Early Holocene occurrence of loess in Folldal,east-central southern Norway: Identification,origin and palaeoclimatic significance

Abstract

Identified for the first time in Norway, windblown yellowish-brown dust in Folldal, east-central southern Norway, fulfils physical and geochemical criteria for sensu stricto loess, including a high carbonate content. Two extant and one relict marl pond located on low-carbonate rocks are investigated. An allochthonous origin for the marl ponds is suggested and explained by the deposition of suspended clastic carbonate as loess by aeolian processes. The scattered geographical distribution of loess and the need for a calcareous source area suggest a W-SW provenance from a relict valley sandur/existing flood plain in upper Grimsdalen. The mean grain size (~22.4 µm) of the aeolian silt is in the finer range of average loess, explained by the up to 25–30 km long transport and approximately 250–300 m airborne uphill move from the suggested source area to the marl ponds. Based on age-depth modelling using 9 AMS radiocarbon dates, loess deposition occurred from ca. 10,390 to 9780 cal. yr BP (610 years). Two prominent periods of loess accumulation occurred from ca. 10,390 to 10,190 (200 years) and ca. 10,020–9950 (70 years) cal. yr BP. Loess deposition results from W-SW wintertime winds in a dry to semi-arid climate. In southern Norway, W-SW winds in winter are associated with a positive mode of the North Atlantic Oscillation (NAO). From source to deposition, the response time to erode, transport and accumulate loess is suggested to be seasonal to 1 year. An immediate increase in organic production followed loess deposition as early initiation of a warmer and wetter Holocene climatic optimum.

Keywords

climatic optimum early Holocene Folldal loess marl ponds North Atlantic Oscillation (NAO)Norway prevailing wind direction provenance Preboreal-Boreal transition windblown dust

Introduction

Aeolian sediments and related landforms reflect variations in atmospheric circulation patterns on continental (Lancaster, 1990; Renssen et al., 2007) to more regional or local scale (Björck and Clemmensen, 2004; Nielsen et al., 2016a, 2016b). Compared to many temperate and subarctic parts of the world like the European Sand Belt (Isarin et al., 1997; Kasse, 1997, 2002; Koster, 1988; Tolksdorf and Kaiser, 2012), however, only scattered areas of windblown deposits are described from Fennoscandia (Lancaster et al., 2016).

Windblown deposits and related landforms occur in Finnish Lapland (Clarke and Käyhkö, 1997; Käyhkö et al., 1999; Matthews and Seppälä, 2014; Seppälä, 1971, 1981, 1995, 2004; Vliet-Lanoë et al., 1993), Swedish Lapland (Seppälä, 1972) and in central Sweden (Agrell and Hultman, 1971; Alexanderson et al., 2016; Alexanderson and Bernhardson, 2016, 2019; Alexanderson and Fabel, 2015; Bergqvist, 1981; Bernhardson et al., 2019; Bernhardson and Alexanderson, 2017, 2018; Högbom, 1913, 1923; Hörner, 1927; Stevens et al., 2022). Along the southwestern coast of Sweden, several investigations have been carried out (Björck and Clemmensen, 2004; de Jong et al., 2006, 2007; Kylander et al., 2013, 2016, 2018; Sjöström et al., 2022).

Many sites with windblown material occur in SE Norway (Klemsdal, 1969). Among these, the aeolian dune fields at Starmoen and in the Jømna valley east of Elverum were investigated more thoroughly (Alexanderson and Henriksen, 2015; Klemsdal, 2010). However, the origin of the dune fields at Starmoen and in the Jømna valley as primarily aeolian has recently been challenged and reinterpreted by Hansen et al. (2020) as the result of catastrophic megafloods from the drainage of ice-dammed lake Nedre Glåmsjø (Longva and Bakkejord, 1990; Longva and Thoresen, 1991). Several aeolian sites exist along the coast of Norway (Klemsdal, 1969). Among these studies have been carried out in Finnmark (Samuelsen, 1935; Sjögren, 2009), Andøya (Nielsen et al., 2016a), Langøya in Vesterålen (Nielsen et al., 2016b), Jæren (Prøsch-Danielsen and Selsing, 2009; Selsing and Mejdahl, 1994) and at Lista (Høyland, 1974).

First described by Leonhard (1824), loess is terrestrial clastic sediment. Formed by the accumulation of windblown dust, loess is predominantly composed of silt-sized grain fractions (0.02–0.05 mm) with smaller amounts of incorporated fine sand and clay (Muhs, 2013; Pye, 1995; Smalley et al., 2011; Smalley and Marković, 2019; Újvári et al., 2016; Vandenberghe, 2013; Vandenberghe et al., 2018). For loess to form, three fundamental time-dependent requirements are necessary to fulfil: (1) a sustained source of dust, (2) available wind energy to transport the dust, and (3) a suitable accumulation site/trap (Albani et al., 2015; Muhs, 2013; Pye, 1995; Seppälä, 2004). Sources of loess may be local, neighbouring or distant, and grain-size distribution appears to shift towards finer grains with increasing distance from the source area (Muhs, 2013; Pye, 1995; Vandenberghe et al., 2018). Loess is usually homogenous, loosely cemented by calcium carbonate (CaCO₃), and often yellowish brown or buff by colour. In addition to a characteristic CaCO₃ content of 5–35%, loess typically consists of 50–70% quartz (SiO₂) and 10–20% feldspars and micas (Albani et al., 2015; Editors of Encyclopaedia Britannica, 2010; Muhs, 2013; Pye, 1995). The composition of loess is more complicated than that of aeolian sand as the mineralogy reflects the geology of the source rocks (Seppälä, 2004; Stevens et al., 2007). Postdepositional processes also make it challenging to find a commonly accepted definition of loess (Smalley et al., 2011; Sprafke and Obreht, 2016). During the Quaternary, loess primarily formed in periglacial environments, but it also occurs in semi-arid parts of some lowland deserts and on the margins of high mountain ranges (Frechen, 2011; Pye, 1995).

