The Norwegian mountains: the result of multiple episodes of uplift and subsidence

The elevation of the mountains in Norway is geologically young. Much of the present‐day land surface was buried below a thick cover of relatively young sediments in the early Miocene, 23 Ma, when Scandinavia started to be uplifted. Big river systems eroded deeply into the rising landscape and transported sand and gravel from Norway and Sweden to Denmark where the detritus was deposited in a large delta. In Norway, the erosion formed an extensive plain near sea level that included the present‐day mountain plateau of Hardangervidda and extended across a thick pile of sediments that covered the present‐day coastal areas of Norway. Hardangervidda was uplifted to its present elevation of about 1200 m after a second phase of uplift that began about 5 Ma, in the early Pliocene. The hard bedrock of Hardangervidda has preserved this part of the plain as an elevated plateau, but the part of the plain that extended across the sediments has been eroded away, exposing the underlying basement rocks. That re‐exposed basement surface was shaped in the Jurassic when the climate was warm and humid. The basement rocks were weathered where rainwater seeped into fracture zones. Erosion of the weathered rocks has left a terrain of fracture valleys and hilly relief that contrasts with the sub‐horizontal plain of Hardangervidda. This weathered landscape is today exposed on the slope between the west coast and Hardangervidda. While the elevation of the mountains is young, today's landscape has a long history.

of 'only' 2.5 km. Those who favour this explanation base their hypothesis primarily on computer modelling that assumes continuous isostasy and a history of long-term erosion. They argue that the mountains (the top of the iceberg, so to speak) have been supported by a crustal root that has persisted since Caledonian times. This hypothesis has, however, been tested by seismic observations that have failed to find a root sufficient to support the present-day mountains (Fig. 2).
A second group-to which we belong-thinks that field observations show that there is good evidence from both Norway and Greenland that the Caledonian Mountains collapsed during the Devonian, shortly after their formation. Eclogite rocks, formed at depths How old are the Norwegian mountains ( Fig. 1)? Why is there a mountain range along the Atlantic margin of Scandinavia? It is surprisingly difficult to answer even the first of these questions, and we can address the second question only after the timing has been established.
Some 420 Ma, Laurentia, Baltica and Avalonia collided forming the Caledonian mountain chain that stretched from Scandinavia and East Greenland through Scotland and Ireland to eastern North America. Some geoscientists argue that the present-day mountains are just the remnants of these Caledonian mountains that have worn down slowly and gradually. The mountains eventually became lower and lower through those hundreds of millions of years, until the highest summit in Norway today reaches an elevation The elevation of the mountains in Norway is geologically young. Much of the present-day land surface was buried below a thick cover of relatively young sediments in the early Miocene, 23 Ma, when Scandinavia started to be uplifted. Big river systems eroded deeply into the rising landscape and transported sand and gravel from Norway and Sweden to Denmark where the detritus was deposited in a large delta. In Norway, the erosion formed an extensive plain near sea level that included the present-day mountain plateau of Hardangervidda and extended across a thick pile of sediments that covered the present-day coastal areas of Norway. Hardangervidda was uplifted to its present elevation of about 1200 m after a second phase of uplift that began about 5 Ma, in the early Pliocene. The hard bedrock of Hardangervidda has preserved this part of the plain as an elevated plateau, but the part of the plain that extended across the sediments has been eroded away, exposing the underlying basement rocks. That re-exposed basement surface was shaped in the Jurassic when the climate was warm and humid. The basement rocks were weathered where rainwater seeped into fracture zones. Erosion of the weathered rocks has left a terrain of fracture valleys and hilly relief that contrasts with the sub-horizontal plain of Hardangervidda. This weathered landscape is today exposed on the slope between the west coast and Hardangervidda. While the elevation of the mountains is young, today's landscape has a long history.
of more than 50 km, were lifted close to the surface onto which Devonian sediments were being deposited. In East Greenland, which made up a coherent landmass with Scandinavia during the formation of the Caledonian mountains, Permian marine sediments cover a peneplain eroded into the Caledonian rocks.
