bureau of Weights and Measures, or BIPM, summarized this problem in the 2022 Resolution 4 of the General Conference on Weights and Measures. The resolution noted that leap seconds create discontinuities that risk malfunctions in critical digital infrastructure, including global navigation satellite systems, telecommunications, and energy transmission. It also pointed out that different methods for introducing leap seconds have not followed a single agreed standard. The problem is not just that a leap second exists. The problem is that a leap second has to pass through a world of systems that do not all digest it the same way.
The 2035 plan changes the compromise
In 2022, the General Conference on Weights and Measures decided that the allowed difference between UTC and UT1 should be increased in or before 2035. In plain language, the world’s main timekeeping authorities agreed to move toward a future in which UTC can continue without frequent leap-second adjustments for at least a very long stretch. The exact details still require coordination with groups such as the International Telecommunication Union, but the direction is clear: make UTC more continuous and less vulnerable to one-second interruptions.
That does not mean civil time will suddenly drift away from the Sun in any noticeable way. The difference being discussed is tiny on human scales. If leap seconds stop being inserted frequently, the gap between clock time and Earth-rotation time would grow slowly. Noon would not suddenly arrive in the dark. The practical aim is to avoid unpredictable jumps while preserving a managed relationship between civil time and Earth’s rotation over the long term.
The change also reflects how much timekeeping has changed since 1972. In the early leap-second era, many people encountered official time through radio signals, clocks, calendars, and broadcast schedules. Today, time is embedded in global positioning, encryption certificates, cloud services, sensor networks, scientific data, and automated control systems. A rule that made sense when digital infrastructure was simpler can become awkward when nearly every part of society depends on synchronized machines.
Why Earth time still matters
If leap seconds are troublesome, it may seem tempting to ignore Earth’s rotation altogether. But the link between time and the sky still matters. Astronomers need to know exactly how Earth is oriented in space. Spacecraft navigation, telescope pointing, geodesy, and some kinds of surveying depend on precise Earth-orientation data. UT1 is not an old-fashioned leftover; it remains a scientific description of the turning planet.
The difference is that not every clock in everyday life has to carry that adjustment as a sudden one-second jump. A future without frequent leap seconds would still allow specialists to track UT1 and publish the correction between Earth-rotation time and UTC. The relationship would be measured and managed, but the world’s main civil time scale would not be interrupted as often by an inserted or skipped second.
This is a common pattern in measurement. One system may serve everyday coordination, while another preserves the deeper physical detail. A road map does not need to show every pebble to be useful, but surveyors still need more precise tools. In the same way, civil time can be stable enough for networks and ordinary clocks while scientific timekeeping keeps careful watch on Earth’s irregular motion.
A tiny correction with a large lesson
Leap seconds are small, but they reveal a surprisingly large truth about modern life: time is both natural and engineered. It comes from the rotation of Earth, the position of the Sun, and the physical behavior of atoms. It also comes from agreements among laboratories, standards bodies, governments, and the engineers who keep networks synchronized. A clock on a wall hides that whole arrangement behind two hands or a glowing screen.
The move away from frequent leap seconds is not a rejection of astronomy. It is a recognition that the world now needs a time scale that behaves continuously for the machines that depend on it, while still keeping track of the planet that gave time its original meaning. The extra second at 23:59:60 was an elegant solution for one era. The next era may need a quieter kind of correction, one that keeps the sky in view without making the world’s clocks stumble over a single second.
Add comment