Except for windblown dust in Iceland and some coastal areas, the occurrence of aeolian silt is not usual, and soils consisting of true loess are extremely rare in NW Europe (Hjulström, 1955). However, Högbom (1899) described the modern formation and deposition of ‘loess’ related to the (artificial/accidental) drainage of lake Ragundasjön in Indalsälven, Norrland, in AD 1796, and this may be the first mention of loess in Sweden (Hjulström, 1955).

Inland dunes are ‘ dunes presently found in the inland and which were formed by inland winds ’ (Alexanderson et al., 2016). Bonäsheden in Dalarne (Högbom, 1923) and Brattforsheden in Värmland (Hörner, 1927) (Figure 1) are significant Swedish inland dunes. At Bonäsheden, aeolian silt (‘ flygmo ’ in Swedish) occurs W-SW of the dune field (Alexanderson and Fabel, 2015; Hjulström, 1955), and at Brattforsheden windblown silt has been mapped W-SW of the dune field by Hörner (1927). According to Alexanderson and Fabel (2015), the latter sediments are loess-like deposits, aeolian silt without a carbonate content (Pécsi, 1990; Seppälä, 2004), consisting of massive coarse silt and fine sand. Stevens et al. (2022) confirmed and extended the previous mapping of the aeolian silt at Bonäsheden and Brattforsheden and described the essential sediment characteristics using scanning electron microscopy (SEM), grain size, mineral magnetism and geochemistry. Despite lacking the characteristic carbonate content, the aeolian dust was termed loess by Stevens et al. (2022). Aeolian silt of suggested Late Glacial age occurs at two sites in the southern Swedish uplands NE sector (Agrell and Hultman, 1971). The windblown silt likely originates from glaciofluvial sediments mapped SW of the aeolian deposits.

Figure 1.

Overview map of southern and central Scandinavia showing study area and sites taken into account when discussing regional implications of the occurrence of early Holocene loess in Folldal.

Mention of loess in the scientific literature from Norway is near non-existent, but Björlykke (1916, 1918) discuss the possible occurrence of loess in silty soils at Romerike (‘ Romeriksmjele ’) NE of Oslo. However, Romeriksmjele is fine flood sediments of late Preboreal age deposited as part of the final drainage of ice-dammed lake Nedre Glåmsjø (Longva and Bakkejord, 1990; Longva and Thoresen, 1991).

The shift from glacial to interglacial conditions occurred at the Younger Dryas-Preboreal transition in NW Europe. This shift dates to 11,530 +40/−60 cal. yr BP at Kråkenes on the outmost coast of western Norway (Figure 1) (Gulliksen et al., 1998). During the first 500 years after the Younger Dryas-Preboreal transition, an abrupt and distinct rise in July temperature of 8–10°C occurred at Kråkenes (Birks et al., 2000; Birks and Ammann, 2000). However, summer (July) temperatures close to present values first took place at about 10,000 cal. yr BP (Birks and Ammann, 2000; Nesje et al., 1991; Nesje and Kvamme, 1991.

In central Scandinavia, the final transition from glacial to interglacial conditions occurred during the late Preboreal/early Boreal (ca. 10,500–9500 cal. yr BP). During this period, the large ice-dammed lakes S-SE of the main watershed drained (Longva and Bakkejord, 1990; Longva and Thoresen, 1991; Stroeven et al., 2016; and references therein), and the pine-tree limit as a proxy of July/tetraterm (June-September) temperature obtained the Holocene maximum altitude from about 9500 to 9000 cal. yr BP (Aas and Faarlund, 1988; Dahl and Nesje, 1996; Kullman, 1980, 1995; Moe, 1979; Paus and Haugland, 2017; and references therein). However, the timing and causes of annual to seasonally prevailing wind patterns (Bernhardson et al., 2019; Bernhardson and Alexanderson, 2017, 2018; Sundborg, 1955) and hydroclimate (Digerfeldt, 1988; Digerfeldt et al., 2013) related to this significant shift in Scandinavia are not fully known or understood.

The main objectives of this study are to verify and date the possible occurrence and deposition of loess in Folldal, east-central southern Norway (Figure 1), and to discuss its origin and local and regional (synoptic scale) climatic implications. In addition, the results are discussed against relevant data reflecting hydroclimate and wind patterns in southern and central Sweden.

