Xi'an joins the timekeepers
A cloud of strontium atoms, cooled to within millionths of a degree of absolute zero and suspended in a lattice of intersecting laser beams, oscillates 429 trillion times per second inside a laboratory at the National Time Service Center in Xi’an, Shaanxi. On February 11, 2026, that particular oscillation was formally recognized by the Bureau International des Poids et Mesures as precise enough to help define what time it is for the entire planet.
International Atomic Time — TAI, for its French initials — is not a single clock. It is a monthly average, computed by the BIPM in Versailles, of more than 400 continuously running clocks at around 80 laboratories worldwide. Each lab ships clock readings to Paris; Paris runs the ensemble through a weighted algorithm to produce a stable, cumulative time scale running unbroken since January 1, 1958. The result is published in a document called Circular T, which is the closest thing humanity has to an official answer to the question “what time is it, objectively?”
Not all 400 clocks are treated equally. A handful are designated secondary frequency standards — approved as precise enough to actively steer the ensemble, trimming its drift. For most of TAI’s existence, those steering clocks were cesium fountains clustered in a small group of Western and Japanese laboratories. In 2013, France’s LNE-SYRTE became the first to steer TAI with an optical clock rather than cesium. Japan’s NICT-Sr1 followed in December 2018, the first strontium lattice clock recognized for the role.
Circular T 457, released February 11, 2026, added two new names. The strontium optical lattice clock NTSC-Sr2, built by the National Time Service Center of the Chinese Academy of Sciences, was used to steer TAI — the first time a Chinese optical clock had done so. In the same bulletin, the National Institute of Metrology’s NIM-Sr1, built in Beijing, was recognized as a separately approved secondary frequency standard. China had held a seat at the TAI table since its cesium fountain NIM5 was approved in 2014; this was different in kind.
The precision involved is difficult to scale. NTSC-SrII achieves a systematic uncertainty of 2×10⁻¹⁸ — meaning run for the age of the universe, it would drift by less than a tenth of a second. Cesium fountain clocks operate at roughly 10⁻¹⁶, two orders of magnitude less precise. The difference is not incremental; it is categorical.
The timing matters. BIPM has been building the case for redefining the SI second using optical standards rather than cesium — a redefinition expected no earlier than 2030. The nations whose clocks are already recognized by BIPM will have contributed the comparative data that makes it possible. The geography of global timekeeping is reordering itself, one approved frequency standard at a time.
Time has always been political. Church bells told medieval Europe when to pray, and who controlled the bells controlled the day. Greenwich became the prime meridian not because it sat at any natural zero but because Britain had the largest merchant fleet and the most precise marine chronometers. TAI was built when a handful of post-war Western nations ran the world’s most advanced atomic laboratories. The arrival of NTSC-Sr2 and NIM-Sr1 in Circular T 457 does not overturn that arrangement, but it changes it: two more hands on the world’s most precise clock.
Sources
- NTSC CAS announcement, February 2026 — official statement that NTSC-Sr2 steered TAI for the first time, via BIPM Circular T 457.
- NIM announcement — NIM-Sr1 approved as secondary frequency standard for steering international standard time.
- NICT Japan press release, 2019 — Japan’s NICT-Sr1 beginning TAI calibration in December 2018, for historical context on the timeline of optical clock adoption.
- People’s Daily, February 2026 — report on China’s NIM-Sr1 contribution to International Atomic Time calibration.
- BIPM FAQ on the redefinition of the second — context on how secondary frequency standards steer TAI and the roadmap toward an optical definition.