JJAG: Comparison between JJA Alpine temperature anomaly, the anomaly derived from the Burgundy GHD (Grape Harvest Date) and the ETA (European Temperature Anomaly, Mariani & Zavatti, 2017). The low pass filters of the series (green, red and blue respectively) have a 50, 20 and 30 years window width. The strong divergence from the beginning of LIA (1350 CE) to 1730 CE has been assigned to local climates by the authors (for one year, 1639 which appears milder in Burgundy than in the Alps) but the differences span for several decades and in Burgundy "kills" the coldest LIA period, so some more structured and stable situation must had been present or we need to assume that grape harvest date cannot be a proxy for temperature in cold periods. After 1730 (last phase of the LIA and the recovery from a cold climate) the smoothed curves are quite similar. It must be noted that the anomaly from GHD has a different base than JJA (April to August mean temperature of Dijon between 1960 and 1989 instead of 1901-2000); ETA has a 1961-1990 base period, so some difference could derive from the bases.
Spectra show some variety: GHD and ETA seem similar at long periods (e.g 300 yr) while JJA and ETA show the same for shorter periods (20-80 yr) also if some coincidence could appear casual.