Premium
Shifts in plant phenology: a look at the sensitivity of seasonal phenophases to temperature in Germany
Author(s) -
Wittich KlausPeter,
Liedtke Marco
Publication year - 2015
Publication title -
international journal of climatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.58
H-Index - 166
eISSN - 1097-0088
pISSN - 0899-8418
DOI - 10.1002/joc.4262
Subject(s) - phenology , climatology , climate change , environmental science , scale (ratio) , seasonality , physical geography , geography , atmospheric sciences , ecology , biology , geology , cartography
The study presented here examines the sensitivity of mid‐spring, mid‐summer and mid‐autumn phenology to effective temperature in Germany for two divergent spatio‐temporal scale arrangements: the regional long‐term (50 years) scale and the nationwide short‐term (annual) scale. The dates of leaf unfolding of common horse chestnut, picking of ripe gooseberries and leaf colouring of common oak were chosen as the seasonal markers. Each of these phenophases was observed nearly nationwide over the 50‐year period 1961–2010. The focal question of the study was: what is the effect of different space and time scales on phenological sensitivity to effective temperature? This question is rarely addressed in phenological studies. We found a distinct contrast in the phenological sensitivities for these scale arrangements: first, when plotting regional 50‐year averaged phenological occurrence dates against corresponding effective temperatures across space, the phenological responses are −1.5, −1.9 and 0.0 days K −1 for mid‐spring, mid‐summer and mid‐autumn phenology. This type of scale combination corresponds to the common climate‐map presentation illustrating spatial climate averages. The second scale arrangement was to regress annual phenological timings, averaged over all regions, against analogue temperatures. Here, we get −4.7, −6.2 and +2.2 days K −1 for the three seasonal phases. This case thus reflects the nationwide response to climate change. As reference, the basic dataset of annual and regional data (without further averaging) provides sensitivities of −2.9, −3.3 and +0.7 days K −1 for mid‐spring, mid‐summer and mid‐autumn phenology, respectively. One should be aware of such differences in sensitivity when different (time and spatial) domains are studied and conclusions are drawn. For example, the space‐for‐time approach applied in ecology uses spatial gradients of phenological responses to temperature as a substitute for phenological estimates along time gradients. Based on our datasets, we found that this approach is of limited use for making predictions.