Longterm Changes In River Runoff

To some extent, long-term changes in river runoff, which contribute to the formation of a system of surface currents in the Arctic Ocean and in the upper low-salinity water layer in the Arctic Basin, can influence fluctuations in sea ice area and distribution.

Zakharov (1996) compares the volume of continental runoff to the Arctic Ocean from Asia and North America for 1940-1968 (Anon. (N)) with the extent of sea ice in the North European Basin from 1946 to 1971 (mean annual data). The correlation coefficients of this relationship at time shifts of 3, 4, and 5 years (the sea ice extent after the runoff) were 0.33, 0.45, and 0.36, respectively (at a 95% significance level of 0.51). However, with 5-year running smoothing of the series and a shift of 2 years, the coefficient value increased to 0.82. We suggest that this relationship cannot provide a convincing argument in favor of the decisive role of continental runoff in sea ice extent changes because, as Zakharov (1996) indicates, the data were obtained using indirect methods. The anomalies of iceberg discharge were not taken into account, and the series compared are short (25 years). The continental runoff that was taken into account comprises only 42% of the freshwater flowing into the Arctic Ocean. A similar correlation of sea ice extent (for a longer period, 1940 to 1999) using observational data on runoff of the largest rivers to the seas of the Arctic Basin, which feed freshened Arctic water directly to the North European Basin, does not confirm this sea ice extent relationship to continental runoff. This author proposes a conceptual scheme of self-oscillation in the atmosphere-ocean-polar ice system that also provokes some strong objections (see Section 5.3).

Based on data in Ivanov (1976) and Zakharov (1996), the total continental runoff to the Arctic Ocean is 5135 km3/year. Ivanov et al. (2004) estimate continental runoff to the Russian Arctic Seas at approximately 2900 km3/year, including 2300 km3/year delivered by the large rivers flowing to the Eurasian seas. River runoff is non-uniformly distributed during the year: in summer (May to October), it comprises 84-85% (in the Barents and Kara Seas) to 99.5% (in the Chukchi Sea). But even in such rivers as the Yenisey with a significant part of their drainage area located outside the permafrost zone, almost half (45%) of the annual runoff is observed during the flooding that occurs in one month (June). Hence, runoff to the Arctic Seas mainly depends on the accumulation of solid precipitation in the winter. The intensity of this process depends on the speed of zonal transport in the troposphere of temperate latitudes that brings relatively warm and moist air from the North Atlantic.

The temporal changes in river runoff to the Eurasian Arctic Seas are characterized by the presence of relatively short-term cyclic fluctuations with durations of 3-5 years (the White, Barents, and Chukchi Seas), 5-6 years (the Laptev Sea), 6-8 years (the East Siberian Sea), and 8-12 years (the Kara Sea) (Ivanov et al., 2004). An analysis of multiyear changes in the annual runoff of large rivers to these seas for 1937-1999 (using data kindly provided by Ivanov et al. (2004)) showed the presence of a noticeable positive trend (except for the Kolyma, where a small negative trend is observed). The long-term changes in river runoff to the Arctic Seas throughout much of the twentieth century are shown in Figure 4.25.

As shown in Figure 4.25, the parameters of linear trends of annual runoff to the western seas (the Barents and Kara) and to the eastern seas (the Laptev and East Siberian) differ significantly. Of interest are typical changes in the river runoff trends in 1967 (western seas) and about 1973 (eastern seas). The trend parameters are shown in Table 4.8.

River runoff to the Eurasian shelf Arctic Seas increased significantly in the last third of the twentieth century (Figure 4.25 and Table 4.8). This phenomenon was caused by intensified west-to-east circulation in the atmosphere of middle and temperate latitudes in the Northern Hemisphere, which is associated with a corresponding decrease in atmospheric pressure at high Arctic Ocean latitudes. The North Atlantic Oscillation (NAO) index is discussed in Section 4.2 as a good indicator of the intensity of North Atlantic west winds; it also reflects a general planetary west-to-east transfer at temperate latitudes in the Northern Hemisphere (Smirnov et al., 1998).

Figure 4.25. Changes in the total annual runoff of the Severnaya Dvina, Pechora, Ob', and Yenisey Rivers (a) and the Lena, Yana, Indigirka, and Kolyma Rivers (b) from 1937 to 1999. Straight-line segments indicate linear trends for the typical time intervals.

Figure 4.25. Changes in the total annual runoff of the Severnaya Dvina, Pechora, Ob', and Yenisey Rivers (a) and the Lena, Yana, Indigirka, and Kolyma Rivers (b) from 1937 to 1999. Straight-line segments indicate linear trends for the typical time intervals.

Table 4.8. Linear trends (km3/year) in river runoff by region and time

Western Seas Region

Eastern Seas Region

Period, years

Trend

Period, years

Trend

1937-1967

-1.34

1937-1973

+0.24

1967-1999

+4.00

1973-1999

+1.10

Figure 4.26 shows fluctuations in the NAO index for the period 1937 to 1994. It depicts linear trend segments that approximately correspond to two time intervals considered above. The parameters were —0.057 and +0.122, which is in satisfactory agreement in sign with the changes in trend parameters of the river runoff volume to the western seas.

Figure 4.26.

Fluctuations in the winter North Atlantic Oscillation (NAO) index for the period 19371994.

Figure 4.26.

Fluctuations in the winter North Atlantic Oscillation (NAO) index for the period 19371994.

Increased west-to-east transports in the atmosphere of temperate latitudes at the end of the twentieth century are confirmed by our zonality index, which characterizes the intensity of these transports between 40° and 65°N, as shown in Figure 4.27, which is a display of part of Figure 4.9. Comparison of the two figures shows that the linear trends plotted in Figure 4.27 mainly represent branches of a cyclic fluctuation with a period of about 60 years. The elevated runoff and the increase in the indexes at the end of the twentieth century compared with their values in the 1930s-1940s are probably determined by cycles lasting more than 100 years.

Figure 4.27.

Changes in the average zonality index for October-March for the period 1930-1994.

Figure 4.27.

Changes in the average zonality index for October-March for the period 1930-1994.

Clear similarities in the trends of zonal transports in the atmosphere (Figures 4.26 and 4.27) and runoff of the Eurasian rivers to the Arctic Ocean (Figure 4.25) confirm the validity of the proposed hypothesis about the causes of climatic change, namely: [increased west-to-east transports in the moderate latitudes] ^ [increased precipitation over the drainage area] ^ [increased Siberian river runoff into the Arctic Basin].

Was this article helpful?

0 0

Post a comment