HOLOCENE ENVIRONMENTAL RECORDS FROM THE NE BARENTS SEA REGION: CLUES TO CLIMATE FORCING ON MULTI-CENTURY-TO-MILLENIAL TIMESCALES IN THE EURASIAN HIGH ARCTIC
LUBINSKI, DAVID J. INSTAAR.
Although recent reconstructions have greatly increased our knowledge of the spatial and temporal variability of Holocene Arctic climate change (eg., Overpeck et al., 1997; CAPE Project members, 2001), much work remains to understand the causes of that variability. This study reviews marine and terrestrial Holocene paleoenvironmental data (published and unpublished) in the NE Barents Sea region, which provide important clues to the mix of forcing factors operating in the Eurasian high Arctic on multi-century-to-millennial timescales. In particular, these data attest to the importance of changes in insolation, Atlantic inflow, eustatic sea level, glacial-isostatic uplift, and atmospheric circulation. Additional factors may have influenced the Eurasian high Arctic but are not readily studied with existing data (e.g., Siberian river discharge, sea ice-albedo feedback, vegetation-albedo feedback on mainland Eurasia, residual ice sheets in arctic Canada, high frequency NAO forcing). To better understand all of these factors, new data must be collected and carefully compared to other regions as well as to new climate model simulations.
The NE Barents Sea region is well situated to help increase our understanding of a host of Holocene forcing factors. It is presently bisected by the summer sea-ice limit, making it specifically useful for studying changes in sea ice cover. It also lies along a track of cyclonic activity that brings warm and moist North Atlantic air masses deep into the Eurasian Arctic during prolonged positive phases of the NAO (North Atlantic Oscillation), the strongest mode of North Atlantic climate variability. This connection to the North Atlantic is demonstrated by periodic covariations between the NAO and a number of environmental measures in both the northern and southern Barents Sea (i.e., glacier mass balance, winter temperature, winter precipitation, sea surface temperature; e.g., Zeeberg, 2001). Furthermore, the NE Barents Sea is strongly connected to the Arctic Ocean and the Russian mainland coast through ocean currents, including those that carry driftwood from rivers ranging from the Dvina (White Sea) to the Lena (Laptev Sea).
In the past decade, a number of terrestrial and marine paleoenvironmental records have been produced in the NE Barents Sea region. These records include those from Franz Josef Land and northern Novaya Zemlya as well as the adjacent NE Barents shelf and the deep St. Anna and Franz Victoria troughs. The entire NE Barents region was glaciated during the Last Glacial Maximum out to the shelf edge. Deglaciation probably began c. 15 14C ka in the outer marine troughs and was completed throughout the marine areas by c. 10 14C ka.
A number of climate forcings - some interconnected - appear to have influenced the Holocene paleoenvironmental record from this region:
* Summer insolation. July insolation at 80N was 11% higher than present
at 10 14C ka (48 watts/m2 higher), making it one of the largest anomalies for
any latitude during the past 100,000 years. Since no solar radiation reached these
high northern latitudes in winter, the seasonality of insolation was also higher.
Both the seasonality and summer anomaly slowly decreased during the Holocene.
Many available proxies are sensitive to summer solar radiation as well as its
influence on air temperature. For example, glacier extent records from Franz Josef
Land, northern Novaya Zemlya, and Svalbard show glaciers behind present limits
soon after deglaciation, retraction until at least c. 4 14C ka, and a series of
subsequent Neoglacial advances (e.g., Svendsen and Mangerud, 1997; Lubinski et
al., 1999; Zeeberg, 2001). Moreover, several marine proxies (dinoflagellate cysts,
benthic foraminifera, Bowhead whalebones?) suggest less-than-present sea ice cover
during the early Holocene. This result is consistent with the much thinner and
less extensive summer sea ice cover simulated by climate models using the early
Holocene insolation anomaly (e.g. TEMPO, 1996).
* Atlantic inflows. Records from the Nordic Seas and Svalbard show higher-than-present
northward transports of Atlantic Water during the early Holocene (e.g., Salvigsen
et al., 1992; Koc et al., 1993). The extension of these transports into the subsurface
of the Arctic Ocean was recently verified with NE Barents Sea records of foraminiferal
species and isotopic data (Duplessy et al., 2001; Lubinski et al., 2001). Although
early Holocene Atlantic transports were high, they do not appear to have entered
the channels of Franz Josef Land (Lubinski et al., 1998). Recently published dinoflagellate
cyst records from the SE Barents Sea suggest that smaller recurrent Atlantic water
variations may occur on a 1-1.5 kyr timescale after 5 ka and possibly earlier
(Voronina et al 2001). Variations on an even shorter timescale are indirectly
suggested by a high resolution glacial marine record on Northern Novaya Zemlya
spanning the past 800 years (Zeeberg, 2001).