We know fairly well how collisional mountain ranges form, like the Alps, Himalayas and Andes. We do not, however, understand why there are mountains along many extensional (passive) continental margins. The present-day mountain chain along the western margin of Scandinavia (the Scandes) is one example and others are found in East and West Greenland, northeast Brazil, western India and southeast Australia. They occur in all climate zones, from the Arctic to the tropics. These mountain ranges have several features in common: the presence of elevated plateaux such as the high plain of Hardangervidda in Norway (see Figs 1 and 3), erosional truncation of the sedimentary strata along the oceanward side of the mountain ranges and a steeper slope on their oceanward side than towards the landward side. There is a debate raging about the age of these mountains.
Together with Paul Green and Johan Bonow (who are experts in thermochronology and geomorphology, respectively), we have published a paper in the Journal of the Geological Society, London, where we show evidence that the topography of the mountains in southernmost Norway is geologically young. We reached that conclusion by combining observations of landscape and geology with analysis of fission tracks in grains of the mineral apatite (see Box 1). Here we outline our key arguments.

Southern Norway was buried below thick covers of rock 23 Ma
Our study area covers a region south of Bergen and west of Oslo that makes up the southern half of an elongated, dome-shaped mountainous massif (Fig. 1). An extensive plain about 1200 m above sea level (masl), known as Hardangervidda, occupies the central part of this massif ( Fig. 3(a)). The bedrock slopes downwards from Hardangervidda towards the coast in the west, south and southeast.
Analysis of our apatite fission-track analysis data (AFTA data; see Box 1) shows that this part of Norway was deeply buried below a cover of rocks in early Miocene times (23 Ma). At this time, a phase of uplift started to affect Scandinavia, north and east of the Kattegat, that lifted the geological strata above sea level and thereby exposed them to erosion that removed many hundreds of metres of this cover. Because temperature increases steadily with depth below the surface, removal of rock by erosion leads to cooling of the underlying rock layers, and this can be detected by AFTA.
AFTA data from 27 rock samples from southernmost Norway combined with results from offshore boreholes and from samples from southern Sweden show that the present-day surface began to cool due to erosion between 23 and 21 Ma (in the early Miocene; Fig. 4).

Fig. 1.
Elevation of southernmost Norway. The mountain plateau, Hardangervidda, at c. 1200 masl, is clearly visible. Note the big northsouth and northwest-southeast trending, glacially over-deepened valleys that probably originated in the Miocene as river valleys feeding sediments to the delta in Denmark (Fig. 4). AFTA: apatite fission-track analysis. masl, metres above sea level; LF, Lysefjorden. Inset map: DK, Denmark; NO, Norway; SE, Sweden. Big river systems, originating in Norway and Sweden, began to deposit sediments in Denmark at the same time, showing that the uplifting area was being eroded. The AFTA data thus tell us when the surface of Earth was uplifted and started to cool because of erosion. The analysis also gives us the temperature of a rock sample when the cooling began. We know that the temperature in the Earth's interior increases with depth, by some 15-30°C/km depending on the rock properties, so we are able to estimate the thickness of the cover rocks that were once present.
We analysed rock samples from the coast and from the interior. The results show that samples from today's coast were at about 60°C when cooling began 23 Ma, whereas the results from the interior indicate that rocks on the surface of Hardangervidda at an elevation of 1200 m were somewhat colder. In both cases, the palaeotemperatures can be explained by burial below a cover of insulating rocks.
We also used the variation in determined palaeotemperatures across southern Norway to estimate the thickness of the rocks that have been removed since 23 Ma (Fig. 5(a)). We find that about 1500 m of sediments covered the coastal areas around Stavanger and Bergen (where Jurassic sediments are preserved in a fracture zone), while about 750 m of Caledonian metamorphics and younger strata covered the rocks that today are exposed on Hardangervidda.
In other words: The high plain of Hardangervidda did not exist 23 Ma (in the early Miocene).

