Learning from the past (Photo: Ina Frings)
Reconstructing past regional climate on longer time scales
Recent anthropogenic climate change is embedded in a background of natural climate and environmental variability. Disentangling the purely anthropogenic trends requires a realistic understanding of past environmental variations and of their mechanisms. The instrumental record is, however, too short to contain the full range of these variations. Together with long-term Climate Model Simulations and the analysis of indirect archives, we aim at reconstructing the natural climate and environmental patterns in the past and estimate their impact in the next few decades. This activity has led to a series of publications in high-impact journals.
These indirect environmental and climate archives comprise biophysical systems, like tree-rings, lake sediments, speleothems corals, but also historical and administrative records of the past centuries. Processing these records to extract the true climate signal, and discard the non-climate noise, is necessary since not all the variations seen in these records are directly related to climate or environmental conditions. This requires sophisticated and innovative statistical techniques that make use of all the information contained in the archives, sometimes also combining natural archives and model simulations, and produce a consistent picture of past climate and environmental conditions.
HZG has developed and/or tested statistical methods to reconstruct several climate aspects over the whole or part of past millennium based on combinations of natural archives and long instrumental records of selected meteorological variables.
Fig. 1 A selection of the multi-proxy network used in our research for hydrological and marine reconstructions, including a newly developed Aeolian proxy at the Baltic Coast from Ludwig et al., 2017 and Bierstedt et al., 2017)
The Atlantic Multidecadal Variability (AMV), a multidecadal quasi-oscillation of the sea-surface-temperature in the , could be reconstructed for the last 1200 years, based on a collection of terrestrial archives under the assumption that the AMV also leaves an imprint on air temperatures in North American and Europe (Wang et al., 2017) . The method used is a multivariate regression technique that links an AMV index with its terrestrial predictors. The method was also tested under controlled conditions using pseudo-proxies in an ensemble of climate simulations over the past millennium, yielding reasonable reconstruction skills.
The reconstruction of the North Atlantic sea-surface temperature field (Pyrina et al., 2017a) is based on records of growth-layers of Arctica Islandica - a very long-lived bivalve mollusc- from a few locations in the Eastern North Atlantic. The statistical method included a variant of principal component regression, which was also satisfactorily tested using pseudo-proxies together with an ensemble of climate simulations. With the help of this test we could establish that the reconstructions are skillful in a wide area in the North Atlantic, but not in the tropical areas.
The spatially resolved global temperature (Gómez-Navarro et al., 2017) was derived from the network of terrestrial archives compiled by the PAGES-2k initiative (HZG is also involved) and applying the analog method - an off-line paleodata assimilation method developed at HZG - to an ensemble of past millennium climate simulation.
Finally, the climate evolution in Northern Europe at very high spatial and temporal resolution, suitable to be used as a driver of Baltic Sea ecosystem models, was completed for the past 150 years from early long instrumental records of sea-level-pressure and temperature (Schenk 2015). The reconstruction was again based on the analog method, combining the instrumental records with high-resolution simulations with regional climate models. This strategy will be pursued in the future, and by including natural archives and historical information we aim at extending this reconstructions to the last 400 years, which will include the Little Ice Age, a climate regime markedly different than the present, and that may represent a mirror image of the expected warming over the next few decades.
A general conclusion from these studies is that climate models tend to produce a simulated climate that is spatially too homogenous and too sensitive to some perturbations such as volcanic eruptions. On the other hand, the simulated internal variability (not related to external perturbations) is too weak compared to reconstructions (Wang et al., 2017).
Fig. 2 The annually resolved AMV reconstruction (Wang et al. 2017)
However, in some cases, climate reconstructions and model simulations agree reasonably well in representing long-range links between remote areas, mediated by standing atmospheric planetary waves. An important example that illustrates the relevance of understanding the mechanisms behind large-scale anomalous periods is the climatic period during the 1st millennium AD, during what is known as the Antique Little Ice Age, showing long-distance relationships between Central Asia and Europe (Büntgen et al., 2016). The frequency of these reconstructed extremes (pronounced anomalous periods) may be relevant for past societies (Lemmen and Wirtz 2014), and for land use-carbon feedback to the atmosphere (Ruddiman et al., 2016).
The relationship between temperature and precipitation, and its relevance for climate impacts, is a critical aspect of climate variations and future climate change impacts. In contrast to future climate simulations, which indicate a general intensification of the mean hydroclimatic gradients, the past millennium hydroclimate reconstructions do not show a pronounced link to past regional temperatures (Ljungqvist et al., 2016) nor an intensification of hydroclimate extremes during the 20th century.