* Eustatic sea level and glacial-isostatic uplift. Ice sheet coverage of
the NE Barents Sea resulted in glacial-isostatically induced deepening by up to
150 m, which probably influenced Atlantic Water flows into the northern Barents
Sea (Lubinski et al, 2001). Foraminiferal stable isotopes show that Atlantic water
transiting the Barents Sea probably entered the Franz Victoria Trough from the
south during the early Holocene, a path that is blocked today by a shallow sill.
Atlantic water flows to the NE Barents Sea may also have been influenced by bathymetric
changes elsewhere in the Arctic (i.e. Nares Strait and other E. Canadian Arctic
Channels, Bering Strait). Lowered sea levels on the unglaciated shallow Kara Sea
and Laptev Sea shelves during the early Holocene led to more northerly river mouths
in the Kara Sea and may have disrupted the Laptev Sea polynya, an important sea-ice
factory and formation of dense water on the shelf. The dense water helps maintain
the halocline/pycnocline in the central Arctic Ocean as well as the NE Barents
Sea. Reduced dense water formation during the early Holocene may have decreased
hydrographic stability, allowing heat exchange between Atlantic intermediate water
and the overlying fresher water layer (and reducing sea ice cover).
* Atmospheric circulation. The Iceland Low, the strongest synoptic feature
influencing the NE Barents Sea, has changed its strength and position during the
Holocene. It appears to have been stronger and displaced to the north during the
early Holocene due to high summer insolation (e.g., Harrison et al., 1992). The
effect of this change on the NE Barents Sea can be inferred from meteorological
data and models of extreme modern climate states (Serreze et al, 1993; Rogers
and Mosley-Thompson, 1995; Proshutinksy and Johnson, 1997). The studies imply
increased storminess and advection of heat from fall through spring, and, possibly,
during summer. Wind and ocean circulation patterns were undoubtedly altered by
the more easterly extension of the low pressure trough into the Eurasian Arctic.
A potential record of this extension is the paucity of driftwood in the NE Barents
Sea until c. 7 14C ka (Lubinski et al., in prep). Expected atmospheric pressure
patterns would cause easterly winds near the Eurasian coasts and transport of
much driftwood alongshore (where it could be trapped) rather than into the open
sea. Data constraining Holocene driftwood provenance after c. 7 14C ka (genera
ratios) suggest at least two recurrent modes of transport and circulation.
There remain several important climate forcing factors for the NE Barents Sea which are not easily studied with existing data. For example, discharge reconstructions for the large Russian Arctic rivers remain rare or controversial despite the influence of their large riverine freshwater fluxes on regional hydrography, sea ice cover, etc. Also needing study are the long-distance effects of the residual Holocene ice sheets in arctic Canada and two important albedo feedbacks. A strong sea ice-albedo feedback is especially important for the timing for fall freeze up while a vegetation-albedo feedback on mainland Eurasia may have a strong influence on spring sea ice conditions far offshore (e.g., Foley et al., 1994;TEMPO, 1996).
Future efforts to improve our understanding of climate forcing factors in the Eurasian high Arctic should focus both on data and simulations. Particularly needed are data-based Arctic river discharge reconstructions and long, high resolution paleoenvironmental records. The ice caps of Franz Josef Land and northern Novaya Zemlya have not been fully exploited and new long ice cores will be invaluable. Paleoenvironmental studies of the last millennium in the NE Barents Sea (when the NAO signal is most easily detected) will also help improve interpretations of earlier portions of the Holocene. Climate simulations should be run to refine our understanding of the mix of forcing factors and physical processes. For example, there are no simulations of the effects of lowered sea level and subsequent flooding of the shallow unglaciated Russian shelves on the arctic pycnocline, sea ice, and circulation. Similarly the influence of glacial isostasy on bathymetry and its impact on ocean circulation has seldom been modeled.
Acknowledgements: This poster will show data from many sources including a number of collaborative projects that Lubinski has done and is doing with Leonid Polyak (Ohio State University), Steven Forman (Univ of Illinois at Chicago), Sergey Korsun (Shirshov Institute of Oceanology), Jaap Jan Zeeberg (Univ of Illinois at Chicago), Stein Johansen (Univ of Sciences and Technology, Trondheim), and Anne de Vernal (GEOTOP, Univ of Quebec at Montreal).
References: All references listed here (and more) will be detailed on the poster.
Previous | Abstract Index | Next
Copyright © 2001 INSTAAR, Univ. of Colorado