Oceanography of the Arctic: Encyclopedia Arctica 7: Meteorology and Oceanography

CONTENTS

OCEANOGRAPHY OF THE ARCTIC

Introduction

The geographic boundaries of the various subdivisions of the Arctic
waters have been dealt with by the International Hydrographic Bureau in its
publication “Limits of Oceana and Seas,” Special Publication No. 23, second
edition, 1937. The proposed boundaries are shown in the map in Figure 1,
in which the different seas are indicated by numbers with the names stated
in the caption.

From as oceanographic point of view the proposed boundaries are satis–
factory, with the following exceptions:

In the case of 12a, the Laptev Sea, the northern boundary should rather
be selected as a curved line inside of the 1,000-meter contour, as indicated
by the heavy dashed line in the figure.

In the case of 12c, the Chukchi Sea, the southern boundary should not
follow the Arctic Circle, but should cut across the narrowest part of Bering
Strait. The International Hydrographic Bureau uses the spelling Chukchee,
but here the name will be spelled Chukchi in agreement with the practice
of the U.S. Navy Hydrographic Office.

The Beaufort Sea, 12d, is tram the oceanographic point of view part
of the North Polar Sea, 12g, and should be dealt with as such. This does not

Untitled/Unnumbered Page

Fig. 1. Bathymetric chart of the arctic seas (after Wüst) and limits of seas according to the
International Hydrographic Bureau, Special Publication No. 23, second edition, 1937.
8 Norwegian or Greenland Sea
9 Barents Sea
10 Kara Sea
11 White Sea
12a Laptev or Nordenskiöol Sea
12b East Siberian Sea
12c Chukchi Sea
12d Beaufort Sea
12e Northwest Passage
12f Hudson Bay (not on map)
12g Arctic Ocean on Polar Sea

Page 002

EA-Oc. Sverdrup: Oceanography

mean that any objection is raised against retaining the Beaufort Sea as a
geographic unit.

The boundaries of 12e, the Northwest Passage, and 12e’ (not shown in
Fig. 1), Davis Strait, are not acceptable from an oceanographic point of
view. It appears more logical to let the eastern boundary of the Northwest
Passage follow a line from Ellesmere Island to Baffin Island and to introduce
two subdivisions to the west of Greenland, Baffin Bay and Labrador Sea, sep–
arated by Davie Strait, Baffin Bay, used in this sense, should include Kane
Basin and Robeson Channel. It seems that even from a geographic point of
view these limits may be preferable.

Area 12f, Hudson Bay, is not shown in Figure 1, but in the chart that
accompanies Publication No. 23 of the International Hydrographic Office the
term Hudson Bay applies only to the bay proper, and Hudson Strait is taken
as part of area 12e, the Northwest Passage. From the oceanographic point of
view it appears better to consider Hudson Bay and Hudson Strait as one region
and let a line across Foxe Channel represent the northern boundary.

Area 12g is in the cited publication called the Arctic Ocean, but here
the name the North Polar Sea shall be used.

In the following discussion the oceanographic conditions within the
adjacent seas of the North Polar Sea shall be dealt with first, beginning
with the Labrador Sea-Baffin Bay and proceeding clockwise around the Polar
Sea to the Norwegian Sea. The reason for this procedure is that the waters
and currents of the Polar Sea are so much dependent upon the conditions in
a number of the adjacent seas and upon an exchange of water with these that
they must be known in order to arrive at an understanding of the conditions
in the Polar Sea.

Unfortunately, no uniform discussion can be presented of the ocean–
ography of all the arctic seas because some have been far less explored
than others and because for several regions the results of more recent
work have not been available to the author. It has not been attempted to
compile a complete list of the oceanographic literature that deals with the
arctic “waters, but a few references are given, several of which contain
long lists of literature.

The Labrador Sea

Limits . The Labrador Sea is bounded to the west by the coast of
Labrador and the southern part of Baffin Island and to the east by Greenland.
To the southwest it is limited by s line from Cape Farewell, Greenland, to
Cape St. Charles at the northern side of Belle Isle Strait and to the north
by a line across Davis Strait in 66° N.

Bathymetric Features . The bottom of the Labrador Sea has the character
of a trough, which is open to the southeast but partly close to the north
by a submarine ridge across Davis Strait. The sill depth (or threshold
depth) of this ridge is about 800 meters. To the south of the ridge the
depth increases rapidly to 2,000 m., and then more gradually to nearly
4,000 m. in the southeastern part of the sea. In a few places the depth
exceeds 4,000 m. The continental shelf is very narrow off the coast of
Greenland, but somewhat wider off Labrador. Thus, the distance to the
200-meter depth contour is about 50 km. off Greenland and about 100 km.
off Labrador.

The Labrador Sea is at all depths in open communication with the

Page 004

EA-Oc. Sverdrup: Oceanography

North Atlantic and might therefore be considered nothing but a bay of that
ocean, but the specific conditions prevailing there justify treating it
as an adjacent sea.

Bottom Sediments . The bottom sediments of the Labrador Sea are char–
acterized by the wide occurrence of rock fragments which have been transported
by floating ice and deposited when the ice melted. Otherwise the sediments
consist of sand, silt, and clay in varying proportio n s. Most of the rock
fragments and pebbles are readily recognized as coming from formations
on Greenland and Baffin Island, but the occurrence of limestone fragments
in Davis Strait seems difficult to explain. The organic content of the
sediments is low, about 1 per cent, as may be expected because of the pre–
ponderance of ice-borne deposits.

Salinity . Off the Greenland coast there is a narrow belt of water of
salinity 31 o/oo to 34 o/oo, but beyond this, off the continental shelf,
water of salinity close to 35 o/oo is encountered, representing Atlantic
water. Low salinity water (30 o/oo to 34 o/oo) is present on and partly
off the Labrador continental shelf. By far the greater mass of the waters
in the Labrador Sea is of salinity around 34.90 o/oo, the highest salinities,
34.92 o/oo to 34.94 o/oo, being found near the bottom. At the surface,
salinities of about 34.60 o/oo prevail in summer, but the seasonal varia–
tions are considerable and higher surface values may be found in winter.

Temperature . The coastal water off Greenland shows a considerable
annual variation of temperature which in summer rises to 5 to 6°C. and in
winter drops to freezing point. The Atlantic water off the shelf has a

Page 005

EA-Oc. Sverdrup: Oceanography

temperature of about 5° at depths below 150 to 200 meters. Off the
Labrador coast the waters are colder, and in midsummer temperatures
somewhat below −1°C. are found at a depth of 100 meters. The bulk of
the waters in the Labrador Sea has a temperature between 3.5°C. and 2°C.
the lowest values being found near the bottom.

Oxygen Content . The oxygen content of the water is generally high.
In the de e p and bottom water the amount is between 6.0 and 6.5 ml/L.

Currents . Off the Greenland coast the West Greenland Current flows
to the north, carrying water of the East Greenland Current (see the Nor–
wegian Sea), the salinity of which has been increased by admixture of At–
lantic water. Beyond the continental slope the water of salinity around
35.0 o/oo represents one of the last braches of currents carrying Atlantic
water, which has been considerably mixed with Arctic water from the East
Greenland Current.

Part of the water from the East Greenland Current continues into
Baffin Bay and part turns around to the south of Davis Strait and joins the
Labrador Current , which flows south off Baffin Island and Labrador. The
cold and low salinity waters of this current come mainly from Baffin Bay,
and in part they represent outflow through the sounds between the islands
to the west of Greenland. In the central part of the Labrador Sea numerous
eddies are present, but these are probably constantly shifting.

Formation of Deep and Bottom Water . The large and essentially uniform
body of water in the central part of the Labrador Sea is formed by the mixing
of Atlantic water which in part flows north off West Greenland and Arctic
water which in part flows around Cape Farewell, in part flows south from

Page 006

EA-Oc. Sverdrup: Oceanography

Baffin Bay. This mixed water has in some areas salinity between 34.88 o/oo
and 34.94 o/oo. When water of these salinities in winter is cooled to tem–
peratures between 3.5°C. and 2°C. it attains a sufficiently high density to
sink from the surface. In general the stratification of the deep water of
the Labrador Sea is stable, that is, the density increases with depth. Water
that sinks from the surface layers will then sink to the depth at which the
density with which it leaves the surface is encountered, and will spread out
at that depth. The high oxygen content of the deep wa t er supports the concept
about its formation. The depth to which sinking takes place by this process
can be expected to vary during the season and from season to season.

This process was first described by Fridtjof Nansen in order to account
for the formation of the deep water in the Norwegian Sea. There the sinking
wa t er is replaced by water that rises to the surface, but in the Labrador Sea
the deep and bottom water that is formed flows south into the Atlantic Ocean
where, after mixing with other water types, it can be traced to the Antarctic
Ocean. The average outflow of deep water from the Labrador Sea is estimated
at 2 million m ³ /sec.

Ice Conditions . Most of the eastern one-third of the Labrador Sea is
generally ice-free the year round except for a region along the coast of
Southwest Greenland. This coast is generally open from August to December,
but in the latter month the ice starts moving around Cape Farewell and spread–
ing to the north. It attains its greatest northern extent in April-May when
the coast is closed from Cape Farewell to south of Godthaab or to about lati–
tude 62° N. Between this latitude and the Arctic Circle the outer coast line
is always ice-free, but the fjords are frozen. The southeast coast of Baffin

Page 007

EA-Oc. Sverdrup: Oceanography

Island and the Labrador coast are generally completely ice-free in September.
From October on the ice belt off these coasts increases in width, the increase
beginning in the north. By the end of December it extends to the southern
limit of the Labrador Sea, in April-May it covers the western two-thirds of
the sea, and in June to August it gradually disappears.

Icebergs are always present both in the West Greenland and the Labrador cur–
rents. Many of the icebergs in ht th e former current have been carried around
Cape Farewell. A few of these are later on carried south by the Labrador Cur–
rent, together with icebergs from the glaciers on West Greenland. The greater
number of these, about 70 per cent, are probably derived from glaciers near
Disko Island and 20 per cent probably come from glaciers that terminate in
Melville Bay.

Baffin Bay (with Kane Basin and Robeson Channel )

Limits . Davis Strait to the south; the northern outlet of Robeson Channel
to the north; a lone across the sounds between Ellesmere Island, Devon Island,
Bylot Island, and Baffin Island to the west; Greenland to the east.

Bathymetric Features . Baffin Bay proper is a basin which to the south
is separated from the Labrador Sea by a submarine ridge across Davis Strait
with a sill depth of about 800 meters. The maximum depths in the sounds to
the north and the west are unknown, but are probably less than 600 m. The
continental shelf, with depths less than 200 m., is narrow all over. In the
central part of Baffin Bay the depth of a wide area exceeds 2,000 m., for
which reason one there finds a large body of water which is closed off from
its surroundings.

Salinity . Down to a depth of about 500 meters the waters in Baffin Bay
have salinities above 34.4 o/oo, but below that depth the salinity is between
34.4 o/oo and 34.5 o/oo, the highest values being found to the south. The
waters show an admixture of Labrador Sea water that flows in along the
coast of Greenland, but must mainly be characterized as Arctic water, the
surface layers of which have been diluted by excessive precipitation.

