Day-to-day temperature variability trends in 160-to 275-year-long Europeaninstrumental records

Citation
A. Moberg et al., Day-to-day temperature variability trends in 160-to 275-year-long Europeaninstrumental records, J GEO RES-A, 105(D18), 2000, pp. 22849-22868
Citations number
29
Categorie Soggetti
Earth Sciences
Volume
105
Issue
D18
Year of publication
2000
Pages
22849 - 22868
Database
ISI
SICI code
Abstract
Day-to-day temperature variability is investigated in eight European series of daily mean temperatures beginning between 1722 and 1833. Eight statisti cal measures of day-to-day temperature variability are compared. The intram onthly standard deviation of daily temperature anomalies is found to be a g ood measure. The absolute change in temperature anomaly from one day to the next is sensitive to changes in observational procedures and is suggested as a diagnostic tool for identification of inhomogeneities in instrumental temperature series. Because many changes in observational procedures have t aken place, quantitative estimates of trends in day-to-day variability, bas ed on all series, could only be calculated for 1880-1998. A trend analysis over this period indicates an increase by 5% in southwest Europe, 0 to -5% change in the northwest, and a decrease by 5 to 10% in northeast Europe. On a longer time perspective, day-to-day temperature variability in winter, s pring, and autumn in northern Europe has decreased over the last 200-250 ye ars. The frequency of extremely cold wister days in northern Europe was low er in the twentieth century than in the eighteenth and nineteenth centuries . Day-to-day temperature variability in winter in northern Europe was negat ively correlated with a North Atlantic Oscillation index in the period 1826 -1997, but some other factor must also have contributed to the long-term va riability decrease. More long daily temperature series, and development of homogenization methods for such data, are needed for an improved knowledge of long-term changes in day-to-day temperature variability.