Jump to Main Content
Digitizing the plant phenological dataset (1750–1875) from collections of Professor Adolf Moberg: Towards the development of historical climate records
- Holopainen, Jari, Helama, Samuli, Väre, Henry
- Agricultural and forest meteorology 2018 v.253-254 pp. 141-150
- Betula, autumn, barley, climate change, data collection, growing season, latitude, leaves, oats, paleoclimatology, phenology, rye, sowing, spring, summer, temperature, Finland
- Long records of phenological observations are commonly used as data in global change and palaeoclimate research and to analyse plants' responses to climatic changes. Here we delve into the historical archives of plant phenological observations (1750–1875) compiled and published previously by Professor Adolf Moberg (Imperial Alexander University of Finland). The digitized dataset represents 44,487 observations of 450 different plant species for their 15 different phenological phases made in 193 sites across Finland, and results in 662 different phenological variables. The five most frequently observed variables are the blooming of rye, the sowing of barley, the blooming of bird cherry, the leaf outbreak of birch, and the sowing of oat. The spring and summer observations demonstrate positive relationships between the onset date and the site latitude, this relationship becoming negative for observations made in the autumn. This latitudinal effect is evident in the raw data as demonstrated by the temporal correlations between the unadjusted mean phenological records and the mean latitude of the sites. After the latitudinal effect is removed from the original data such correlations are much reduced and the new set of phenological records based on the adjusted dates can be computed. The resulting mean phenological records correlate negatively and statistically significantly with the mean temperatures from April through July. Linear trends indicate (i) summer onsets having become delayed by more than one week over the full period and (ii) shortening of the growing seasons since 1846. The dataset is made available in an open repository.