Temperature . Except for a submerged tongue which extends north from
Davis Strait, the temperatures in Baffin Bay are below 0°C., and in the
tongue they are only slightly above that value. At the greatest depth the
temperature is about −0.45°C. Thus, there exists a striking difference
between the water at a depth of about 2,000 meters in the Labrador Sea and
Baffin Bay; in the former the temperature and salinity are 3.0° and 34.9 o/oo,
respectively, and in the latter they are −0.45° and 34.5 o/oo.

Oxygen Content . The upper layers are, as usually, rich in oxygen. But
the deep water of Baffin Bay is deficient. Near the bottom the content is
about 3.5 ml/L.

Currents . There appears to be little known about the currents of Baffin
Bay. The fact that icebergs from the Disko Island area and from Melville Bay
are carried south by the Labrador Current suggests, however, that the surface
currents are part of a counterclockwise circulation.

Formation of Deep and Bottom Water . The deeper water in Baffin Bay must
be renewed in the bay itself, because water of this character cannot flow
into the basin from the north or the south. It seems probable that there
exist areas where in winter the salinity of the surface water is increased
so much by freezing of ice that it attains a sufficiently high density to

Page 009

EA-Oc. Sverdrup: Oceanography

start sinking. When sinking it is mixed with water of a somewhat higher
temperature of −0.45°C., which lies somewhat above freezing point. The low
oxygen content indicates that the renewal of the deep water is a slow or
intermittent process. The formation of deep and bottom water as a result
of increased salinity by freezing of ice plays a large part in the Antarctic,
particularly in the Weddell Sea, but, in the Arctic, Baffin Bay appears to
be the only region in which this process takes place.

Ice Conditions . The ice conditions which are most favorable to naviga–
tion are encountered in August and September. In these months the eastern
part of Baffin Bay is generally free from ice up to Etah, at the entrance to
Kane Basin, in latitude 78°20′ N. A fairly wide lead is found off Baffin
Island, but heavy ice masses are found in the eastern part of the bay between
latitudes 68° and 75° N. In October the ice cover rapidly increases in extent,
and from November to April the bay is completely ice-covered. A peculiar
development starts in April when an ice-free area appears to the south of
Etah. This area persists during May and June, and in ~~Jlly~~ July it joins the ice–
free region which in that month advances from the south along the west coast
of Greenland, north of Disko Island.

Numerous icebergs are discharged into Baffin Bay, mainly from the glaciers
in the vicinity of Disko Island, but also from glaciers at the head of Melville
Bay.

Hudson Bay and Hudson Strait

Limits . From the oceanographic point of view it seems rational to deal
with Hudson Bay and Hudson Strait as one unit that is limited to the northwest
by a line across Foxe Channel.

Bathymetric Features . The greater part of the bay has depths between
100 and 200 meters. Off the western shore of the bay the 100-meter depth
contour is reached at distances of 20 to 50 miles from the coast. The south–
eastern part, between Churchill and James Bay, is somewhat shallower, the
100-meter depth being reached at distances of 50 to 100 miles from the coast.
Off the eastern shore the depth decreases rapidly to values in excess of
100 meters but the bottom is very irregular and at some greater distance
numerous islands are found. In the central and northern parts of Hudson
Bay there is found an irregularly-shaped region with depth exceeding 200
meters. From this region a narrow trough with depths greater than 200 m.
extends into the western entrance of Hudson Strait but it is not known if
the trough continues uninterrupted to the deep e r parts of the strait. In
the eastern part of the strait the depths increase toward the mouth. A
maximum depth of 620 m. has been recorded at the eastern entrance. In Foxe
Channel the depths are less than 200 m.

Salinity . Observations of salinity and temperature of the water of
Hudson Bay are available from summer only. In this season the runoff from
the rivers that empty into the southern portions of the bay, particularly
into James Bay, is of great importance to the salinity of the upper layer.

In August 1931, the average salinity of the upper 10 meters of water
increases from SE to NW from a value of about ~~[ ]~~ 24 o/oo off James Bay to a value
of 30.5 o/oo in the northwestern area. Simultaneously the salinity in the
upper 10 m. increased from 28.2 o/oo at the western entrance of Hudson Strait
to 32.4 o/oo at the eastern entrance.

In Hudson Bay the salinity increases rapidly with depth in the upper 50 m.

Page 011

EA-Oc. Sverdrup: Oceanography

At a depth of 50 m. the average value of the salinity is 32.1 o/oo. Below
50 m. the salinity increases more slowly to an average value of 33.0 o/oo
at a depth of 150 m.

Winter values are not available, but it is probable that in winter the
salinity of the surface layer is increased somewhat because of the formation
of ice. It is, however, improbable that the winter values of the surface
layer in Hudson Bay proper approach the values at 50 m. or deeper, but in
the eastern part of Hudson Strait they can be expected to do so.

Temperature . At the end of August the surface temperatures in Hudson Bay
may lie between 7°C. and 10°C. but the temperature decrease rapidly with
depth and is all over negative at 25 meters. At and below 50 m. the available
observations show temperatures between −1.8° and −2.2° and the averages lie
about 0.1° below the freezing point as computed from the salinity. The
latter result may be due to errors in the observed temperatures, which have
been recorded to the nearest one-tenth of a degree only. Therefore, it
appears that all deeper parts of the bay are filled by water that has been
cooled to the freezing point. The salinity of this water equals that which
is found in the eastern part of Hudson Strait, for which reason it is probable
that the bottoms water in the bay is formed in winter by sinking of water in
Hudson Strait. It is not known if this process occurs annually such that
the bottom water is partly renewed every year or if it occurs at rare
intervals such that the bottom water is more or less stagnating.

At the eastern entrance to Hudson Strait the surface temperature in
August is only about 1°C. but t e mperatures slightly above freezing are found
to a depth of more than 200 m., indicating effective stirring by currents.

Currents . In summer the surface waters flow out of Hudson Bay and out
through Hudson Strait. In the bay the currents are weak and numerous eddies
are present. In the strait outflowing currents of a speed up to 3 knots
have been recorded.

Ice Conditions . Hudson Bay is nearly free of ice in June, but Hudson
Strait is rarely navigable prior to July and even then only by heavily
built ships. In August, September, and October both Hudson and Hudson
Strait are nearly ice-free, but from November through June the strait is
closed. In Hudson Bay belts of ice form along the shores, reaching their
widest extension in March, but the greater area of the bay remains open even
in midwinter, except for scattered fields of drifting ice.

The Northwest Passage

There exists practically no oceanographic information from this entire
region. The few available soundings indicate that the bottom is quite ir–
regular and that some of the sounds between the numerous islands are quite
deep. The features appear similar to those which are found in previously
glaciated regions.

Ice Conditions . Navigation in part of the area of the Northwest Passage
is in general possible in August and September only. Ice-free water is en–
countered nearly every year in Lancaster Sound, between Devon Island and
Baffin Island, and in the extreme southern part of the region, that is, in
Amundsen Gulf to the south of Banks Island and in the straits and sounds to
the south of Victoria Island and King William Island. The waters between
Devon Island and Banks Island (south of 75° N.) are navigable in some years,

Page 013

EA-Oc. Sverdrup: Oceanography

but in other years even large icebreakers cannot make any progress. The
northern regions to the west of Ellesmere Island are much more difficult
to penetrate by ship, and up to 1948 only one icebreaker had ever been able
to floow the west coast of Ellesmere Island to 80° N.

The Beaufort Sea

The oceanography of this region will be discussed when dealing with
the North Polar Sea, of which it is an integral part from the oceanographic
point of view.

The Chukchi Sea

Limits . The limits adopted by the International Hydrographic Bureau
are shown in Figure 1. According to these, the Arctic Circle represents the
southern limit of the sea, but from the oceanographic point of view it is
preferable to consider a line across Bering Strait, from Cape Dezhnev to
Cape Prince of Wales, as the southern limit. The northern limit corresponds
approximately to the known or assumed northern border of the continental
shelf.

Bathymetric Features . Over the greater part of the Chukchi Sea the
depth varies between 40 and 60 meters. Off the Siberian coast the depth
generally increases to 40 m. or more within a distance of less than 10 miles,
but off the Alaskan coast the depth increase more slowly, and the 40-meter
contour is met with at distances ranging from 30 to 60 miles from the coast.
The two large bays, Kolyuchin Bay on the Siberian coast and Kotzebue Sound
in Alaska, are both shallow, being less than 20 meters de p e p. In Bering Strait
the maximum depth is about 60 m.

The bottom shows some irregular features. About halfway between Wrangel
Island and Alaska lies Herald Shoal, with a minimum depth of 15 mm m., and due
east of Wrangel Island the small and steep Herald Island rises out of the
water. To the east of Herald Island a trough with depths exceeding 60 m.
and with some isolated de e per parts extends north-south and can be traced
from latitude 73° N. to 69° N. A small depression in latitude 68° N. is
perhaps connected with this trough.

Bottom Sediments . The bottom of Bering Strait is rocky and stony. In
the Chukchi Sea the bottom is generally muddy, with scattered pebbles and
larger stones which must have been carried to their position by ice rafting
(see chapter on ice).

Salinity . No large rivers empty directly into the Chukchi Sea, but
quantities of fresh or low-salinity water flow into the sea, especially in
summer, when water from the Yukon River runs north through Bering Strait, fol–
lowing the coast of Alaska. Because of this inflow and because of the summer
runoff from land, the surface waters of the Chukchi Sea are of low salinity.
In summer the average salinity between depths of 10 and 30 meters is about
30 o/oo along the coast of Alaska and about 32 o/oo west of the meridian
through Bering Strait. Where there is scattered ice the me l t water from the
ice may reduce the salinity much below these values above a depth of 10 m.,
and in lanes a less than 1-meter-thick layer may be nearly fresh. Where there
is open water and mixing has taken place, the salinity in the upper 10 meters
may be about 28 o/oo. The waters are often stratified, the salinity increasing
abruptly from one layer to another. Below a depth of about 40 m. higher
salinities are generally encountered, but rarely in excess of 33 o/oo.

No winter observations are available, but is analogy with condi–
tions in the East Siberian Sea one may expect that freezing leads to the
development of a well-mixed upper layer of constant salinity, and that
below this a sharp transition is found to a bottom layer of somewhat higher
salt content.

Temperature. In summer the surface temperature is ice-free waters
say rise to 6 or 8°C., but is scattered ice it generally lies close to 0°.
To the west of the meridian through Bering Strait the temperature at depths
of 10 to 30 m. lies between −1.5°C and −1.6°C., that is, only a little above
freezing point of water of salinity 32 o/oo, −1.68°C. Near the bottom the
water of salinity around 33 o/oo has generally a somewhat higher temperature,
ranging from −0.2°C. to −0.8°C, but water of temperature −1.7°C. may also
be present. Off the coast of Alaska and in the eastern part of Bering Strait
positive temperatures may be found all the way to the bottom, provided that
the salinity is less than 32 o/oo. All over, the variation of temperature
with depth is more or less irregular. No observations are available from
winter, but again analogy with conditions in the East Siberian Sea suggests
that in winter the temperature remains uniform at freezing point from the
surface to a depth of about 30 meters, below which depth somewhat higher
temperatures are found.

Density . The density depends mainly on the salinity, because at
temperatures close to freezing the density varies little with temperature
and because the salinity varies within wide limits. At depths of 10 to 30
meters the density is around 1.0255 except off the coast of Alaska where it
drops to less than 1.024. Close to the bottom it is about 1.0265 in the
western part of the sea.