Box 1 Apatite fission-track analysis
The Norwegian mountains consist of rocks that were formed at very high temperatures, deep below the surface of the Earth. This implies that the bedrock has cooled from those high temperatures over geological time as the kilometre-thick cover rocks were eroded until the rock became exposed at the surface. We can investigate this cooling/erosion history by analysing fission tracks in grains of the mineral apatite from a rock sample.
Apatite fission-track analysis (AFTA) is a method for determining thermal histories of rocks at temperatures less than 130°C. The method depends on analysis of radiation damage features ('fission tracks') in apatite grains because apatite contains radioactive uranium: when a uranium atom spontaneously splits into two positively charged fragments, the fragments will be repelled through the lattice and create a linear damage zone consisting of displaced atoms.
The number of tracks in unit area of a grain surface depends on the uranium content, the time over which tracks have accumulated and the distribution of track lengths in the sample. Tracks are formed with a fixed, initial length, but once formed they shorten at a rate which depends on temperature. As the temperature increases, all tracks are reduced to the same length regardless of when they were formed.
This process is irreversible, so if the temperature drops, all tracks formed up to that time are effectively 'frozen' at the length attained at the maximum temperature. Tracks that form after cooling are longer, due to the lower temperatures. Thus, at the end of a history involving heating and cooling, a sample will contain two populations of tracks; shorter tracks formed up to the onset of cooling and longer tracks formed after cooling. The proportion of short to long tracks will reflect the time of cooling relative to the total time over which tracks have been retained, while the mean length of the shorter component of tracks reflects the maximum palaeotemperature attained prior to cooling. It is thus possible to estimate timing and magnitude of the erosion that results from a phase of uplift of the Earth's surface.  Hilly basement terrain at an elevation of about 900 m, south of Lysefjorden. Kvernafjellet to the left, rises 190 m above the lake in the foreground. The hilly relief was shaped by weathering processes in the warm and humid climate in Jurassic times when the bedrock was exposed 175 Ma (Fig. 5). Photo locations in Fig. 1.

Hardangervidda: a plain graded to base level by Miocene rivers
Hardangervidda is a relatively flat plateau at an elevation of about 1200 masl (Figs 1 and 3). However, at the beginning of the Miocene, a thick cover laid on top of the rocks that today are exposed on Hardangervidda. This cover was removed by river erosion, which began to cut into the landscape as a consequence of the early Miocene uplift. The erosion continued for millions of years until Hardangervidda had been graded to the well-defined plain, which is seen today.
At the time when Hardangervidda had been eroded to a plain, it must have been close to either a regional resistant level or the adjacent ocean. Since there is no resistant level in Norwegian geology, this implies that the plain must have been close to sea level by the time of its final formation ( Fig. 5(b)) and must have been raised to its present elevation of 1200 m afterwards (Fig. 5(c)). This conclusion also implies that today's inclined slopes between Hardangervidda and the coast were buried below sediments and that the plain at the level of Hardangervidda must have cut across those sediments too.
Today, deep valleys and fjords cut into the edge of Hardangervidda (Fig. 6). The deeper parts of these valleys are clearly formed by glacial erosion, but remains of V-shaped river valleys are still present high on the valley sides. The preservation of these valley shoulders provides evidence that the initial incision of the uplifting plains was by rivers, in which the resulting fluvial valleys were then over-deepened by glaciers and that the glacial erosion of the plateaus was insignificant.
In other words: Hardangervidda started to form during the Miocene as part of a vast plain. The plain extended across the basement rocks of Hardangervidda and also across the pile of sediments that then covered the present mountain slopes. The plain was uplifted to its present altitude during the Pliocene and Pleistocene, but the part of the plain that extended across the sediments has been eroded away.

Hilly basement relief: remnants of a landscape from the time of dinosaurs
Our AFTA data also show that an earlier phase of uplift and erosion affected Scandinavia in the Middle Jurassic, 175 Ma, which resulted in the formation of an extensive erosion surface across basement rocks (Fig. 7). That episode has left a lasting impact on the landscape.
