The isthmus with its surroundings has undergone a dramatic transformation since the end of the LIA (Figs. 1, 2; and Figs. 4–6 in: Blaszczyk et al. 2009, 2013), due to glacial recession under the climate warming that intensified since the 1980s.
It should be emphasized that the frontal retreat of the isthmus’ ice coasts has been a final result of the negative net mass balance, i.e., a loss of ice mass, of its former and present tributary glaciers (Ziaja 2001; Pälli et al. 2003; Sharov 2006a, b; Blaszczyk et al. 2009). The uplifts of both the equilibrium (of the net mass balance) line altitude (ELA) and the snow line for each glacier have been the direct effect of a (local) rise in temperature, due to the global influence. This caused a lessening of the accumulation zone of each glacier (including its declining and then disappearing, if situated below the new ELA), which has decreased the volume of the ice outflow from this zone. In addition, higher summer temperatures and a longer season without frost have intensified ablation of ice on the glaciers’ surface. As a result, a lowering of the surface elevation and a decrease in the ice thickness have occurred on all the glaciers. This gives rise to the effect of the so-called retreat of the glaciers (which not only is the greatest at their fronts, but also occurs on their sides). Frontal retreat of tide-water glaciers is additionally activated by sea action, intensifying due to a great shortening of the sea–ice season.
During 1899–1900, the isthmus was 28 km wide and fed by 21 glaciers (Hornbreen by 15 glaciers and Hambergbreen by 6 glaciers); in 1936, 22.4 km wide and fed by 15 glaciers (Hornbreen by 12 glaciers and Hambergbreen by 3 glaciers); and in 2005, 7.9 km wide and fed by 5 glaciers (Hornbreen by 4 of them and Hambergbreen by only one). Feeding from the south was very insignificant in 2005 because two tributary glaciers’ tongues from this area (Mikaelbreen and the unnamed glacier north of the 584 m peak, which flew to Hornbreen) had narrowed and thinned considerably since the 1980s (Wassiliew 1925; C12 Markhambreen 1956; C13 Sørkapp 1986; Table 2; Figs. 1, 2, 3). The Sykorabreen glacier (Fig. 4), previously the main tributary glacier of Hambergbreen from the south, became the tide-water glacier between 1990 and 1995 (Ziaja 1999).
It is worth noting that the glacial-ice isthmus’ width was longer than the shortest line between the most incised ice cliffs in both glaciers’ heads during the twentieth century (because this line went across the Ostrogradskifjella mountain massif).
Nowadays (2013), the isthmus is only 6.2 km wide and only three glaciers are weakly supplying the isthmus with ice, all of them from the north (Isingbreen and Flatbreen are feeding Hornbreen, and Hambergbreen is being fed by Skjoldfonna). Moreover, the delivery of ice from these glaciers is becoming smaller and smaller due to decreases in their thickness and surface elevation (Fig. 3).
The Hambergbukta fjord did not exist, and its valley was filled with the impressive tongue of the Hambergbreen glacier in 1900. Moreover, its lowest part, comprised a great glacial lobe several km long, protruded into the Barents Sea (Fig. 1).
This protrusion resulted in a huge surge of this glacier just toward the end of the nineteenth century. Afterward, its tongue was shrinking, and its cliff was retreating quickly—by ca. 8 km until 1936, and by 14 km in total until 1957. From 1957 to 1961, the glacier front advanced by 2 km (Lefauconnier and Hagen 1991). We suggest that this may have been the beginning of the next surge. According to Lefauconnier and Hagen (1991), this next surge “likely started in 1961” and persisted until 1970, with a total frontal advance by 5 km in the period between 1961 and 1970. Then, the glacier’s front retreated by 5 km until 1988. From 1988 to 1990, the next smaller surge occurred (Jania 2001), i.e., the lengthening of the glacier by 1.5 km. Subsequently, the glacier was thinning, and its front was retreating quickly, by more than 4 km, until 2013 (Figs. 1, 2). Hence, the ice flux down to the ablation area has been very irregular due to the surges.
Generally, the occurrence of the surges is independent of climatic variation (Lefauconnier and Hagen 1991). However, an increase in the quantity of water at the bottom of a glacier, due to ablation of ice under the warming, may stimulate its surge, which could happen in the study area (Jania 2001). Undoubtedly, a dramatic shrinkage of Hambergbreen and its tributary glaciers has become the main environmental and landscape–seascape implication of the surges there. This happened because the periods of intensified climate warming (or the so-called secondary warm fluctuations) followed just after all the three aforementioned surges. Each time, the higher summer temperature enabled the relatively quick ablation of the huge glacier tongue formed by the surge and hindered rebuilding the accumulation zone (firn field) at the upper parts of the tributary glaciers.