Oxygen Content . In summer the oxygen content is high at all depths,
ranging from more than 9 ml/L at the surface (a little over 100% saturation)
to about 7 ml/L (about 80% saturation) near the bottom. Thus, the waters
are well ventilated throughout.

When freezing begins is the fall, it has beers observed that the oxygen
contests directly below the ice increase far above the saturation value.
This phenomenon can probably be ascribed to the presence of phytoplankton,
which, for some time continues assimilation of carbon dioxide accompanied by
production of oxygen, which, because of the ice cover, cannot escape to the
atmosphere. The supersaturation that results reaches a maximum about 2 weeks
after beginning of freezing and lasts for 5 or 6 weeks, indicating that the
phytoplankton gradually disappears.

Currents . It has already been mentioned that warm water of relatively
low salinity, which contains a large admixture of Yukon River water, flows
in through Baring Strait along the American coast. At the same time an outflow
often takes place along the Siberian coast, but it is probable that on an
average for the whole year there is inflow amounting to about 300,000 cubic
meters per second, or about 6,300 cubic miles (statute miles) per day.

The warn water continues to the north and northeast along the coast of
Alaska, where it can be traced to a distance of 80 to 100 miles from the shore.
It gradually loses its identity because it mixes with the waters to the west.

Except off the coast of Alaska, the currents in the southern part of the
Chukchi Sea appear to be variable, being go v erned by the variable winds. At
the northern boundary of the sea the winds are on an average directed to the
west, and the average drift takes place in that direction. ( Karluk , 1914).

Page 017

EA-Oc. Sverdrup: Oceanography

To the east of Wrangel Island there seems to be a tendency for the currents
to flow north within the region of the trough in the bottom ( Jeannette ,
1879; Maud , 1922).

Ice . Bering Strait generally becomes navigable at the end of June,
and along the Siberian coast a coastal lane mostly develops in early July.
Along the coast of Alaska the waters generally become navigable to Point
Hope in early July, but farther toward the northeast the ice clings to the
coast, and without icebreakers Point Barrow is seldom reached prior to the
middle of August. From the middle of July and through August the southern
part of the Chukchi Sea is generally more or less open, but strong northerly
winds can at any time carry the ice down to the coa s ts. In most years it is
possible to reach Wrangel Island by navigating through scattered ice, and
in some years ships have reached latitude 72° or more.

The ice that is encountered is the ordinary broken-up drift ice, com–
posed of rather small ice floes. The thickness of the flat ice floes is
frees 2 to 3 meters, but below pressure ridges the ice has been telescoped
so much that it extends to an average depth of 8 to 10 m. and in places to
15 and 20 m., leaving a space of only 20 to 25 m, between the lower portions
of the ice and the bottom.

The East Siberian Sea

Limits . The limits proposed by the International Hydrographic Bureau
are shown in Figure 1. No comments are needed.

Bathymetric Features . The East Siberian Sea is even shallower than the
Chukchi Sea and, except in the eastern part where depths exceeding 20 meters
are reached at a short distance from the coast, the depth increases very slowly

Page 018

EA-Oc. Sverdrup: Oceanography

when departing from land. The 40-meter depth contour is met with at distances
ranging from 180 to 300 miles from the mainland, and the 60-meter contour at
distances of 280 to 320 miles.

The New Siberian Islands and De Long Islands are located in the western
part of the sea. The former are not rocky, but large parts of them are formed
by masses of ice which are covered by a layer of stones and gravel, sand and
soil, that is only a few feet thick. These parts retreat rapidly because in
summer the ice melts wherever it is exposed.

Bottom Sediments . Bottom samples from the outer part of the shelf, in
latitudes 74° to 76° N., show very soft, dark mud, except to the north of the
New Siberian Islands, where the bottom is hard sand.

Salinity . Two large rivers, the Kolyma and the Indigirka, empty into
the East Siberian Sea and, in addition, part of the waters from the mighty
Lena River flow into the Sea through Laptev Strait between the mainland and
Fadeevski Island. Consequently waters of salinity below 15 o/oo are in summer
found to considerable distances from the coast. Off the mouths of the rivers
the waters are nearly fresh. Where the sea remains ice-covered in summer a
thin layer of brackish water, representing meltwater from the ice, ocvers all
openings in the ice.

In winter the salinity in the upper layers is increased by the freezing
of ice. At a distance of about 60 miles from the mouth of the Kolyma River,
the winter value of the salinity in the upper 10 meters was 26.7 o/oo. At a
distance of about 250 miles from the mainland, where the depth to the bottom
is 40 to 60 m., the salinity is in winter constant from the surface to a depth
of 30 or 40 m., having a value of about 29.6 o/oo. Closer to the New Siberian

Page 019

EA-Oc. Sverdrup: Oceanography

Islands the value drops to about 28.5 o/oo, probably because of greater ad–
mixture of Lena River water.

Near the bottom there is generally present water of a higher salinity.
Tim isohalines appear to follow the bottom contours; thus, the 34 o/oo
is o haline nearly coincides with the 40-meter isobath.

Temperature . In summer the surface temperature in open water near the
coast may rise to a few degrees above freezing, but is scattered ice it will
not exceed 0°C. At depths below 5 to 10 m. the temperature rises to about
0.1°C. above the freezing temperature, that is to a temperature of about
−1.50°C. In winter the temperature in the upper layer of constant salinity
is at freezing point. The freezing point of sea water depends no only upon
the salinity of the water, but also upon the pressure and is lowered by
0.0074°C. for every 10 m. of depth. The observations during the Maud ex–
pedition indicate that down to a depth of 20 m. the temperature of the water
is on an average equal to the computed freezing point, taking both salinity
and pressure into account. Thus, at a depth of 20 m. the freezing point com–
puted from salinity alone was −1.592°C., but, taking the effect of pressure
into account, the freezing point was −1.607°C. The average observed tempera–
ture was −1.604°C. This feature suggest that down to a depth of 20 m. the
water is in contact with ice, meaning that the ice masses that are accumulated
below pressure ridges must, at least in sense places, extend downward to a
depth of 20 m. It is probable that similar conditions exist all over the
Polar Sea.

The temperature of the relatively high salinity bottom water lies a few
tenths above freezing point and appears to increase slightly when advancing from

Page 020

EA-Oc. Sverdrup: Oceanography

the border of the shelf toward the coast. All differences are small and
over the entire area the temperature of the bottom probably varies between
−1.6° and −1.4°C. The slight increase in temperature may be caused by heat
that is generated by dissipation of tidal energy or by heat flowing through
the sea bottom from the interior of the earth.

Density . In summer the density of the upper layer increases slightly
with depth and in scattered ice the density of a very thin surface layer
appr ao oa ches that of fresh water. In winter the density remains practically
constant down to a depth of about 35 meters or less where the water is shallow.
At a distance of about 250 miles from the coast the water of the upper layer
has a den is si ty of about 1.0238, but close to the coast it is lower. It is
all over lower than the density of the corresponding layer in the Chukchi
Sea, because of greater admixture of river water.

The transition from the upper homogen e ous layer to the bottom water is
generally so sharp that the density increases from, say, 1.024 to 1.026 on
a vertical distance of less than 5 m. The density of the bottom water depends
upon its salinity and toward the northern border of the shelf, where the depth
exceed 60 m., it reaches values in excess of 1.027.

Oxygen content . The oxygen content of the upper layer remains high
throughout the year, but shows an annual variation ranging from 100% satura–
tion in September to about 90% in April. This variation is related to the
development of phytoplankton in summer, but this development appears to be
less marked in the ice-covered areas of the East Siberian Sea than in open
parts of the Chukchi Sea, as might be expected because the ice cover reduces
the amount of light at subsurface depths.

Where the layer of bottom water is quite thin, 5 m. or less, the bottom
water is very low in oxygen, probably because oxygen is being consumed by
the organisms living on the bottom. Where a th i ck bottom layer is present,
a minimum in the oxygen content generally appears at the top of the layer,
probably because of spreading of water whose oxygen content has been greatly
reduced by contact with the bottom.

It is of interest to observe that over the western part of the East
Siberian shelf there exists a marked relation between the temperature and
the oxygen content of the water that has a salinity of .32 to 33 o/oo and
occurs as a thin layer of bottom water over wide areas. On an average an
increase in temperature of 0.1°C. corresponds to a decrease in oxygen con–
~~[ ]~~ tent of 3 ml/L. This feature suggests that the water is being heated simul–
taneously with being deprived of oxygen by the respiration of bottom organisms.
The energy for heating can be supplied by dissipation of tidal energy or by
flow of heat from the earth’s interior.

Currents . The flow of the waters of the East Siberian Sea is so slow and
sluggish that the term “currents”' used to describe this flow may be misleading.
The distribution of salinity near the coast indicates that there the surface
waters tend to move to the east and away from the coast. Water with a strong
admixture of Lena River water flows east through Laptev Strait, and turns
gradually north and northwest, flowing around the New Siberian Islands and
joining the general drift toward west and northwest along the border of the
continental shelf. The waters from the Indigrika and Kolyma rivers also
appear first to flow east and then to turn north and west. There is no evi–
dence that any appreciable amount of these waters enters the Chukchi Sea.

The fresh water from the rivers is mixed with such large quantities of
sea water that the tipper mixed layer that flows out of the northwestern region
of the East Siberian Sea has a salinity of about 29 o/oo. The flow that has
been described carries salt out of the area, but there is no reason for be–
lieving that the salinity of the waters on the shelf is decreasing. There
must, therefore, be present a flow by which salt is carried into the region,
and this flow takes place along the bottom. Mixing takes place between the
different waters, as evident from the fact that the salinity of the north–
flowing surface layer increases with increasing distances from the coast,
whereas the salinity of the south-flowing bottom water decreases with increas–
ing distance from the continental border. It was mentioned that along the
continental border the average flow is directed toward west and northwest.
This flow is maintained by the prevailing winds.

The Wind Drift of the Ice . Except for the slow outflow of surface water
which has been diluted by river water and the slow creep toward the coast of
bottom water of higher salinities, the currents in the East Siberian Sea are
all maintained by the winds. The wind drift of the ice was first studied by
Fridtjof Nansen on the basis of the observations during the drift of the Fram
in 1893-96, and subsequently by H. U. Sverdrup and N. N. Zubov, who used the
observations during the drifts of the Maud, 1922-24, and the ~~Sedov~~ Sedov , 1937-40,
respectively. The most detailed discussions are those based on the records
from the Fram and the Maud , and since the latter drifted along the northern
part of the East Siberian Sea the subject to the wind drift can appropriately
be taken up here.

The analysis has in all cases been carried out by comparing the drift,
as determined by astronomical observations, with the resultant wind in the

Page 023

EA-Oc. Sverdrup: Oceanography

period between observations. Nansen found that over the greater part of
the Polar Sea there also existed a weak permanent current, the effect of
which had to be eliminated in order to find the effect of the wind alone.
The available account of the drift of the Sedov does not give any details
of the wind drift, but the general results are in complete agreement with
those obtained from the Fram and the Maud . The drift of the Maud was not
complicated by the presence of any permanent current and took place so far
from land that the drift was not distorted by the effects of coast. A more
detailed examination of the wind drift could, therefore, be based on the
results of the Maud expedition.

