IGOR POLYAKOV

Contact Information

Go Back To Home

Arctic Ocean variability derived from historical observations

Igor Polyakov¹, David Walsh¹, Igor Dmitrenko¹, Roger L. Colony¹, and Leonid A. Timokhov²

¹International Arctic Research Center, University of Alaska Fairbanks
²Arctic and Antarctic Research Institute

Details of this research may be found in our paper,
[download pdf - Polyakov et al, 2003]

Motivation

Significant changes in the Arctic Ocean water temperature and salinity were observed in recent decades and were discussed in numerous studies. Russian historical data (see, for example, [Gorshkov, 1980; Treshnikov, 1985]) were used in these studies as reference data to define thermohaline anomalies. However, recent measurements are generally based on "snapshot" CTD profiles, and reliable evaluation of anomalies requires an accurate knowledge not only of climatic means, but also of variability. The Environmental Working Group (EWG) [1997] Atlas is the only climatology providing measures of water temperature and salinity variability. In this paper we examine Russian historical data collected from the central Arctic Ocean during the 1970s, as well as a data archive from the Laptev Sea collected in the 1960-90s. The goal of the research was to estimate background variability derived from these historical observations for future comparison with recently observed anomalies.

Data

The data used in this study consist of temperature and salinity measurements from Russian winter surveys of the central Arctic Ocean between 1973-79, and 40 years of summer and winter observations in the Laptev Sea (for geographical notations see Figure 1). Seven Russian winter aircraft surveys (1973-79) occupied 1034 oceanographic stations. These central Arctic Ocean surveys constitute the most complete set of arctic observations. The number of observations during these surveys was approximately the same from year to year, with some variation in the total area of observation. Nansen bottles were typically used to take water samples and to measure water temperature at standard levels. An extensive archive of Laptev Sea temperature and salinity, extending from 1960-90, was produced by the Arctic and Antarctic Research Institute. These data are complemented with data from joint Russian-German summer (1993-95 and 1998-2000) and winter (1996, 1999) expeditions.

Variability from historical observations

Figure 2 shows vertical profiles of water temperature measured in winter 1977 at two locations (60km apart) in the Eurasian Basin near Spitsbergen. The deviation of the temperature in 1977 from the EWG climatology, also shown in the figure, is striking: in the Atlantic Layer (AL) core (150-300m), the difference exceeds 1^oC, exceeding the σ given in EWG Atlas by a factor of 3. These AL Core Temperature (ALCT) anomalies are of the same magnitude as the anomalies found in the 1990s. Strong positive and negative ALCT anomalies measured in 1977 were also found at several nearby locations, suggesting the spatial variability apparent in Figure 2 is not a result of measurement error.

We speculate that the reason for the “suppressed” variability given by EWG is the multi-stage quality control procedure used. At the first stage this procedure eliminates data with anomalies exceeding three standard deviations (σ); at the next stage means and σ's re-computed for the reduced data, and anomalies exceeding two σ are eliminated. Data reduced for the second time is used to calculate means and standard deviations given by the EWG. This approach may lead to a substantial underestimate of σ, thereby reducing the utility of the dataset.

These considerations led us to further analyze Arctic Ocean variability apparent in the data. Figure 3 demonstrates ALCT and sea surface salinity (SSS) variability in the 1970s. The largest ALCT variability is seen close to Fram Strait, diminishing along the Nansen Basin slope towards the Laptev Sea (Figures 3). The σ and ALCT anomalies are maximum in the Eurasian Basin, while for the Canadian Basin historical data from the 1970s show smaller variations (Figure 4). Figure 3 also demonstrates that the Laptev-East Siberian seas shelf is a source of strong SSS variations. These SSS anomalies are advected eastward to the East Siberian Sea and northward to the Amundsen Basin interior roughly following the Lomonosov Ridge (Figure 3, left bottom).