Recession of the western part of the isthmus, i.e., the shrinkage and frontal retreat of the Hornbreen glacier, was much more even and unidirectional during all the periods (since the beginning of the 20th century), in spite of surge episodes mentioned by Pälli et al. (2003). A more regular constant outflow of ice has prevailed in Hornbreen and its tributary glaciers. This may be due to their small inclination (Hambergbreen and its tributary glaciers are significantly steeper). Some exceptions, like the small frontal advance of Hornbreen from 2010 to 2011, did not change this general regularity.
Over a period of ca. 70 years (since 1936), the isthmus’ ice thickness decreased by ca. 60 m in the ice shed between both glaciers and by more below it, down to their fronts. Decrease in the ice thickness reached ca. 125 m near the present (2013) Hambergbreen’s front and 150–175 m near the present (2013) Hornbreen’s front (Fig. 2). Hence, a significantly smaller volume of ice melted on the eastern side of the present isthmus’ area which remains under the influence of the cold East-Spitsbergen Current.
Ice thickness on the isthmus is a basic question in the discussion of the future of both the entire isthmus and its neighborhood in the case of a further glacial recession. Is the bedrock under Hornbreen and Hambergbreen below or above the sea level? If the bedrock is below the sea level, then both fjords, Hornsund and Hambergbukta, will be connected into a sound after the melting of the mentioned glaciers (which is likely in the case of the climate stabilizing or further warming) and Sørkapp Land will be transformed into an island. If not, the isthmus will change its character (from icy to rocky) but Sørkapp Land will remain as the peninsula.
After the discoverer Poole who mistakenly named the fjord “Hornsund”, thinking that it is a sound, in 1610 (The Place-Names of Svalbard 2001), the Russian glaciologist Koryakin (1975) put forward the idea that there is a sub-glacial (buried under ice) sound “dividing Sørkapp Land from the main island of the archipelago” cut in the bedrock of the Hornbreen and Hambergbreen glaciers. He had no direct data on the glaciers’ thicknesses. However, he noticed three indirect circumstances which supported his view: low elevation of the glaciers’ surface, deep sea at the glaciers’ fronts, and a lack of any nunataks (isolated rocky peaks which protrude above the glaciers) near the ice-shed.
The next part of this discussion uses the results of radio echolocation surveys of the isthmus’ ice thickness taken from a helicopter. The Norwegian–British team (Drewry et al. 1980) suggested that “Hornbreen is situated on its bedrock in the vicinity, but generally above the sea level and there will be no deep-water connection” by a strait if the glaciers melted. The Russian team (Macheret and Zhuravlev 1985) provided evidence for “the possible existence of a strait under the ice (filled with these two glaciers) separating Sørkapp Land from the rest of (…) Spitsbergen.” It is worth noting that the isthmus’ width was already 12.2 km in 1990 (Table 2).
According to Ziaja (1999, 2001, 2002)—after comparative studies of the papers, topographic maps, air photos, and Landsat satellite images— there appears to be at least a shallow and narrow sound between the Barents Sea and Greenland Sea, and thus the changing of the Sørkapp Land peninsula into an island, is very probable in the event of the glaciers’ melting. He also forecasted the second possibility (if the bedrock would be situated above the sea level): the isthmus being preserved in the form of low (up to a few dozen m) and narrow (up to a few km) land belt (Ziaja 1999).
The “high-resolution ground penetrating radar surveys” (of the isthmus’ ice thickness) at 50 MHz were made by the Finnish–British–Polish team (Pälli et al. 2003) from the surface of the isthmus’ glaciers in April 2000. The surveys’ data, completed with the studies of changes in the glaciers’ surface elevations since 1900 on the topographic maps, led to the following conclusions: (1) Hornbreen and Hambergbreen glaciers are situated below 200 m a.s.l., (2) their beds “lie from –25 to 25 m a.s.l.,” (3) “the low-lying glaciated valley filled by” the glaciers “may become a partially inundated ice-free isthmus within perhaps 100 years,” and (4) “but there is no continuous sub-sea-level channel between Torell Land and Sørkapp Land” (Pälli et al. 2003).
