Sixty degrees for 140 years: Luxembourg's must weight since 1839

Journal No. 5 measured a rise of twenty degrees Oechsle in the average must weight of Luxembourg's harvest between the 1980s and the ten harvests to 2025. It could not say whether the 1980s were themselves unusual, because STATEC's current file begins in 1980. Euro-Climhist, the historical climate database at the University of Bern, holds a longer series that starts in 1839. It files the series under Elbling at Remich. For 1839 to 1988 it is drawn from the viticulture chapter of STATEC's Statistiques historiques, and where it overlaps STATEC's current national file, the nine harvests 1980 to 1988 and the single years 1938, 1960 and 1970, the two agree to within one degree. We therefore read them as the same measurement and join them, using the Euro-Climhist series up to 1979 and STATEC's current file from 1980. One year, 1880, has no reading. From 1989 Euro-Climhist switches to the Remich wine institute's own figures for Elbling alone, which ran on average 4.4 degrees below the national average across 1989 to 2000; we keep those twelve years out of the joined line.

The joined line is flat for a very long time. The thirty harvests from 1839 to 1868 averaged 58.3 degrees Oechsle; 1869 to 1898, 60.9; 1899 to 1928, 59.3; 1929 to 1958, 61.2; and the twenty-one harvests from 1959 to 1979, 61.6. Across all 140 harvests before 1980 the average was 60.2. Single years ranged widely, from 40 in 1879 to 84.5 in 1865, but the stretches of thirty years barely moved. The 1980s, at 61.0, sat on the same line.

Then the line leaves it. The ten harvests from 1990 averaged 67.5, from 2000 76.4, from 2010 78.9, and the ten to 2025 81.5. Before 1980, four of 140 harvests reached 76 degrees: 1846, 1865, 1868 and 1921. Between 2001 and 2025, twenty of 25 did, and in the ten harvests to 2025 all of them did. Two harvests before 1980 reached 80, in 1865 and 1921; eight have done so since 2001. The lowest reading since 2001 is 67, in 2001 itself, a figure that 34 of the 140 earlier harvests reached.

Our weather record for the Mosel grid point begins in 1950, so it covers the last thirty years of the flat stretch. The mean temperature from April to September averaged 13.76 °C in the 1950s, 13.61 in the 1960s and 13.44 in the 1970s, and 13.70 in the 1980s. In 2016 to 2025 it was 15.65. While the season stayed where it was, so did the must weight.

That earlier period also bears on the question Journal No. 5 left open. Over 1980 to 2025 a warmer season and a vineyard replanted away from Elbling and Rivaner moved together, and comparing each harvest with the one before gave about 3.8 degrees Oechsle for each degree Celsius. Before 1989 the vineyard changed much less: Elbling and Rivaner covered 73 per cent of it in 1975 and 64 per cent in 1988. Across the 38 consecutive pairs of harvests from 1950 to 1988, a season one degree warmer than the last went with a must weight 6.3 degrees higher, with a correlation of 0.77, and 31 of the 38 pairs moved in the same direction. The largest moves were all of this kind: 1977 was 2.12 degrees cooler than 1976 and the must weight fell 16; 1972 was 1.77 degrees cooler than 1971 and it fell 16; 1965 and 1960 were each nearly two degrees cooler than the year before and each fell 14; 1959 was 1.76 degrees warmer than 1958 and it rose 14.

Applied to the 1.95 degrees by which the season has warmed since the 1980s, the older rate would account for about 12 of the 20.5 degrees Oechsle gained, against about 7.5 at the rate measured since 1980. The two figures come from different vineyards and different decades, and these files cannot say which applies today or why they differ. Between them they put the share that goes with the warmer season at somewhere between a third and three-fifths of the rise. What the longer record adds without any calculation is the starting point: the 1980s were not a low, but the last decade of a level that had held since 1839.

Must weight 1839–1979 and the Elbling figures for 1989–2000: Euro-Climhist, wine must quality, Remich (Luxembourg), 1839–2000. Sources: STATEC: "Viticulture". In: Statistiques historiques 1839–1989. Luxembourg: éditpress 1990: 129–136; and Staatliches Weinbauinstitut Remich: Das Weinjahr [1966–2000] und seine Ernteergebnisse (ed. Ministère de l'Agriculture, de l'Alimentation et de la Viticulture). Remich 1966–2000. Data: Euro-Climhist, University of Bern, https://www.euroclimhist.unibe.ch. Must weight 1980–2025 and vineyard area by variety: STATEC, LUSTAT, Indicateurs viticoles (tables D2200 and D2201), CC0. Weather: Contains modified Copernicus Climate Change Service information 2026.