All observations show that the wind drift is, on an average, directed
to the right of the wind, and the speed is proportional to the wind speed
as measured 6 or 8 meters above the ice. According to the observations on
board the Fram , the Maud , and the Sedov , the angle of deflection and the
wind factor, that is, the ratio between drift speed and wind speed, have
the following average values:

Nansen ascribed the deflection to right to the effect of the rotation
of the earth. On his suggestion Ekman carried out a theoretical analysis
of the wind-driven currents in the open sea and, introducing certain simplify–
ing assumptions, arrived at the conclusion that there the surface current
deviates 45° from the wind. Subsequently Ekman’s assumptions have been mod–
ified, but in all cases the angle of deflection of the surface current in

Page 024

EA-Oc. Sverdrup: Oceanography

open water lies close to 45%. The smaller value that characterizes the
ice drift must be due to the resistance that arises when the ice is
pressed together. The detailed observations of the Maud expedition con–
firm this view. They show that the angle of deflection and the wind fac–
tor are subject to a pronounced annual variation, with maximum of both
values at the end of the summer in August-September, and minimum at the
end of the winter in April-May.

In summer, when the ice is relatively open and severe jamming of the
ice is rare, the angle of deflection approaches the theoretical value for
open water, 45°, and the wind factor is large, but at the end of the winter,
when the ice is tightly packed, both angle and wind factor are small. Roughly
the conditions can be accounted for by introducing an “ice resistance” which
is proportional to the speed of the ice drift. This factor of proportionality
which can be called coefficient of resistance, reaches a minimum in August
and a maximum in April. The summer values are in part substantiated by results
from the records of Andr e é e’s journey over the ice to the northeast of Spits–
bergen in 1897. These records were found on White Island in 1930. According

Page 025

EA-Oc. Sverdrup: Oceanography

to these somewhat uncertain data, the wind drift was, in the period July 26
to September 9, characterized by the values: angle of deflection 59°,
wind factor 2.76. These values are larger than those from the Ea s t Siberian
Sea and indicate, if correct, that in summer the ice to the northeast of
Spitsbergen is more open and more easily set in motion by the wind.

Ice Conditions . The greater part of the East Siberian Sea is
ice-covered all year. In summer open water is found off the mouth of the
large rivers and Laptev Strait, through which Lena River flows, is ice-free.
In general navigable lanes occur along the coast, but without the use of
ice - breakers the navigable season is short, lasting with interruptions from
the beginning of July to the middle of September. In the western part of
the sea icebreakers have been able to travel north and have reached the
northern coasts of the New Siberian Island.

In winter the water on the very shallow areas between Laptev Strait
and Bear Island probably freezes to the bottom. When melting takes place in
summer, the ice floes rise and carry with them quantities of mud, stones, and
shells that are frozen fast to the under side of the ice.

The Laptev Sea

Limits . The limits proposed by the International Hydrographic
Bureau are shown in Figure 1. From the oceanographic point of view the
deeper waters to the north should rather be considered part of the Polar Sea.

Bathymetric Features . The inner part of the Laptev Sea is very
shallow, with large areas of less than 20 meters depth. North of latitude
75° N. the bottom has the shape of a very gentle trough whose axis runs about

Page 026

EA-Oc. Sverdrup: Oceanography

north-south along the 124th east meridian. In the northwestern part of
the sea the shallow shelf is quite narrow, the 100-meter depth contour
being reached a few miles to the north of Cape Chelyuskin, whereas it
lies about 130 miles north of Kotelny, the western of the New Siberian
Islands. A few of the western New Siberian Islands and some small islands
off the Taimyr Peninsula are located in the Laptev Sea.

Salinity and Temperature . The Lena River and smaller rivers like
the Khatanga and Olenek carry large quantities of fresh water into the
Laptev Sea. The influence of the Khatanga and the Olenek does not appear
to extend to any great distance from the coast, but the Lena water exerts
a great influence over a wide area. Part of the Lena water flows through
Laptev Strait into the East Siberian Sea and part flows to the north along
the western coasts of the New Siberian Island, as evident from the rela–
tively low surface salinities (less than 20 o/oo) in these regions. Along
the northeastern coast of the Taimyr Peninsula much higher surface salin–
ities, ranging from 30 to 33 o/oo have been observed in the early autumn.

In summer the surface temperature off the Lena Delta may rise to 5°C.
of slightmore more, but these relatively high temperatures are limited to
the waters in the immediate vicinity of the coast. In winter the tempera–
tures are at freezing point.

Very little information is available about the vertical distribution
of salinity and temperature. It is probable that in winter the freezing
of ice leads to the development of a homogeneous upper layer below which
water of higher salinity exists. The Fram observations in the northern
part of the Laptev Sea demonstrated this stratification.

Currents . There probably exists a counterclockwise circulation in
the Laptev Sea, the surface waters being carried to the north off the
western coasts of the New Siberian Islands and to the south off the eastern
coast of the Taimyr Peninsula. Nansen arriv e d at this conclusion by exam–
ining the surface salinities. In the northeastern part of the sea the
drift is mainly directed to the north, according to the experiences of the
Fram and the Sedov . In addition to this horizontal circulation, water of
of relatively high salinity, 32 to 34 o/oo, probably creeps toward the south
in a thin layer near the bottom.

Ice . The southern and southwestern parts of the Laptev Sea become free
from ice in summer because of the influence of the large rivers. Along the
eastern coast of the Taimyr Peninsula the ice is broken, and leads are often
present. The season when navigation without the use of icebreakers is pos–
sible along all or part of the coast lasts from the end of June to the middle
of September.

The Kara Sea

Limits . The limits can be taken as those shown in Figure 1, with the
reservation that from the oceanographic point of view the northern limit may
have to be modified when the location of the continental slope can be
established. It has been suggested that from the oceanographic point of
view the Kara Sea should be dealt with as two regions, an eastern which is
shallow and where the waters are of relatively low salinity, and a western
where much greater depths are found and where the waters are of higher
salinity wad closely related to the waters of the Barents Sea.

Bathymetric Features . The eastern part of the Kara Sea is very shallow,
a depth of 50 meters being reached at distances of 60 to 150 miles from the
coast. Several small islands are found in the northeastern part of the sea.
In this western part an irregular, deep trough is found off the eastern coast
of Novaya Zemlya with depths exceeding 500 m. in latitude 74° N. The region
is which this deep is located, that is, the sea between Novaya Zemlya and
the Yamal Peninsula, is often referred to as the Kara Sea proper.

Salinity . Large amounts of fresh water are carried into the Sara Sea
by the rivers Ob and Yenisei. As a result, the surface waters of the eastern
portion of the Kara Sea are of low salinity but water of higher salinity
creeps toward the coast along the bottom.

The waters in the western part of the Kara Sea, off Novaya Zemlya are
of much higher salinity, the surface layer having salinities of 30 to 34 o/oo
and the deeper layers salinities in excess of 32 o/oo.

Temperature . In summer large regions of open water may be found in the
southeastern part of the sea and there the surface temperature may rise to
5 to 10°C., but where scattered ice is present it lies close to 0°C. Nega–
tive temperatures are found at shallow depths, appr o aching freezing point at
depths of 20 to 30 m. In winter the entire layer down to 20 or 30 m. has
probably a temperature close to freezing point.

To the east of the northern part of Novaya Zemlya there is found a layer
between 75 and 150 m. which has slightly higher temperatures than the over–
lying and underlaying strata. It is probable that this layer contains an
admixture of Atlantic water that barely penetrates into the Kara Sea from the
Barents Sea.

Currents . The surface waters of the Kara Sea are generally carried
to the north but it is possible that a counterclockwise circulation is present
in the southwestern part where relatively high salinity water is found off
the coast of Novaya Zemlya. The deeper and more saline water must move toward
the coast because the net transport of salt away from the coast must be zero.

Ice Conditions . From November through June the Kara Sea is closed to
navigation, but in July the southern part of the sea becomes partly ice-free
or covered by scattered but heavy ice. Navigation through the three passages
leading into the Kara Sea from the west, Yugor Strait, Kara Strait, and
Matochkin Strait, is generally possible. The best ice conditions are found
in August and September when the greater part of the southern and southeastern
areas are open. In October icebreakers are needed, except perhaps in the
southwestern part.

The Barents Sea

Limits . The limits can be taken as those proposed by the International
Hydrographic Bureau (Fig. 1).

Bathymetric Features . The bottom of the Barents Sea has more the char–
acter of a continental borderland than of a continental shelf, because it is
quite irregular with many banks and depressions. Fairly flat shelf areas
are found to the east and southeast of the Spitsbergen islands and in the
entire southeastern part of the sea. Bear Island and Hopen Island lie on
the Spitsbergen shelf. An east-west ridge connects the shoal areas around
Franz Josef Land with those around Svalbard and a north-south ridge, approx–
imately between longitudes 32° and 40°E., separates the western depressions

Page 030

EA-Oc. Sverdrup: Oceanography

from the eastern. The saddle depth of the former ridge is less than 200
meters, that of the latter is greater than 300 m. The western depression,
the Bear Island depression, runs between Bear Island and the north coast
of Norway and branches to the east of longitude 30° E. The greatest depths
are in excess of 400 m. The eastern depression, with depths in excess of
300 m. and a few depths greater than 400 m. runs southwest between Franz
Josef Land and Novaya Zemlya and continues across a region with sill depth
less than 300 m. into a north-south irregularly shaped trough between
longitudes 40° and 47° E. In the extreme northern part of the Barents Sea
a third depression extends south between North East Land and Franz Josef
Land, sending branches toward the west and the east.

Salinities . The salinities of the waters of the Barents Sea are
greatly influenced by the inflow of Atlantic water, of salinity slightly
above 35 o/oo. Thus in the Bear Island depression the salinity of the water
lies between 35.0 o/oo and 35.1 o/oo except in the surface layers in the
south and in the north. In the south, surface salinities between 34 o/oo
and 35 o/oo are present because of admixture of fresh water from rivers,
and in the north the surface salinities may drop below 30 o/oo because of
melting of ice. The deeper parts of the eastern depression are filled by
water of salinity around 35 o/oo.

To the north, between North East Land and Franz Josef Land, the surface
waters have salinities between 32 o/oo and 34 o/oo, whereas salinities
slightly above 34.9 o/oo are encountered below a depth of 200 m. This water
contains a considerable admixture of Atlantic water which has moved around
the north coast of the Spitsbergen islands (see under “Currents”).

The waters on the extended shelf areas to the southeast of Spitsbergen
and off the coast of Russia are of salinities below 34 o/oo. In the former
region they represent arctic surface waters, in the latter region coastal
waters that are diluted by runoff.

Temperature . The bulk of the Atlantic water in the Bear Island depres–
sion, which is open to the west, is of a temperature between 5° and 7°C.
The temperature decreases toward the east, and in longitude 33° E., on the
north-south ridge, the temperature of the water lies between 2 and 5°C.
In the northern part negative temperatures are found near the bottom.

The deeper parts of the eastern depression are filled by water of tem–
perature −1.5° to −1.8°, but at intermediate depths of 100 to 300 meters
temperatures between 0° and 2°C. are encountered, especially in the southern
and eastern parts. The water of that temperature has a considerable admix–
ture of Atlantic water, part of which appears to continue around the north
point of Novaya Zemlya into the Kara Sea.

