 |
| Titel |
A probabilistic approach to attribute warming to changes in atmospheric circulation |
| VerfasserIn |
Irene Brox Nilsen, James Howard Stagge, Lena Merete Tallaksen |
| Konferenz |
EGU General Assembly 2016
|
| Medientyp |
Artikel
|
| Sprache |
en
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 18 (2016) |
| Datensatznummer |
250129458
|
| Publikation (Nr.) |
 EGU/EGU2016-9575.pdf |
|
|
|
|
|
| Zusammenfassung |
| Europe has been warming over the past decades, especially in southern Europe in summer
and northern Europe in winter. To understand the causes of regional warming, it is common
to separate the temperature change signal into changes in atmospheric circulation (or
dynamic causes) and other factors, so-called within-type changes (or thermodynamic causes).
For example, increasing temperatures due to greenhouse gases may alter the position and
strength of the polar jet stream, thus causing a change in the atmospheric circulation signal.
On the other hand, warming may be entirely independent of circulation, occurring as a
general increase in surface temperature. With the aim to detect regions and time of the
year in Europe in which recent warming can either be explained by changes in
atmospheric circulation or by within-type changes, we suggest a novel probabilistic
approach to calculate the circulation-induced trend, the part of the temperature
trend that is induced by changes in atmospheric circulation. Through the use of
resampling, in combination with a Monte Carlo test, we assessed the likelihood that the
observed temperature trend can be explained entirely by changes in atmospheric
circulation frequency. The temperature data originate from the gridded 0.5∘ Watch
Forcing Data Era-Interim (WFDEI), and cover the period 1981–2010. The SynopVis
Grosswetterlagen catalogue of circulation types was used to detect circulation-induced trends
in the same time period. We analysed trends on the monthly time scale to reveal
short-term responses, such as those related to snow or greening. The most wide-spread
observed warming was found in summer, with up to 60% of the European land area
experiencing a significant warming trend during August, most notable in eastern and
northern Europe. In spring and autumn, the percent area with significant temperature
increases reduced to 10–30%. In December and January, only 5% of the land area
experienced significant warming, most pronounced in northern Scandinavia. The
probabilistic approach revealed that changes in atmospheric circulation could not
account for all the observed warming. Regions where the observed trend likely
can be explained entirely by changes in atmospheric circulation include western
Europe in May, eastern Europe in August and Scandinavia in September. In most of
the regions and months experiencing significant trends, however, warming must
be caused by other factors as well; such within-type changes potentially driven
by feedbacks between the land surface and atmosphere. For two cases where the
magnitude of the temperature trend was strongest, northern Scandinavia in December
and in the Black Sea region in November, circulation types warmed over time.
This confirms the role of within-type change, i.e., that circulation types changed
their properties over time. This study provides an important contribution toward
improved understanding of the causes of synoptic-scale temperature change in Europe,
more specifically, the relative role of circulation-induced changes and within-type
changes. |
|
|
|
|
|
|