1976 would be an ordinary year now

In the weather record Barrelheads keeps — ERA5-Land daily data for twenty-nine northern-hemisphere wine regions, reaching back to 1950 — the summer of 1976 stands where living memory puts it. Measured by growing degree days above 10 °C, summed from 1 April to 31 October, it was the hottest growing season of the three decades from 1961 to 1990 in seven of the twenty-nine regions: Champagne, Chablis, the Côte de Beaune, the Côte de Nuits, the Mâconnais, the Rheingau and the Mosel. The heat of that year sat squarely over north-eastern France and the German river valleys. Further south the benchmark season of those decades falls elsewhere: 1989 on both banks of Bordeaux, in Sauternes and in the Rhône; 1961 in Piedmont, Tuscany and Roussillon.

What has happened since is simple to state. In Champagne, eight of the ten complete seasons from 2016 to 2025 were hotter than 1976. The same is true in Chablis, the Côte de Beaune and the Côte de Nuits. In the Mâconnais, the Rheingau and the Mosel, nine of the ten were hotter. The Météo-France station at Dijon-Longvic, measured the same way, agrees that 1976 was the hottest season of those three decades and puts the recent count at seven of ten rather than eight; at Bordeaux-Mérignac, where the mid-century benchmark is 1990 rather than the grid's 1989, the count is also seven of ten. A season that had no equal in thirty years of mid-century weather has been exceeded, in the regions where it set the benchmark, in eight or nine summers out of the last ten.

The pattern is not confined to the 1976 belt. Give every region its own benchmark — the hottest season it recorded between 1961 and 1990, whichever year that was — and count the recent seasons that beat it. Across the twenty-nine regions that makes 290 region-seasons, and 227 of them, close to four in five, were hotter than the region's old benchmark. In five regions — Piedmont, Rioja, Roussillon, Tuscany and Veneto — all ten of the last ten complete seasons were hotter. In seventeen of the twenty-nine, at least eight were. The lowest counts in the set are four of the last ten, in the Douro and the Willamette Valley: even where the shift is weakest, the old benchmark has fallen four times in a decade.

Rank 1976 in the full record and the same picture appears from the other direction. Our record now holds seventy-six complete seasons per region. In Champagne, 1976 stands fourteenth. In Chablis it is fifteenth; in the Côte de Beaune and the Côte de Nuits, sixteenth; in the Mosel, seventeenth; in the Rheingau, nineteenth; in the Mâconnais, twenty-second. In the Mâconnais the comparison can be put more plainly still: the 1976 season accumulated 1,321 growing degree days, and the 1991–2020 average for the same window is 1,359. The hottest season of the mid-century record sits below what the following thirty years treat as normal there.

The medians of the current decade tell the same story to the unit. The middle season of 2016–2025 in Champagne accumulated 1,362 growing degree days against 1,219 in 1976; in the Mosel, 1,145 against 1,032. These are not freak years set against a freak year; they are the ordinary middle of the last decade, sitting well above the extreme of the earlier one.

The season now ending belongs in the same series, though it is not yet complete. Counted over the same calendar window — 1 April to 25 September, the latest date for which every region has data — 2026 is running ahead of 1976 in all twenty-nine regions.

None of this is a claim about wine. What the record shows is narrower and, we think, more useful: the season that defined heat for a generation of European growers no longer stands out in the series at all. A grower who planted vines in the spring of 1977 and worked a full career to 2016 retired just as seasons of that kind stopped being remarkable. The reference seasons have moved, and the measured record says exactly how far.