In the extreme northern portion of the Barents Sea, the high salinity
water that is found between Franz Josef Land and North East Land below 150
to 200 meters shows temperatures between 1° and 2.5°C., except near the bottom
where temperatures slightly below 0° are encountered.

In the north and northeast the warm waters of the Barents Sea are covered
by Arctic water which in summer shows temperatures of about 0° near the sur–
face and temperatures close to freezing point at depths of 25 to 50 m. In
winter the waters above these depths are probably cooled to freezing point.

In the southwest the surface temperature rises in summer to values
above 10°C. and remains positive throughout the year.

Oxygen distribution. The waters of the Barents Sea are well aerated.
In summer a slight supersaturation is generally found to a depth of 25 to
50 m., and near the bottom the waters are above 80% saturated.

Currents . The principal exchange of water between the Barents Sea
and the adjacent seas consists in an inflow of Atlantic water, mainly
between the north coast of Norway and Bear Island, and an outflow of Arctic
surface water, mainly close to Bear Island, on the southern side, and
between Bear Island and West Spitsbergen. The North Cape Current is
responsible for the larger inflow. This current runs east off the extreme
northern coast of Norway, continues at some distance from the Murmansk
Coast and can be traced to the southwestern coast of Novaya Zemlya along
which glass balls used as floats by Norwegian fisherman and frequently found.

Part of the water of this current is cooled and spreads to the north
in the eastern depression of the Barents Sea, where water of salinity about
35 o/oo and of temperature around −1.5° to −1.8°C. is found. In the south–
eastern part low salinity water, with admixture of river water, flows east
and northeast.

To the north of North Cape Current, Atlantic water also flows east and
northeast, but the flow is more sluggish, and over the north-south ridge the
current branches and extensive mixing with Arctic surface water takes place.

Between North East Land and Franz Josef Land the current is generally
directed to the south into the Barents Sea. The upper layers of this current
carries Arctic water, but below a depth of about 150 meters there is a con–
siderable admixture of Atlantic water which has traveled around the north
coasts of Spitsbergen and North East Land.

The Bear Island Current that follows the bottom slope to the south of
Bear Island carries Arctic water which has a relatively high salinity because
of admixture of Atlantic water. This current is narrow and probably carries
only a small amount of water. The greater outflow of Arctic water takes place
south of the Spitsbergen islands where the Spitsbergen Current flows around
South Caps and turns north, following the west coast of West Spitsbergen.
In addition, the salinities of the Kara Sea indicate that some Atlantic
water flows cut of the Barents Sea around the north point of Novaya Zemlya.

Ice Conditions . In August ice conditions are more favorable to naviga–
tion than in any other month. The coasts of the sea are generally ice-free,
except the eastern coasts of the Spitsbergen islands and the north coast of
Franz Josef Land. The ice conditions do, however, in nearly every year permit
vessels to reach as far as to North East Land in the Spitsbergen group. In
some years North East Land can be circumnavigated.

In September the situation is nearly similar, but from October on the
ice starts advancing from north to south and from east to west. Gradually
the northern and eastern regions become ice-covered. The ice reaches its
widest distribution in March-April, when only the southwestern part remains
ice free. The area of open water varies a great deal from one region to an–
other, but on the average the ice limit runs east-southeast from Bear Island
to about latitude 70° N., longitude 45° E. and then southwest toward the
eastern part of the Kola Peninsula between the Barents Sea and the White Sea.
Thus, the north coast of the Kola Peninsula is ice-free.

The variations in the extent of open water in the Barents Sea in April or
May have been correlated with the volumes of Atlantic water entering the

Page 034

EA-Oc. Sverdrup: Oceanography

Norwegian Sea to the north of Scotland. There are indications that with
a lag of two years a smaller or larger inflow will be reflected in the
amount of open water. Thus the i n flow was very large in the spring of 1929
and two years later, is May 1931, the ice-free area in the Barents Sea
was exceptionally great.

A few small icebergs may be observed in the northern part of the
Barents Sea, coming from glaciers on Novaya Zemlya, Franz Josef Land, or
Spitsbergen.

The Norwegian (or Greenland) Sea

The waters and currents of the Norwegian Sea shall be discussed only
to the extent to which they influence conditions or are influenced by
conditions in the Polar Sea.

Limits . The limits of the Norwegian (or Greenland) Sea are shown
on Figure 1.

Bathymetric Features . The bottom topography of the Norwegian Sea
is characterized by two deeps, the Greenland Deep to the north and the
Norwegian Deep to the south. Within each of them the depths exceed 3,000
meters. They are separated by an east-west ridge in about 72° N., the Jan
Mayan ridge, with a maximum depth of about 2,200 m. The small volcanic
island of Jan Mayen rises at the western end of this ridge. To the north
the Greenland Deep is separated from the deep parts of the Polar Sea by
the Nansen Ridge, the deepest part of which, according to the latest
estimate, is about 1,750 m.

Salinity . The Atlantic water which flows north off the west coast

Page 035

EA-Oc. Sverdrup: Oceanography

of Norway, enters the Norwegian Sea with a salinity of 35.4 o/oo. As
the waters continue along the Norwegian coast, its salinity decreases
somewhat by admixture of coastal water and water from the central parts
of the sea. Off Finmark in northern Norway the salinity is about 35.1 o/oo
and off West Spitsbergen the salinity is around 35.0 o/oo.

Arctic water which is found off the east coast of Greenland has, on
the other hand, salinities between 31 o/oo and 34.5 o/oo. The lowest
value applies to water that contains a considerable admixture of melt water.
In summer the surface salinity in lanes or openings can be very much lower.

The waters of the large central parts of the sea show a uniform salinity
of 34.90 to 34.94 o/oo, except that the surface waters way be slightly
diluted by excessive precipitation.

Temperature . All the Atlantic water which flows into the Norwegian
Sea, has a temperature above 4°C. and the larger part has, is general, a
temperature between 8° and 9°C. The temperature decreases as the water flows
north, but off northern Norway the maximum temperature is still around 6°C.,
and even off Spitsbergen, in latitudes 78° to 79° N., temperatures above 5°C.
have been observed. These values apply to conditions below a depth of about
100 m., where an annual variation related to processes of heating and cooling
practically disappears.

The Arctic water off eastern Greenland has temperatures between 0° and
−1°C. In the entire central part of the Norwegian Sea low temperatures are
encountered, except in summer when a thin surface layer is heated. Below a
depth of 600 m. the water has a temperature close to −1°C. In the northern
basin, the Greenland Basin, the temperature at the bottom is −1.1° to −1.2°C.,

Page 036

EA-Oc. Sverdrup: Oceanography

and in the southern, the Norwegian Basis, the temperature is about −1.0°C.

Currents . The circulation in the Norwegian Sea is counterclockwise;
along the coast of Norway the current is directed to the north, and off the
coast of Greenland it is directed to the south. Between these two main
currents there exist a number of eddies and swirls, some of which are
probably stationary, whereas others may represent traveling disturbances.
The current which flows north along the coast of Norway enters the Norwegian
Sea north of Scotland and represents a branch of the Gulf Stream system of
the North Atlantic. The total volume of Atlantic water which flows in
amounts, on as average, to about 3 million cubic meters per second, but it
varies considerably from one year to another. Off the Norwegian coast the
surface velocities reach values of about 30 cm./sec. or 0.5 knots. As the
current flows north, water is lost to the large eddies of the open Norwegian
Sea, but some water is added by outflow from the Skagerrak and by runoff
from the western coast of Norway. When reaching the extreme northern coast
of Norway the current branches, one branch flowing east into the Barents Sea
(see above) and another branch flowing north along the continental slope that
represents the western border of the Barents Sea. The latter branch con–
tinues along the west coast of West Spitsbergen, where it mixes with the cold
waters of the Spitsbergen Current coming from the Barents Sea. North of
Spitsbergen the last branch of the Atlantic water turns east and submerges
under the lighter, less saline surface layer of the Polar Sea. Part of this
water turns south between North East land and Franz Josef Land. The total
amount of water that is carried north off Spitsbergen is not known.

The East Greenland Current carries Arctic water of low salinity and low

Page 037

EA-Oc. Sverdrup: Oceanography

temperature. This current represents a continuation of the slow drift
of surface water across the Polar Sea, and as the flow becomes concen–
trated off the coast of Greenland, the speed of flow increases. The
highest velocities are reached along the continental s l ope where, in the
surface layer they amount to about 0.5 knots or 12 miles in 24 hours. At
the very surface northerly winds may lead to much greater speeds. The
current flows out of the Norwegian Sea through Denmark Strait between
Iceland and Greenland, and must on an average carry nearly the same amount
of water out of the sea as is carried into it by the branch of the Gulf
Stream system because only small amounts of water flow through the other
opening between the Polar Sea and the Atlantic and Pacific oceans.

Formation of Deep and Bottom Water . Nansen has shown that the deep
and bottom water of the Norwegian Sea is formed in winter in the western
part of the sea, probably north and south of Jan Mayen Island. Water of
salinity 34.90 o/oo to 34.94 o/oo is formed by mixing of Atlantic and
Arctic water and in winter when this water is cooled to a temperature of
about −1.0°C. it attains a sufficiently high density to sink to great depths.
This concept is based on the fact that in March to may surface salinities
down to 34.94 o/oo have been observed in the neighborhood of Jan Mayen,
and at the same time surface temperatures down to −1.9°C. have been recorded.
The cooling must lead to slightly lower temperatures to the north because
the bottom water in the Greenland Basin has a slightly lower temperature
than that in the Norwegian Basin, −1.2°C. against −1.0°C. The process
varies in detail from one year to another. In years with intense cooling
the water may sink all the way to the bottom, but in other years it may

Page 038

EA-Oc. Sverdrup: Oceanography

attain a slightly lower density and spread out at a higher level.

Below a depth of less than 1,000 meters the deep and bottom water of
the Norwegian Sea is completely cut off from any communication with the
Atlantic Ocean. Deep and bottom water of a higher temperature is formed
inthe Atlantic (see the Labrador Sea) and differs characteristically from
that of the Norwegian Sea. The latter is, however, in communication with
the Polar Sea down to a depth of about 1,750 m., and this feature will
be shown to be of great importance to conditions in the Polar Sea.

Oxygen Content . The waters of the Norwegian Sea are well aerated.
The deep wa t er is about 88% saturated, which shows that it must be fre–
quently renewed by vertical convection reaching from the surface to great
depths.

Ice Conditions . Thanks to the influence of the warm Atlantic water,
navigable waters extend farther to the north in the northern Norwegian Sea
than in any other part of the Arctic. In late summers luxurious ships take
their passengers into the fjords of West Spitsbergen and continue north
until they sight the drifting ice, often in 80°N. or even farther north.
In some years the sea north of Spitsbergen may be ice-free to 82° N., and
the entire Svalbard group of islands may be circumnavigated.

The west coast of West Spitsbergen generally remains ice-free through
October, but toward the end of that month ice is being carried around
South Cape by the Spitsbergen Current and starts forming a belt of more
or less scattered ice off the southwest coast. The ice becomes more and
more open the farther north it advances, and therefore the northwestern
part of the coast can be reached much later in the season than the southwestern.