Discussion

Substantial changes have occurred in the Arctic region over the last few decades. Shifts in atmospheric circulation patterns have resulted in increased transport and temperature of Atlantic waters entering the Arctic via Fram Strait [Rudels et al., 2000]. This intrusion of warm water from the North Atlantic found its way to the Arctic Ocean interior along the continental margins of the Eurasian and Canadian basins. According to these observations, the Atlantic water was nearly 1^oC warmer compared with climatological data [Quadfasel et al., 1991; Carmack et al., 1995; see also SCICEX website]. Salinification in the Eurasian Basin [Steele and Boyd, 1998] and freshening by as much as several salinity units in the Canadian Basin [McPhee et al., 1998] were observed.

Most of these estimates are based on synoptic measurements, hence reliable anomaly estimates require knowledge of the means and standard deviations derived from historical data. We found that these data allow the maximum temperature of the AL to be computed quite accurately. The EWG climatology provides statistical estimates of variability, but our analysis shows that the EWG dataset considerably underestimates variability. Using Russian historical data we calculated basic statistical parameters for selected Arctic Ocean regions. The estimates of s demonstrate a high level of ALCT variability in the Nansen Basin, and SSS fluctuations on the Siberian shelf spreading into the Amundsen Basin. If we accept these statistical estimates, they place strong constraints on our ability to define long-term means, and hence the magnitudes of ALCT and SSS anomalies computed using synoptic measurements from the 1990s referenced to means from climatologies.

References

Carmack, E. C., R. W. Macdonald, R. G. Perkin, F. A. McLaughlin, and R. J. Pearson, Evidence for warming of Atlantic water in the southern Canadian Basin of the Arctic Ocean: Results from the Larsen-93 expedition, Geophys. Res. Lett., 22, 1061-1064, 1995.

Environmental Working Group (EWG), Joint U.S. - Russian Atlas of the Arctic Ocean [CD-ROM], Natl. Snow and Ice Data Cent., Boulder, Colorado, 1997.

Gorshkov, S. G., Atlas of Oceans. Arctic Ocean. (in Russian). Moscow, Military Defense Publishing House, 199 pp, 1980.

McPhee, M. G., T. P. Stanton, J. H. Morison, and D. G. Martinson, Freshening of the upper ocean in the Arctic: Is perennial sea ice disappearing? Geophys. Res. Lett., 25, 1729-1732, 1998.

Polyakov, I., D. Walsh, I. Dmitrenko, R. L. Colony, and L. A. Timokhov, Arctic Ocean variability derived from historical observations, Geophys. Res. Lett., 30 (6) : 1298(doi:10.1029/2002GL016441, 2003) [download pdf]

Quadfasel, D. A., A. Sy, D. Wells, and A. Tunik, Warming in the Arctic, Nature, 350, 385, 1991.

Rudels, B., R. Meyer, E. Fahrbach, V. Ivanov, S. Osterhus, D. Quadfasel, U. Schauer, V. Tverberg, R. A. Woodgate, Water mass distribution in Fram Strait and over the Yermak Plateau in summer 1997, Annales Geophysicae, 18, 687-705, 2000.

Steele, M., and T. Boyd, Retreat of the cold halocline layer in the Arctic Ocean, J. Geophys. Res., 103, 10419-10435, 1998.

Treshnikov, A. F., Atlas of the Arctic, Moscow, 204 pp., 1985. (in Russian).

Figure 1. Map of the Arctic Ocean. Labeled regions are used for analysis of the water temperature and salinity from the 1970s. Dashed lines show the Laptev Sea region used for analysis of data from the 1950-90s.

Figure 2. Vertical profiles of water temperature measured in 1977 in the Eurasian Basin northward Spitsbergen (blue) and from EWG climatology [EWG, 1997]. Note the large 1977 temperature difference (>1.5oC) in the core of the Atlantic Water (150-300m).

Figure 3. Surface salinity (left) and AL core temperature (oC, right) from Russian surveys in the 1970s. Top: climatological mean fields. Bottom: difference (sign is omitted) between instantaneous values and climatological means.

Figure 4. AL core temperature (T, oC) and surface salinity (S) means (top), their s (second top), maxima (third top), and minima (bottom) for 13 regions of the Arctic Ocean (depth>300m) shown in Figure 1. Numbers above the top panels denote the number of measurements used for analysis.

Last modified: May 06, 2004. 14:47:39 pm