Fig. 3 Derived hydroclimate anomalies relative to the average for the millennium selected. Above: values from proxy data. Below: values from climate simulations. (from Ljungqvist et al., 2016)
European summer precipitation is mainly driven by the position of the North Atlantic storm tracks and its position appears to be independent of external natural climate forcings – for example, volcanic eruptions and solar variations (Gagen et al., 2016). These results prompted us to initiate a more detailed analysis of the link between wind extremes and the mean climate state in northwestern Europe. In several global and regional models, this frequency is not strongly related either to the mean regional temperature nor to major atmospheric variability patterns, such as the North Atlantic Oscillation (Bierstedt et al., 2016). For wind extremes, the different parametrizations of the boundary layer dynamics, including deforestation, are more important, with clear implications for future projections of extremes.
Bierstedt, S. E., Hünicke, B., Zorita, E., and Ludwig, J. (2017): A wind proxy based on migrating dunes at the Baltic Coast: statistical analysis of the link between wind conditions and sand movement. Earth System Dynamics, 8, 639-652, doi:10.5194/esd-8-639-2017
Bierstedt, S.E., Hünicke, B., Zorita, E., Wagner, S. and Gomez-Navarro, J.J. (2016): Variability of daily winter wind speed distribution over Northern Europe during the past millennium in regional and global climate simulations. Climate of the Past 12, 317-338, doi:10.5194/cp-12-317-2016
Büntgen, U., Myglan, V.S., Ljungqvist, F.C., McCormick, M., Di Cosmo, N., Sigl, M., Jungclaus, J., Wagner, S., Krusic, P.J., Esper, J., Kaplan, J.O., de Vaan, M.A.C., Luterbacher, J., Wacker, L., Tegel, W., Kirdyanov, A.V. (2016): Cooling and societal change during the Late Antique Little Ice Age from 536 to around 660 AD. Nature Geoscience 9, 231–236, doi:10.1038/ngeo2652
Gagen, M. H.; Zorita, E.; McCarroll, D.; Zahn, M.; Young, G. H. F.; Robertson, I. (2016): North Atlantic summer storm tracks over Europe dominated by internal variability over the past millennium. In: Nature Geosciences 9 (8), S. 630–635. doi:10.1038/ngeo2752
Gómez-Navarro, J.J., Zorita, E., Raible, C. and Neukom, R. (2017): Pseudo-proxy tests of the analogue method to reconstruct spatially resolved global temperature during the Common Era. Climate of the Past 13, 629-648, doi:10.5194/cp-13-629-2017
Lemmen, C., and Wirtz, K.W. (2014): On the sensitivity of the simulated European Neolithic transition to climate extremes, Journal of Archaeological Science 51, 65–72, doi:10.1016/j.jas.2012.10.023
Ljungqvist, F. C., P. J. Krusic, H. S. Sundqvist, E. Zorita, G. Brattstrom, and D. Frank (2016): Northern Hemisphere hydroclimate variability over the past twelve centuries, Nature 532(7597), 94–98, doi:10.1038/nature17418
Ludwig, J. , Lindhorst. S., Betzler C., Bierstedt, S., Borówka, R. (2017): Sedimentary rhythms in coastal dunes as a record of intra-annual changes in wind climate. Aeolian Research 27: 67-77, doi:10.1016/j.aeolia.2017.06.003
Pyrina, M., Wagner, S., and Zorita, E. (2017): Pseudo-proxy evaluation of climate field reconstruction methods of North Atlantic climate based on an annually resolved marine proxy network, Climate of the past 13, 1339-1354, doi:10.5194/cp-13-1339-2017
Ruddiman, W. F., D. Q. Fuller, J. E. Kutzbach, P. C. Tzedakis, J. O. Kaplan, E. C. Ellis, S. J. Vavrus, C. N. Roberts, R. Fyfe, F. He, C. Lemmen, and J. Woodbridge (2016): Late Holocene climate, Natural or anthropogenic?, Rev. Geophys. 54 (1), 93–118, doi:10.1002/2015RG000503
Schenk, F. (2015): The analog -method as statistical upscaling tool for meteorological field reconstructions over Northern Europe since 1850. Dissertation, University of Hamburg.
Wang, J., Yang, B., Ljungqvist, F.C., Luterbacher, J., Osborn, T.J., Briffa, K.R., and Zorita, E. (2017): Internal and external forcing of multidecadal Atlantic climate variability over the past 1200 years. Nature Geoscience 10, 512-517, doi:10.1038/ngeo2962
Paleoclimate – Learning from the Past
Historical atmospheric reconstruction