Phenological maps of Europe

Abstract

The geographical distribution of the timings of phenological phases is a precondition for detecting regional trends of the annual timings of phenological phases and finding their relationships to climate changes. Therefore phenological maps of Europe have been computed showing long-term means, trends and annual timings of extreme years. In this article maps of the beginning, the end and the length of the growing season as means over the years 1961-1998 as well as for the warm year 1990 are presented. Strong dependences on altitude, longitude and latitude were computed both for single phenological phases and the beginning and end of the growing season. The goodness of fit for the regression equation was between 32% for the end and 83% for the beginning of growing season. The results were consistent with those of similar investigations. KEY WORDS: Phenology · Maps · Growing season · GIS · Europe Resale or republication not permitted without written consent of the publisherClim Res 18: [249][250][251][252][253][254][255][256][257] 2001 37°longitude, from Ireland to Finland in the north and from Portugal to Macedonia in the south, the IPGs cover wide regions of Europe (Chmielewski 1996). From the establishment of the first garden in 1959 until 1998, about 60 000 phenological data measurements have been collected in up to 66 gardens.Before the phenological phases could be mapped, a validity check of all the observation data had to be done. All observations differing from the stations' means for more than 30 d were selected and checked. After the observers' remarks had been checked, the selected data were compared with the written records and with the same phenophases of the nearest stations and natural regions. In addition, the data were compared to the observations of the pre-and postphenophases as well as to similar terminated phenophases of the same stations.Gaps in the time series were filled by using linear regression models with 1 independent parameter (= regressor). Regressors were either the same phenophase of the nearest station, the same phenophase of the natural region (Chmielewski & Rötzer 2001), the pre-or post-phenophase at the same station or a similar terminated phenophase at the same station. For the calculation of a missing value the best regression model with regard to the coefficient of correlation was chosen. Missing values were only calculated if the correlation coefficient was higher than 0.70.The annual timing of phenophases is influenced mainly by air temperature. In spring the temperature across Europe decreases from south to north and from the maritime regions in West Europe to the continental regions in East Europe. With increasing altitude, the average air temperature decreases 0.65°C per 100 m. Thus altitude, latitude and longitude provide the possibility of mapping the beginning of phenological phases by using multiple regression analysis.The relationships between the annual timing of a phenological phase and altitude, latitude and longitude could be determined using the ob...

Geographical pattern in first bloom variability and its relation to temperature sensitivity in the USA and China

“…In this study, we found a 1.86 to 4.38 days delay in FBD per degree of latitude from south to north, which is consistent with Hopkins' law. Similar latitudinal patterns in spring phenophase were also found in Europe (Rötzer and Chmielewski 2001). Thus, strong dependences on latitude for spring plant phenophases exist in all continents of the Northern Hemisphere.…”

“…In the western parts of the Table 3 Shift of temperature sensitivity of first bloom date (FBD) with latitude (a x ), longitude (a y ), and altitude (a z ) n number of phenological observation stations; the regression coefficients (a x , a y , a z ) and constant (c) of multiple regression model are shown. R 2 : goodness of fit; **P<0.01, *P<0.05 continents, the impact of longitude is significant in Europe, where the FBD delayed from the maritime regions in West Europe to the continental regions in East Europe (Rötzer and Chmielewski 2001). However, the impact of longitude on FBD is not obvious in the Western US (Table 2), which may be because the mountain range in the Western US blocked the warm maritime airflow (Fig.…”

“…The geographical pattern of mean FBD was examined by using the method of multiple regression analysis (Rötzer and Chmielewski 2001):…”

“…Thus, the phenological variability includes two aspects: mean date and magnitude of variability. Geographical distribution in mean date of plant phenophases could reflect the dynamics of seasons across space (Rötzer and Chmielewski 2001). Hopkins' bioclimatic law firstly demonstrated the impact of geographical factors on phenology in temperate North America (Hopkins 1918).…”

“…Hopkins' bioclimatic law firstly demonstrated the impact of geographical factors on phenology in temperate North America (Hopkins 1918). The phenological maps of several typical plant species were also computed in China and Europe (Dai et al 2014;Rötzer and Chmielewski 2001). However, to date, the studies about geographical pattern in the magnitude of phenological variability are scarce.…”