Further studies, based mainly on the satellite interferometry and altimetry (ERS-1/2-SAR interferograms, ICESat-GLAS altimetry data and ASTER-VNIR imagery by 2004), led to the conclusion that “under current environmental conditions” the Hornbreen–Hambergbreen “ice isthmus will disappear by 2020” (Sharov 2006a) and “Sørkapp Land might become a separate island” (Sharov and Osokin 2006).
Transformation of the Arctic peninsulas to new islands due to the glacier shrinkage is usually described in preliminary or summarizing notes on the Internet or in popular press cited below in this paragraph, being omitted, as it is considered to be an obvious phenomenon, in the main stream of scientific literature on glacial recession. The appearances of new islands in both the eastern Greenland and the western Greenland were noted on the Internet (e.g., World Climate Report 2008; Pelto 2011), whereas glacier recession determining their origin was described in the best glaciological journals (Box and Decker 2011; Howat and Eddy 2011). Such a process was observed in the field by one of the authors of the present study in northwest Spitsbergen where Blomstrandhalvøya was transformed from the peninsula to the new island due to the Blomstrandbreen glacier’s recession by 1995 (Fig. 5). In Franz Josef Land, RussianArctic explorers—first hypothesized that “the new island could have split away from larger Northbrook Island back in 2006” (RIANOVOSTI 2012); discovered the island (Fig. 6) in 2008 (Ostrov Yuriya Kuchieva 2013); and “proved that a new strait has formed” in 2012 (RIANOVOSTI 2012). However, all these islands do not exceed 20 km2 and are at least ca. 70 times smaller than Sørkapp Land which would become the biggest new Arctic island in the event of its splitting out from the rest of Spitsbergen.
We made our field survey during a trip on foot across the isthmus: from the Hambergbukta’s south-western coast through the lowest part of the Sykorabreen glacier, and further to the Hornsund fjord along the southern (lowest) isthmus’ edge at the foot of the Ostrogradskijfjella mountainous group, from August 21 to 24, 2005 when the lower parts of the glaciers were devoid of snow (Figs. 1, 2, 3, 4; Ziaja and Ostafin 2005; Ziaja et al. 2007, 2009). Owing to these snowless conditions, the authors of this paper could confirm general regularities discovered by their predecessors, and correct some inaccuracies.
The altitude of the ice pass between Hornsund and Hambergbukta decreased (lowered) from 241 m during 1899–1900 (map 1:200 000: Wassiliew 1925) to ca. 230 m (estimated in the old 1:100 000 map: C12 Markhambreen 1956) in 1936, to 205–210 m in 1961 (estimated in the new 1:100 000 map: C12 Markhambreen 2008), and to ca. 180 m a.s.l. in 2005 (measured by the authors in field). Hence, the mean annual rate of the (glaciers in the) pass lowering has been ca. 0.6 m, similar to the rate ca. 0.5 m reported by Bamber et al. (2005) from “lower elevation glaciers in south Spitsbergen” between 1996 and 2002. The pass also was removed to the south by at least 2 km, just to the foot of the Ostrogradskijfjella’s northern slope (Fig. 3). The thickness of the Sykorabreen glacier (Figs. 1, 2, 4) in the middle of its lower part decreased by ca. 50 m during the period of 1936–2005. This value was obtained by comparing the glacier’s elevation in 1936—red in the old 1:100 000 map (C12 Markhambreen 1956)—with the elevation measured by the authors in the field in 2005.
We also saw that the ELA on Hornbreen and Hambergbreen was significantly underestimated by our predecessors: 250–300 m (Pälli et al. 2003), or 220–250 m (Sharov and Osokin 2006). In fact, the ELAs on all the glaciers explored by us in and near the isthmus (Hornbreen, Hambergbreen, Sykorabreen, Professorbreen, one unnamed glacier, and Mikaelbreen) were surely not lower than 300 m, even in the most shaded northern slopes or valleys of Ostrogradskijfjella. We observed an intensive melting and fissuring of the snowless glaciers below 300 m a.s.l. Moreover, our analysis of the satellite images (LandsatMSS, TerraASTER: USGS) excludes such a significant rise of the ELA during the period of 2000–2005.
On the northern slopes of Ostrogradskijfjella, the traces of a significant lowering (by several dozen m from 1900 to 2005) of the Hornbreen and Hambergbreen glaciers’ surface were very clear (first of all slope incisions and glacial moraines of different types). Contrary to the maps elaborated by Sharov (2006b, c, d), the Professorbreen glacier is completely separated from Hambergbreen due to its shrinkage and shortening. The same refers to the former tributary glaciers of the lower Sykorabreen glacier (Fig. 4).