Page 039

EA-Oc. Sverdrup: Oceanography

The fjords generally freeze in November and until that time ships can sail
to Kings Bay in latitude 79° N., and in many years they can also reach
Advent Bay, at Icefjord, in latitude 78° 15′ N.

In spring Kings Bay can be reached as soon as the bay ice breaks up,
generally by the middle of May, but the date when shipping can be resumed
to Icefjord depends not only upon when the bay ice breaks up, but also
upon the amount of ice that is carried around South Cape. Since 1907,
when coal mining in Advent Bay has been carried out regularly, the length
of the season of navigation has varied a great deal and has increased con–
siderably. In the three decades, 1911-20, 1921-30 and 1931-40, the length
of the shipping season was 100 days, 148 days, and 178 days, respectively.
This change is related to the change in the climatic condition in the
European Arctic, which is also reflected in the general oceanographic con–
ditions (see below).

South of Spitsbergen the ice extends at the end of the winter to Bear
Island, but in summer the region between Bear Island and the Spitsbergen
islands is generally navigable.

Off East Greenland a belt of ice is present the whole year. It reaches
its greates extension in February-March when Jan Mayen often is surrounded
by ice, and when the ice may be sighted at the north coast of Iceland. In
August-September the ice belt is sufficiently open to permit specially built
ships to reach the coast south of latitude 70 to 42° N. and to follow the
coast to about latitude 77° N.

Changes in Oceanographic Conditions . The changes in ice conditions
which are mentioned above appear to be related to the general increase in

Page 040

EA-Oc. Sverdrup: Oceanography

the air temperature which during the last decades has occurred in northwestern
Europe. The temperature increase is particularly evident in winter and is
less conspicuous in summer. It is greater in the northern areas and is
especially great in Spitsbergen, where in the decade 1921-30 the average
temperature in the months November to March was 5.0°C. higher than in the
decade 1911-20, and in the decade 1931-40 it was 8.9° higher. At the same
time the southerly winds have become more frequent and it is probable that
the higher temperatures are caused by the increase in the frequency of winds
from the south. Such an increase must also be reflected in the oceanographic
conditions, particularly in an Increase in the transport to the north of
warm Atlantic water.

The oceanographic observations from the waters to the west of Spitsbergen
are not systematic enough to permit a detailed analysis, but they indicate
that considerable changes have taken place in the expected direction.

In the years 1910, 1912, and 1922 the maximum temperature of the Atlantic
water off West Spitsbergen varied between 2.57° and 4.47°C., but in 1931 it
reached 5.04°. In the earlier years the maximum salinity varied between
34.95 o/oo and 35.06 o/oo, but in 1931 it reached 35.14 o/oo. Evidently
the temperatures and salinities were higher in 1931 than in 1912 to 1922,
and the higher values Indicate a greater or faster transport to the north
of Atlantic water. It may be added that in the early part of this century
no cod was caught off West Spitsbergen, but about 1927 the cod, which stays
away from very cold water, appeared and until 1947 profitable fisheries were
conducted. In 1948, the cod fishery off Spitsbergen failed, [ ] but it is
still too early to decide if the failure represents a temporary setback or
the reversal of a trend.

The Polar Sea

Limits. The limits of the Polar Sea are shown in Figure 1, except that
from the oceanographic point of view the Beaufort Sea shall be considered
an integral part of the Polar Sea.

Bathymetric Features . The Polar Sea represents a deep basin between
the American and European-Asian continents, which was discovered by Fridtjof
Nansen during the drift of the Fram in 1893-96. Hansen based his plan for
that famous drift on the well-founded idea that the at that time large
unknown part of the Arctic was covered by water; but, in common with the
few of his contemporaries who agreed, he believed the waters to be shallow.
Therefore, he brought only short sounding lines, and when he found that he
could not reach bottom with these, the crew of the Fram had to untwine
heavy steel cables and splice the strands together again in order to make
a long enough line. With this cumbersome equipment Nansen not only carried
out a number of deep soundings, but he also lowered thermometers and brought
up water samples from great depths. He and his men used a week to collect
oceanographic observations which with present-day methods can be obtained
in a few hours. Nansen’s discussion of his observations were carried out
with such an Insight and ingenuity that all of his major conclusions have
been substantiated by subsequent results. His chart of the bathymetric
features of the Polar Basin has been modified, but has not been radically
altered.

The present concept of the bottom configuration of the Polar Basin is
shown in Figure 1. The depth contours of 3,000 meters or more are based
on less than 100 soundings from the following sources:

Nansen, Expedition Fram drift, 1893-96, wire soundings

Storkerson, ice journey, 1918, Beaufort Sea, wire soundings

Amundsen, flight, 1925 (88° N., 10° W.) echo soundings

Wilkins, flight, 1925 (78° N., 175° W.) echo soundings

Wilkins, Nautilus , 1931 echo soundings

Papanin, Drifting Pole Station, 1937-38, wire soundings

Papanin, Sedov drift, 1938-40 wire soundings

Cherevichnyi, flight, 1941 (78° to 80° N., wire soundings
170° to 180° W.)

Over all ocean areas the numerous soundings from recent years have re–
vealed that the bottom is far more irregular than assumed on the basis of
a few scattered data. The same is true for the Polar Sea, but there only
the region to the north of Spitsbergen is sufficiently well known to give
a few details. Only in this region are the depth contours plotted as full
drawn lines, otherwise they are dashed to indicate their uncertainty.

Nansen first established the probability that the Polar Basin is
separated from the Greenland Basin by a submarine ridge, the saddle depth
of which he estimated to be about 1,200 m. After the Nautilus expedition
in 1931, H. U. Sverdrup estimated the depth to be about 1,500 m. and
believed the ridge to run in a direction WNW to ESE between the extreme
northeastern corner of Greenland and the northwestern corner of West Spits–
bergen. From an examination of the soundings and oceanographlc observations
of the Soviet expeditions, the Drifting Pole Station and the Sedov , Wüst
concluded that the ridge known as the Nansen Ridge runs nearly east-west
along the 80th parallel and that the saddle depth probably is about 1,750 m.

The latter soundings also reveal that to the north of the ridge lies
a small isolated basin with depths exceeding 4,000 m. This is demonstrated
by two soundings, one which gave a depth of 4,l60 m. only 80 miles to the
northeast of the northeast corner of Greenland, in latitude 82°54′.N. (Pole
Station), and one which showed a depth of 4,100 m. farther east in nearly the
same latitude ( Sedov ) . To the north of this small depression another sub–
marine ridge, which Wust calls the Polar Ridge , appears to run about east–
west in latitude 84° N. Here the soundings of the Pole Station gave a min–
imum depth of 2,380 m, but the sa d dle depth of the ridge may be different. To
the north of the Polar Ridge lies still another minor depression which is
separated from the Polar Basin proper by a low ridge in about 86° N.

The extent, maximum depth, and detailed features of the Polar Basin
itself still have to be explored. It is, however, ascertained by 26 sound–
ings, that depths exceeding 4,000 m. are found north of 84° N. between longi–
tudes 0° and 90°E. In 86°26′5″ N., 39°25′ E. the Sedov reports having paid
out 5,182 m, of wire without reaching bottom, but on the following day,
when the position was 86°23′ N., 33°35′ E. a bottom sample was brought up
from a depth of 4,977 m. The drift of the Sedov was not planned, but the
ship was accidentally caught in the ice and drifted for 2 l/2 years with a
group of volunteers on board. The ship was not equipped for deep sea work,
and the crew had, like the crew of the Fram , to splice together strands of
heavy steel wire ropes in order to make long sounding lines. Is spite of
the difficulties under which the soundings were made, there is no reason
to believe that the results are in error, particularly because they agree
with other soundings when such are available. They demonstrate, therefore,

Page 044

EA-Oc. Sverdrup: Oceanography

that in about latitude 86°30′ N. and longitude 40° E. depths exceeding 5,000 m.
are present.

The maximum depth reported from the Polar Sea, 5,440 meters, is, however,
doubtful. This depth was obtained by Wilkins in 1927, using a portable echo–
sounding gear, when he and Eielson made a landing on the ice in 73° N., 175° W.
In 1941, the Russian aviator Cherevichnyi landed in three localities in the
neighborhood and found depths ranging from 1,850 to 3,400 m. These soundings
do cast doubt on Wilkins’ result and demonstrate that, if the latter is cor–
rect, it must represent an isolated deep. The sounding of 4,684 m. by
Storkerson (in the Beaufort Sea, lat. 72° N., long. 147° W.) also needs con–
firmation, because soundings to the north and northwest in the immediate
vicinity show depths of 2,000 and 3,500 m.

Evidently only the depths in the small section 10° W. to 90° E. are
reasonably well known; otherwise the information is very scanty. From the
geographic point of view the establishment of the true shape and the detailed
bottom topography of the Polar Basin is one of the greatest remaining tasks
of exploration.

Bottom Sediments . Our knowledge of the bottom sediments is based on
examination of samples which were collected on board the Fram and the Sedov
by means of tubes that were attached to the sounding lines. All samples from
depths exceeding 3,000 m. showed soft mud, most of which varied in color from
light to dark brown, and some of which contained small grains. A few samples
showed gray mud.

Salinity . The known part of the Polar Sea is co v ered by a layer of water
of low salinity, the Arctic Water. This layer can again be divided in two,

Page 045

EA-Oc. Sverdrup: Oceanography

an upper layer within which the salinity and temperature are influe n ced by
the local processes of freezing and melting, and a lower layer which repre–
sents a transition to the deeper strata.

The upper layer appears to be very similar to the corresponding layer
on the north Siberian shelf (see East Siberian Sea). It has an average
thickness of about 30 meters and toward the end of the winter the salinity
is very nearly uniform from top to bottom. Nansen’s observations on the
Fram expedition of salinities in this layer show less uniformity than the
corresponding observations during the Maud expedition, but this feature can
be ascribed to the fact that the technique of exact salinity determinations
was not developed at the time of the Fram expedition. In summer, in the
openings in the ice, there is often found a 1- to 2-meter-thick layer in
which the salinity may be greatly reduced because of admixture of salt water.
In small openings the salinity may be only 1 or 2 o/oo, but in wider lanes
it is greater. The admixture of melt water also leads to a small reduction
in the salinity down to a depth of 30 to 40 m., but Nansen points out that
below a depth of 40 m. there is no trace of mixing from above in summer or
vertical convection in winter. This feature is probably of importance to
the question of the organic productivity of the waters of the Polar Sea
because it implies that vertical convection in winter cannot bring plant
nutrients, phosphates, and nitrates, from greater depths to the surface.
[ ] Thus, the deep vertical convection, which in the Subarctic is of such a
profound importance to the productivity, is lacking in the Polar Sea. A
slow mixing from below must, however, take place because the average
salinity increases from east to west, that is, in the direction of flow,

Page 046

EA-Oc. Sverdrup: Oceanography

and because this increase cannot be ascribed to the effect of freezing since
the average ice thickness does not increase. No measurements of phosphates
and nitrates are available and there is therefore no basis for estimating
the productivity in the Polar Sea.

To the north of the Laptev Sea the average salinity of the upper layer
is 29 o/oo to 30 o/oo, but it increase to about 33 o/oo to the north of Spits–
bergen. These values are, however, very rough because of the few available
data.