Scandinavia's climate at that time was sub-tropical, warm and humid. Water from the warm rain seeped into fractures in the basement and transformed the feldspars in granitic rocks into clays, particularly kaolin, resulting in the formation of a so-called saprolite (a Fig. 4. Early Miocene sediments in Denmark. a. Palaeogeographical reconstruction of the delta and delta plain created by huge amounts of sediments coming from Norway and Sweden to Denmark (NO, SE and DK, respectively). For many millions of years prior to the early Miocene (23 Ma), sea covered all of Denmark. After that time, the sea became filled with deltaic sediments transported by braided rivers originating in an uplifted area to the north in present-day southern Norway and Sweden. b. A thick layer of sand and gravel primarily originating from Norway, deposited by a major river system during the early Miocene. c. The river systems (blue lines) that transported sediment to Denmark in the early Miocene. Their location has been determined by analysing heavy minerals in the deltaic sediments and matching them with the chemistry of possible source areas of the sediment. (Images (a) and (b) courtesy of Erik S. Rasmussen.) (Figs 3(b) and 7(c)). Our AFTA results show that these rocks were buried below a thick cover prior to the onset of early Miocene uplift ( Fig. 5(a)). We therefore infer that Late Jurassic and younger sediments covered the bedrock along the coast between Stavanger and Bergen 23 Ma. This cover protected the fracture valleys and hilly relief until they were re-exhumed after later episodes of uplift.
In other words: The basement rocks were exposed and weathered in Jurassic times and subsequently buried below a protective cover of sediments. Today, we see the remnants of this weathered landscape as hilly relief and fracture valleys at elevations up to 1000 masl.
The mountains of southernmost Norway: uplifted Miocene plains above re-exposed Jurassic landscapes Hardangervidda was formed by river erosion to a plain near sea level during the Miocene. Today it lies 1200 masl (Fig. 5c). We infer that it must have been raised to this level at some time after the Miocene. Our data from Norway do not define the timing of this late uplift, but AFTA data from deep wells in Denmark provide evidence for uplift and erosion that began about 5 Ma, in the early Pliocene. Seismic profiles show images of large quantities of sediments that started to prograde westwards away from Scandinavia around that time. That phase of erosion could not be due to glacial action because it began long before the onset of the Ice Age, 2.7 Ma, an observation confirmed by the presence of the remains of pre-glacial river valleys on the flanks of the deep glacial valleys (Fig. 6) and fjords.
We therefore conclude that the southern Scandes were uplifted in the early Pliocene. The deep valleys that were eroded during this uplift and the subse-'rotten rock'). Clay saprolites occur at many localities in Scandinavia; for example at Porsgrunn where kaolin has been mined to manufacture porcelain (Fig. 1). The weathering of the basement rocks took place mostly in the fracture zones (Fig. 7a) and the basement between these zones was affected much less by the weathering processes. Subsequent erosion washed some of the clays out of the fracture zones, leaving fracture valleys and rounded hills (hilly relief) formed where fracture valleys intersected.
The fracture valleys and hilly relief were subsequently buried below a sedimentary cover when the sea transgressed Scandinavia in Late Jurassic and Cretaceous times (160 Ma, Fig. 7b). We know the timing because seismic data offshore shows hilly relief buried below Upper Jurassic and younger sediments and because a remnant of the Jurassic cover was discovered in a fracture that was penetrated while tunnelling in the Bergen area. These sediments contain organic material whose maturity shows that it has been heated to about 60°C.
We can see fracture valleys and hilly relief on the slopes leading down from Hardangervidda to the coast where the hills appear as hundreds of small islands Ma. Uplift that began in the early Miocene has led to deep erosion by rivers and to the formation of a flat landscape near sea level. The present-day Hardangervidda (red line), is part of this lowlying landscape where basement rocks are exposed (see Fig. 3(a)). c. Present day. Renewed uplift that began in the early Pliocene (5 Ma) has raised Hardangervidda to its present elevation of about 1200 masl. The sedimentary cover above the basement rocks along the coast has been eroded, and the basement hilly relief that had formed in Middle Jurassic times (175 Ma) is now re-exposed (see Fig. 3(b)).
quent Ice Age amplified the effect of the initial uplift by removing a huge load of rocks from the land areas and isostatic compensation for the loss of this load caused further uplift.
