# Ten Winters Out of Sixty-Five > **Author**: Marius Comper > **Date**: 24 August 2026 > **Canonical edition**: [https://mariuscomper.uk/zece-ierni/](https://mariuscomper.uk/zece-ierni/) > **Romanian version**: [https://mariuscomper.uk/zece-ierni/](https://mariuscomper.uk/zece-ierni/) --- At Tulcea, someone logged the Danube's condition every winter, for one hundred seventy-nine years. Until around 1950, the river froze almost every year, as reliable as winter itself. Since then, across sixty-five winters, the ice has caught only ten times. ## Key figures - **179 years** — the Danube Commission's diary, kept at Tulcea, from 1837 to 2016. - **101 days** — the duration record, winter 1879–1880, the longest unbroken ice on record. - **10 of 65** — frozen winters since 1951, under 16% of winters, against almost all before 1950. - **+1.06 °C/century** — warming at Sulina, the strongest of three stations measured against the ice. ## One Hundred and One Days The record belongs to the winter of 1879–1880: ice held the river shut for one hundred and one days, from December deep into March. For people in the Delta, it was an annual condition, varying only in how many weeks the ice lasted. Households, carts, and people crossed the river on foot where only steamers ran in summer. The winter of 1879–1880 was simply the longest entry in an almost unbroken run that would not break for another seventy years. ## A Border That Held for a Millennium Long before the systematic diary, chronicles and old records trace the river's freezing back toward the year 850. Winters like 1621, 1669, 1755, 1779, and 1823 survived in documents precisely because of how far the ice reached: in the harshest of them, it extended along the Black Sea coast all the way to the Bosphorus. These are eyewitness accounts, kept in old documents. They show something simple: a freezing river was part of Romanian winters for centuries. ## The Threshold That Moved The diary itself opens in 1837 and closes, in the analysed record, in 2016: one hundred seventy-nine consecutive winters, measured against temperatures at three stations: Sulina, Bucharest, and Călărași. The correlation between winter mean temperature and ice duration is tight (r = -0.73, at 99% confidence), and all three stations show clear warming: Sulina, closest to Tulcea, warmed 1.06°C per century; Bucharest, 1.01°C; Călărași, 0.80°C, below the threshold of statistical significance. Somewhere around 1950, the series pivots abruptly from one regime into another, like a step on a graph: before, ice held nearly every winter; after, in fewer than one winter in six. ## What Didn't Move One detail breaks any tidy explanation. The freeze-up and break-up dates did not shift significantly across the whole record: the earliest freeze on record arrived on 6 December 1902, the latest thaw on 29 March 1929, both before the 1950 turn. If temperatures had climbed gradually, year after year, we'd expect the freezing window itself to keep narrowing, starting later and ending earlier with each passing decade. The diary shows no such thing. It is like a train that runs less and less often, but always from the same station, at the same times: the calendar window stayed put, and what changed is how many winters still got to use it. ## The Last Evidence, 2012 The most recent frozen winter documented publicly is 2012: ice blocked shipping traffic at Tulcea, and commercial vessels were halted on that stretch, while the ports of Galați and Brăila stayed partly open. It is one of the ten winters in the post-1951 diary, the only one whole generations born or raised since have gotten to see with their own eyes. The frozen river grandparents describe was, year after year, an almost certain state of affairs, until seventy-five years ago. Since then it has become a rarity that returns, on average, once in more than six winters. --- **Source and method.** The systematic ice diary (1837–2016) was kept at Tulcea by the Danube Commission and analysed by M. Ionita, C.-A. Bădăluță, P. Scholz, and S. Chelcea in "Vanishing river ice cover in the lower part of the Danube basin — signs of a changing climate," *Scientific Reports* 8, 7948 (2018), Alfred Wegener Institute and Romania's National Institute of Hydrology and Water Management. Entries before 1837 come from chronicles and historical sources compiled by the same authors and are treated separately, as historical testimony.