At the bottom of the upper layer the salinity Increases abruptly, gen–
erally by an amount of 1 o/oo to 3 o/oo. Nansen’s observations, which admit–
tedly are uncertain, show at 60 m. a salinity of about 33.8 o/oo regardless
of the salinity in the upper layer. His observations were not spaced closely
enough to reveal details as to the increase, but from analogy with conditions
in the East Siberian Sea it is assumed that the increase actually takes place
within a few meters. Such a nearly discontinuous increase in salinity means
a corresponding increase is density, which effectively prevents further pen–
etration of vertical convection.

With increasing depth the salinity increase, but details are not known.
To the north of Spitsbergen a salinity maximum slightly above 35 o/oo has
been observed at depths of 200 to 400 m., but it is not known if an intermediate
maximum is present over the greater part of the Polar Sea or if over large
areas the salinity gradually increases to the characteristic value of the
deep and bottom water.

The only available and reliable observations of the salinity of the deep
and bottom water were made north of Spitsbergen in 1922 ( Ringsel ) and 1931

Page 047

EA-Oc. Sverdrup: Oceanography

( Nautilus ). Below 1,500 m. the salinity is close to 34.92 o/oo, probably
increasing toward the bottom by an amount of about 0.01 o/oo in 1,500 m.
Nansen’s observations gave considerably higher values, but are systematically
is error. They showed, on the other hand, no variations from station to
station, for which reason it is probable that the entire Polar Basin is
filled by water of salinity 34.92 to 34.94 o/oo.

Temperature . The temperature measurements from the Polar Sea are more
accurate than the salinity observations. This applies particularly to the
data which were so very carefully collected and analyzed by Nansen. Still,
so few observations are available that some of the following statements may
be modified when more information has been gathered.

The entire Polar Sea is covered by a layer of water which in winter is
cooled to freezinb g point corresponding to its salinity. The thickness of
this layer is probably about 30 meters. In summer the surface temperature
in openings in the ice may rise to about 0°C. and at depths down to 30 m.
it probably rises to about 0.1° above freezing point.

Below this surface layer the temperature shows no annual variation, but
it rises gradually as the depth increases. Between longitudes 10° W. to
120° E., from which observations are at hand, a temperature of 0° is reached
at a depth of 100 to 250 m. The depth to the 0° surface increases from west
to east and from south to north. Thus, along the 80th east meridian a tem–
perature of 0° is in 72° N. encountered at a depth of 100 m., and at the
North Pole at a depth of 250 m.

A temperature maximum is revealed at depths which also decrease from
south to north and simultaneously the extreme value of the temperature at

Page 048

EA-Oc. Sverdrup: Oceanography

the maximum also decreases. Along the 80th east meridian the Sadko observed
in latitude 82°42′ N. a maximum temperature of 2.6°C. at 300 m., whereas
at Drifting Pole Station a maximum temperature of 0.74°C. was observed at
400 m. when the party was in latitude 88°53′ N.

Below the maximum the temperature decreases and 0°C is reached at depths
varying between 1,000 and 700 m., d e pending upon the latitude, the greatest
extent of the warm water being found to the south. The decrease in tempera–
ture continues to a depth of 2,000 to 3,000 m., where a minimum value of
about −0.85°C. is found. At still greater depth an increase takes place
such that at 4,000 m. the temperature is about −0.65°C.

The increase of the temperature toward the bottom was first discovered
by Nansen and was later on verified by observations at the Drifting Pole station.
Nansen first (1902) thou g ht that the increase was an evidence of a flow of
heat from the interior of the earth, but later on (1912) he advanced the idea
that the temperature increase was the effect of adiabatic processes in a water
mass of uniform salinity. Such processes are related to the fact that water
is slightly compressible, for which reason a small mass of water that is
brought from a great depth (great pressure) to the surface (atmospheric pres–
sure) increases its volume slightly. If the change in pressure is brought
about adiabatically, that is without heat being added to or removed from the
water mass, the work against pressure must take place at the expense of the
internal energy of the water whereby its temperature must sink. Vice versa,
if a water mass is brought adiabatically under great pressure its temperature
must rise.

Page 049

EA-Oc. Sverdrup: Oceanography

In order to eliminate these effects of pressure the potential temperature
has been introduced and defined as the temperature which a small water mass
would have if it were brought adiabatically to the surface of the sea. The
potential temperature is always lower than the temperature in situ , and the
difference depends on the depth of the sample and its salinity and tempera–
ture. Tables have been prepared for computing potential temperatures. Applied
to the Polar Sea, it is found that the potential temperature at the bottom is
0.93°C. according to observations of Nansen and of the Polar Station. It is
also found that even the potential temperature increases slightly toward the
bottom, but the increase amounts only to 0.04° on a vertical distance of 1,000
to 2,000 m. According to the Nautilus observations, it is probable that the
salinity also increases slightly toward the bottom and if this is generally
true the stratification is stable and corresponds to conditions in other deep
basins.

Oxygen Content . The only available observations of the oxygen content
of the waters of the Polar Sea were made north of Spitsbergen during the
Nautilus expedition in 1931. From these, and from analogy with the conditions
on the East Siberian Shelf, it may be concluded that the surface layers are
rich in oxygen and that the saturation value hardly drops below 80%. To the
north of Spitsbergen the relatively warm water at Intermediate depth showed
90 to 95% saturation, and the deep water, below 2,000 m., was 80 to 83% saturated.
Information about the oxygen content of the deep water in other areas will be
of great interest because it may tell something about the movement of the water.

Currents. The surface waters of the Polar Sea move slowly from the Alaska–
East Siberian side of the sea to the wide opening between Spitsbergen and

Page 050

EA-Oc. Sverdrup: Oceanography

Greenland, where their speed increases and where they unite in the rapid East
Greenland Current. This concept was first advanced in 1884 by the Norwegian
meteorologist H. Mohn and was based on the remarkable finding off Julianehaab
Greenl i a nd, of articles from the ill-fated Jeannette expedition. The Jeannette
expedition left San Francisco on July 8, 1879, under the command of Lieutenant
C. W. De Long. in September 1879, the ship was caught in the ice to the west
of Wrangel Island in latitude 71°35′ N., longitude 175°06′ E., and during the
next two years it drifted with the ice toward the west-northwest, more or less
paralleling the Siberian coast until, on 12th June 1881, it had to be aban–
doned to the northwest of the New Siberian Islands, in latitude 77°15′ N.,
longitude 154°59′ E. Of the 33 men on board, 22, including De Long himself,
succumbed in their attempt to reach settlements at the Lena Delta.

Three years later, in the summer of 1884, Eskimos who were hunting off
Julianehaab, near the southern point of Greenland, found on an Ice floe several
articles which were readily identified as belonging to the Jeannette or her
crew. They comprised pieces of clothing marked with names of members in the
party and a list of provisions signed by De Long himself. Mohn concluded
that these relics must have drifted across the Polar Sea.

Mohn’s conclusion gave Firdtjof Nansen his idea for the drift expedition
of the Fram . He reasoned that the logical way to reach the unexplored part of
the inner Arctic, and perhaps the North Pole, would be to work with the current
and not against it. When preparing his plans he examined all possible evidence
for the existence of a current which would carry a ship across the Polar Sea,
provided it was built is such a manner that it would not be crushed by the ice.
In February 1890, Nansen presented his plan for his expe id di tion in an address to

Page 051

EA-Oc. Sverdrup: Oceanography

the Norwegian Geographical Society and went into great detail regarding the
probability of such a current as was suggested by Mohn. He pointed out
that at Godthaab, in West Greenland, there had been found a “throwing stick”
with decorations, which, according to ethnographers, is used exclusively by
Alaskan Eskimos near Bering Strait. Furthermore, that much of the driftwood
commonly found on the east coast of Greenland can only have come from Siberia
while a small portion may possibly have come from the American continent.

Besides this direct evidence, Nansen advanced good reasons for the
existence of the current. It is indeed interesting to observe that his
analysis has been fully confined by subsequent observations except in cases
when he assumes the Polar Sea to be shallow. In order to account for the
outflow of low salinity water from the Polar See, he points out that ( 1 ) large
quantities of fresh water are discharged into the Polar Sea (or its adjacent
seas) by the Siberian and American rivers, ( 2 ) over the Polar Sea precipita–
tion greatly exceeds evaporation, and ( 3 ) the prevailing winds over the Polar
Sea must be easterly and must help in carrying the surface waters to the west.
It will be shown that all these features are important.

The drift of the Fram from 1893 to 1896 fully confirmed Nansen’s concept
of the current both as to direction and speed. The Fram drifted between the
North Pole and Franz Josef Land and in the general direction of her drift she
covered an average distance of 1.6 nautical miles in 24 hours. Subsequently
the character of the surface current has been demonstrated by the drifts of
other vessels, particularly the Karluk (Point Barrow to Wrangel Island, 1913),
the Maud (Wrangel Island to the New Siberian Islands, 1922-24 ) , and the Sedov
(New Siberian Islands to Spitsbergen, 1937-1940). The drifts of vessels across

Page 052

EA-Oc. Sverdrup: Oceanography

the Polar Sea and its adjacent seas are shown in the “Historical Map of
the Arctic,” by Breitfuss ( Arktis , 1935) in which also the drift of the
Soviet Pole Station has been entered. In addition some information as to
the surface currents has been obtained from the drift cask experiment by
Admiral G. W. Melville and Mr. H. G. Bryant, and from the Andr e é e records
which were found on White Island in 1930, and drift buoys released by
Aadr e é e.

In the years 1899 to 1901, 35 specially designed casks were placed
on ice floes in the vicinity of Point Barrow, Alaska, and at points farther
east. Three of these casks were recovered, one on the Siberian coast, one
on Iceland, and one on the coast of northern Norway. The cask that was
found on the Siberian coast had drifted a short distance only and is of
no int e rest. For the two others the data are: One cask released to the
northwest of Point Barrow, Alaska, on September 13, 1899, was found on the
north coast of Iceland on June 7, 1905. One cask released at Cape Bathurst,
Canada, on July 24, 1900 _f was found on the coast of northern Norway on Nov. 3,
1908.

The first cask may have been carried directly to Iceland after it had
reached the opening between Spitsbergen and Greenland and may have used the
six years between release and recovery to complete the drift. The total
length of time of the drift of the Karluk , the Maud or the Jeannette , and
the Fram represents about six years.

The second cask was probably [ ] carried into the North Atlantic by the
East Greenland Current and then north by the branch of the Gulf Stream
system that flows into the Norwegian Sea. In that case the drift across
the Polar Sea way have taken six or seven years. In both instances it is

Page 053

EA-Oc. Sverdrup: Oceanography

probable that the casks had drifted on the Siberian side of the North Pole,

In 1897 Andr e é e had to abandon his balloon is latitude 82°58′ N., longi–
tude 30° E. From that locality Andr e é e and his two companions tried to
reach North East Land, but as they advanced the current apparently carried
them to the southeast. They missed North East Land and reached instead
White Island, where they succumbed. Their journey demonstrated, however,
that the surface currents tend to move in a clockwise direction around Sval–
bard. This was further demonstrated in 1928 by the drift of the survivors
from the Italian dirigible, the Italia . The ice floe on which they
established camp when the Italia crashed in about latitude 81°15′ N., longi–
tude 25° E., drifted slowly to the southeast.