The late uplift phase in the Pliocene raised the sediment-covered slopes below Hardangervidda to well above sea level, so that rivers could erode their cover and wash it into the sea, leaving the hard bedrock of Hardangervidda as a remnant of the plain that once extended across all of southern Norway. The hilly relief formed by weathering in the Middle Jurassic was exhumed to the surface. Deep valleys and fjords have cut into the uplifted landscape, but there are sufficient remnants of the Mesozoic terrain to map its presence over wide areas.
In other words: A long history involving phases of erosion, burial and uplift explains the contrasts between the plains of Hardangervidda, the hilly relief on the slopes of the rock massif and the deep valleys, which cut the massif.

New dimensions to old conclusions
The notion that the Norwegian mountains are relatively young is not new. The Norwegian geologist Hans Reusch argued for this hypothesis in 1901. He highlighted the contrast between the elevated plains and the deep valleys. He referred to the plains as the palaeic (old) surface because he considered that rivers andlater on-glaciers must have incised into an older landscape defined by the elevated plains. Reusch therefore concluded that the plains had been lifted to their present elevation in more recent geological time.
We have added several new dimensions to the old story: Multiple episodes of uplift and subsidence have shaped the landscape since the collapse of the Caledonian mountains. Uplift, erosion and weathering in Middle Jurassic times were followed by subsidence and burial. Finally, two phases of uplift and erosion affected the region in the early Miocene and early Pliocene. Hardangervidda was formed as part of an extensive plain near sea level by uplift and erosion starting in the Miocene. The plain was afterwards raised to its present elevation, but the only parts of the plain that remain today are where it cut across hard rocks forming the present-day mountain plateau. The sloping bedrock surface between Hardangervidda and the coast, which was buried below a thick sedimentary cover until this late stage, was re-exposed and the hilly relief, the result of weathering processes almost 200 Ma, was revealed.
Are the Scandes remnants of a 400-Myr-old mountain chain or did they reach their present elevation only after uplifts during the last 20 Myr? We have presented observations and arguments in favour of Fig. 6. Deep incision a by river valley followed by over-deepening by glacial erosion. a. View from the summit of Gaustatoppen into the Rjukan valley. The mountain plateau of Hardangervidda (h) can be seen on both sides of the valley. The valley sides consist of slopes at two different angles; an upper slope (r, outlined in orange) that is less steep than the lower slope (g, outlined in green). b. Cross-section through the Rjukan valley. The extension (red dashed line) of the lesssteep upper slopes (r) shows that they are remnants of a V-shaped river valley that has been overdeepened by a glacier to form the present-day U-shaped lower valley (g). This configuration of glacially deepened river valleys (and fjords) is common in southern Norway. Incision of the river valleys below Hardangervidda took place during uplift of Hardangervidda after 5 Ma, but before glacial erosion during the Pleistocene. Location of Rjukan is shown on Fig. 1.  Fig. 7. Formation and preservation of a hilly relief in basement rocks. a. Middle Jurassic, 175 Ma. Basement rocks are exposed at the surface and the climate is warm and humid (also in Norway). Rainwater seeps into the fractures and chemically alters feldspars in the basement rocks that are transformed into a deeply weathered rock, a saprolite. b. Late Jurassic, 160 Ma. Most of the saprolite is gone leaving fracture valleys with a hilly terrain between them. The ocean has extended across the region and sediments have accumulated above the bedrock protecting the hilly relief from further erosion. c. Present day. The protective cover has been eroded after a final phase of uplift, leading to reexposure of the hilly relief terrain. The hills are typically about 200 m high (see Fig. 3(b)). Remnants of the saprolite remain in the fracture zones between the hills.