Andr e é e carried several buoys which were dropped from the balloon between
latitude 82°08′ N. and 72°18′ N. and longitudes 19° E. and 25° E. Of these
one was found on King Charles land, on the east side of the Spitsbergen
islands, thus also demonstrating that the drift in this region is southerly.
Three others which were found on Iceland and one which was recovered on the
coast of northern Norway were probably carried around the southern point of
West Spitsbergen island by the Spitsbergen Current.

The available data show clearly that north of the arctic coast of Siberia
the ice, and with it the surface layer, move slowly from east to west, forming
part of a clockwise circulation around the central part of the Polar Basin.
In the adjacent seas the direction of flow is generally to the north, except
in the western part of the Barents Sea, where the flow between Franz Josef
Land and Svalbard is directed to the south. The major current to the south is
found off eastern Greenland, where the surface waters of the Polar Sea flow out.

Over the known area of the Polar Sea drift of the ice is mainly deter–
mined by the prevailing winds. These are easterly, but the direction of
the ice drift deviates about 30 degrees from the wind direction (see under
East Siberian Sea), la the northern part of the East Siberian Sea the wind
drift dominates completely ( Maud) , and in the region from the New Siberian
i I slands to the longitude of Franz Josef Land the wind drift is by far the
most important factor ( Fram , Sedov ). To the west of that longitude the
wind drift is aided by an independent current which in the Polar Sea is
broad and weak, but becomes narrower and faster as it moves south and finally
flows as the well-defined East Greenland Current,

The character of the drift within the central region of the Polar Sea
and off the Canadian Arctic Archipelago can only be estimated. It is probable
that to the north of the Arctic Archipelago the ice drift is also directed to
the west, in which case there exists a large anticyclonic eddy in the central
region. This eddy is probably o n ly semipermanent, for which reason one may
assume that the ice is carried out from that region as well and is ultimately
carried south by the East Greenland Current. it must also be remembered that
the ice of the Polar Sea is all o v er subject to deformation by changing winds
(see Arctic Sea Ice), for which reason the appearance of the ice may be sim–
liar in all areas.

In the northern part of the East Siberian Sea the average speed of the
drift in the main direction of progress is about 1.0 nautical mile in 24
hours ( Jeannette and Maud ). From the region to the north of the New Siberian
i I slands and to the north of Spitsbergen the average speed of the drift was 1.6 n.
miles in 24 hours in the years 1893-96 ( Fram ) and 2.4 n. miles in 24 hours in
1937-39 ( Sedov ) The difference in the speeds observed on board the ~~Fram~~ Fram and

Page 055

EA-Oc. Sverdrup: Oceanography

the Sedov is quite striking. in discussing the observations of the Sedov,
N. N. Zubov points out that these showed no effect of a permanent current
to the southwest or south before the Sedov was north of Spitsbergen in about
latitude 83° N., and longitude 20° E. Nansen, on the other hand, found a
weak current toward the west from longitude 90° E. and on. If both these
results are correct, they indicate that the average winds from the east
must have been considerably stronger in 1937-39 than they were in 1893-96,
Regardless of any explanation, the fact remains that the ice drift was
faster during the drift of the ~~Sedov~~ Sedov , and this must imply that the outflow
of Arctic water was greater in 1937-39 than in 1893-96. The inflow of Atlantic
water into the Polar Sea must also have been greater, and this feature is,
as will be shown, indicated by the temperature observations.

The large amount of water that is carried out of the Polar Sea by the
East Greenland Current must be compensated for by a corresponding inflow.
Some water flows in through the narrow and shallow Bering Strait, but by far
the greater inflow takes place north of Spitsbergen where the last branch of
the Atlantic water submerges below the less saline Arctic water and continues
as an intermediate relatively wars current.

This wars intermediate water was first discovered by Nansen, who offered
the correct explanation for its presence. Subsequent observations have con–
firmed Nans e n’s observations and conclusions. They have shown that an inter–
mediate current carrying Atlantic water flows off the continental slops on the
Asiatic side of the Polar Sea, at least as far as to the New Siberian Islands.
A vertical section along the meridian 80° E., running from 80° N. to the North
Pole, shows that temperatures above 2°C. are found in the southern part of the

Page 056

EA-Oc. Sverdrup: Oceanography

section and that temperatures above 1° are not encountered to the north of
84° N. These features demonstrate that the current is concentrated south
of 84° N. It is, however, probable that the warm water spreads over most
of the Polar Sea, but measurable currents need not occur except in the
Svalbard-Franz Josef Land sector.

The inflowing warm water must flow out again, but the only major out–
flowing current carries no warm water. It is, therefore, probable that
the warm water as it spreads mixes with the surface layers and perhaps dis–
appears completely off the Canadian and Alaskan coasts. That such mixing
takes place is evident from the facts that the temperature of the maximum
intermediate water is much lower near the North Pole (0.7°) than off Franz
Josef Land (2.5°) and that the salinity of the outflowing surface layer is
considerably higher north of Spitsbergen than north of eastern Siberia.

It was pointed out that, according to a comparison of the drifts of the
Fram and the Sedov, the speed of the flow toward Spitsbergen was in 1937-40
nearly 1.5 times that in 1893-96. There exist no estimates of the velocities
with which the intermediate warm water moves, but a comparison of the tempera–
ture observations from the Fram and the Sedov at localities where the drifts
nearly intersect indicates that the maximum temperature in the warm layer
was in 1937-40 from 0.3° to 0.4°C. higher than is 1893-96. This feature
suggests a more rapid inflow and indicates a change comparable to that which
is known from the Spitsbergen area (see Norwegian Sea),

The Deep and Bottom Water of the Polar Sea Nansen pointed out that the
deep and bottom water of the Polar Sea cannot be formed locally in the Polar
Sea by sinking of surface water because of the salinity of the surface water

Page 057

EA-Oc. Sverdrup: Oceanography

is everywhere lower than that of the deep water. He concluded that the deep
water flows into the Polar Sea across the ridge between Spitsbergen and
Greenland, the existence of which he surmised. On the basis of the observa–
tions which were available to him, Nansen placed the sill depth of this
ridge at 1,200 to 1,500 meters. Subsequent observations ( Nautilus , 1931,
Drifting Pole Station, 1938) have fully confirmed Nansen’s view except that,
according to Wüst’s discussion of the latest data, the sill depth of the
ridge, which is now called the Nansen Ridge , is probably 1,750 to 2,000 m.
At depths of 1,250 to 1,750 m. in the northern part of the Norwegian Sea,
the deep water which forms there by winter convection has a salinity slightly
above 34.9 o/oo and a potential temperature close to −1.0°C. The salinity
is similar to that observed in the Polar Sea on board the Nautilus , and the
temperature corresponds to the temperatures according to observations of the
Fram , the Nautilus , and the Drifting Pole Station.

Since the deep and bottom water flows in across the Nansen Ridge, some
of the deep water must also flow out. At present one can only advance a
hypothesis as to the mechanism by which deep water can leave the Polar Sea.
It seems possible that off the Canadian and Alaskan coasts the intermediate
warm water is absent, and that there mixing may take place between the surface
water and the deep water, such that part of the latter may ultimately be
carried out by the East Greenland Current. This question is one of the many
unsolved questio n s pertaining to the oceanography of the Polar Sea.

Ice Conditions . The Polar Sea as defined here is ice-covered the entire
year, except that a small area north and northwest of Spitsbergen is ice-free
in late summer and that in the same season a narrow lane is navigable along

Page 058

EA-Oc. Sverdrup: Oceanography

the north coast of Alaska. The ice cover is practically unbroken is winter,
but during the months of July and August many open spacesand leads are present.
A further description of the character of the ice is found in the article on
“Arctic Sea Ice.”

Summary

The Polar sea, with its adjacent seas off Europe and Asia, can be con–
sidered a large basin which practically is in communication with the North
Atlantic only, The Polar Sea receives appreciable additions of fresh water,
partly from the large American and Siberian rivers and partly because the
precipitation exceeds the evaporation. The amounts have been estimated at
averaging 0.16 million tons per second and 0.09 million tons per second,
respectively.

The fresh water that flows out from the large rivers must, because of
its low density, spread to the north. In doing so it mixes with the sea water
and changes gradually into a surface layer of relatively low salinity, which
on the whole moves away from the coasts. This layer carries salt away from
the coasts and to compensat for this transport there must exist a deeper flow
toward the coast of more saline water. These considerations account for the
general character of the currents and the distributions of temperature and
salinities within the large shallow areas off Siberia.

Turning to the deep Polar Sea, similar considerations apply. The surface
layers move toward the opening between Spitsbergen and Greenland and, to com–
pensate for the outflow, Atlantic water flows in to the north of Spitsbergen
and spreads over large parts of the sea as a warm layer at depths between 200 m.

Page 059

EA-Oc. Sverdrup: Oceanography

and 1,000 m. The surface drift across the Polar Sea is chiefly maintained
by the prevailing winds which deliver the energy that is needed to overcome
friction, but one of the principal features, the outflow of cold, low–
salinity water and the inflow of warm, saline water, depends upon the ad–
dition of fresh water to the surface layers and is independent of wind con–
ditions.

The deep Polar Basin is filled by water that is renewed by a slow, con–
tinuous or intermittent flow across the Nansen Ridge of water from the Nor–
wegian Sea.

This general picture is well established, but there are numerous un–
solved problems of which a few may be mentioned:

What depths exist in the large unexplored part of the Polar Sea? What is
the surface drift in these areas? What total amounts of water flow out and in?
How does the warm intermediate water spread? Are there are regions where no
trace of the intermediate water is found? How does the deep and bottom water
move? What are the amounts of plant nutrients, phosphates, nitrates, etc.,
at different depths? How rap i dly are the plant nutrients of the surface layer
renewed? What do these last features suggest as to the productivity of the
waters?

In order to answer these questions and many others it is necessary to
explore the entire Polar Sea and to use the best available scientific equipment.
The writer believes that during the months of July and August such exploration
can be conducted from a submarine. From the point of view of oceanographic
exploration, the study of the Polar Sea is one of the major remaining tasks.

Page 060

Herald U. Sverdrup

	Fram	Maud	Sedov
Angle between directions of drift and wind	28°	33°1	30-40°
Wind factor	1.82 × 10 ⁻²	1.77 × 10 ⁻²	2 × 10 ⁻²

Month	Angle of deflection, degrees	Wind factor, × 10 ^-2	Month	Angle of deflection, degrees	Wind factor, × 10 ^-2
January	31.4	1.66	July	39.7	2.16
February	27.4	1.68	August	42.0	2.25
March	22.6	1.51	September	40.4	2.35
April	13.2	1.36	October	38.4	2.25
May	18.5	1.48	November	30.3	1.90
June	27.4	1.85	December	31.3	1.68

	Page
Introduction	1
The Labrador Sea	3
Hudson Bay and Hudson Strait	9
The Northwest Passage	12
The Beaufort Sea	13
The Chukchi Sea	13
The East Siberian Sea	17
The Laptev Sea	25
The Kara Sea	27
The Barents Sea	29
The Norwegian (or Greenland) Sea	34
The Polar Sea	41
Summary	59
Bibliography	60

Oceanography of the Arctic: Encyclopedia Arctica 7: Meteorology and Oceanography