Study area

Folldal is a ca. 70 km long and generally broad valley from Hjerkinn on Dovrefjell to Alvdal in Østerdalen, Innlandet County, east-central southern Norway (Figure 1). The W-NW to E-SE orientated valley has a thick bedrock cover of superficial sediments dominated by till and large deposits of glaciofluvial, fluvial and glaciolacustrine sediments (Sollid and Carlson, 1980; Sørbel and Tolgensbakk, 2006; Thoresen and Follestad, 1999). The lack of over-deepened bedrock basins, common in Norway because of glacial erosion, is striking in Folldal. Water-filled kettle holes, however, exist in glaciofluvial terraces in Folldal/Grimsdalen (Sollid and Carlson, 1980) (Figure 2). Among these, Butjønna, Kroktjønna and Røtjønna at Stormoen are calcareous lakes based on the occurrence of charophytes (Figure 2) (Klepsland, 2007; Larsen et al., 2014; Miljødirektoratet Naturbase Faktaark, 2013). Next to Kroktjønna and Røtjønna at Stormoen, there are several peat-filled kettle holes. A core from a kettle hole named Stormoen Rus 15-01 is studied in detail (Figures 2 and 3a). There are no observations of colluvial activity or inlets in the catchment of the kettle hole.

Figure 2.

Overview map of study area with key sites and place names. Geographical extent of calcareous relict valley sandur in upper Grimsdalen is indicated. Overflow gaps (OGs) with altitude for meltwater drainage to upper Grimsdalen during the deglaciation and related calcareous bedrock bands (modified from Bjerkgård et al., 2002; Nilsen and Wolff, 1989) influencing on the relict valley sandur are shown. To visualise the impact of calcareous material in the study area the reported observations of the calcicolous plant mountain aven (Dryas octopetala L.) are shown (Artsdatabanken, 2020). A-A′ represents the schematic transect in the conceptual model from source area to loess accumulation in Figure 8.

Figure 3.

(a) Photo showing the relict marl pond (kettle hole) at Stormoen which is now filled up with peat. The Russian core Rus 15-01 was retrieved from central parts of the kettle hole. (b) Photo showing the shallow extant marl pond Lomslåtjønne seen from south. (c) Photo showing the shallow extant marl pond Myrlitjønne (our name) seen from south. (d) Photo showing details from the shallow bottom of the extant marl pond Myrlitjønne. The tracks seen on the photo are suggested to be after a moose (Alces alces L.). All photos: S.O. Dahl.

From the watershed towards Gudbrandsdalen at an altitude of 1160–1112 m to the west, Grimsdalen is a ca. 50 km long tributary valley which coalesces with Folldal at Stormoen (alt. ca. 660 m). The valley has a general W-SW to E-NE orientation (Figure 2). The only exception is a few km in the middle part with a NW to SE direction to pass Streitkampen (mountain) (alt. 1214 m). The upper ca. 14 km is narrower and more V-shaped, while the remaining valley is generally wide and without over-deepened bedrock basins. Exposed bedrock is typical in the uppermost V-shaped part and between the lower and middle Grimsdalen. The remaining valley has a high bedrock coverage of superficial sediments. Up to 50 m high terraces dominated by glaciofluvial and glaciolacustrine sediments surround the valley bottom along river Grimsa in the lower part. In the middle, significant accumulations of glaciofluvial sediments dominate along the southern valley side. Upper Grimsdalen has a ca. 11 km long stretch with a relatively flat valley bottom (altitudinal range from ca. 960–880 m) with terraces along Grimsa. These terraces consist of glaciofluvial, glaciolacustrine and fluvial sediment, and high-lying parts are suggested remnants after a relict glaciofluvial valley sandur. The occasionally active low-lying terraces along the river represent a modern flood plain (modified from Sollid and Carlson, 1980). Sediment input through meltwater overflow gaps from W-SW and N-NW during the last deglaciation likely contributed to the relict sandur in upper Grimsdalen. Just E-NE of Streitkampen, and between Folldal and Grimsdalen, two small and shallow ponds named Lomslåtjønne and Myrlitjønne (our name) have been investigated as part of this study (see Figures 2 and 3b–d). There are no observations of colluvial activity or inlets in the catchments of the two ponds.

The bedrock in Folldal and Grimsdalen is strongly affected by local allochtons (nappes) from the Caledonian orogen (500–400 Ma). Most of the area south of Folldal and the lower and middle parts of Grimsdalen consists of metasedimentary rocks from the middlemost allochthon with sandstone, quartzite, some conglomerate, and others in places. To the north of central Folldal and lower and middle Grimsdalen, and on both sides of upper Grimsdalen, the uppermost allochthon dominates. This allochthon consists of an ophiolite complex with metasedimentary and metavolcanic rocks, including phyllite, mica schist, marble, graphite schist, calcareous mica schist, calc-silicate schist, and others (Norges Geologiske Undersøkelse, 2021). Based on Nilsen and Wolff (1989) and Bjerkgård et al. (2002), Figure 2 shows the extent of calcareous phyllite and phyllite/calcareous schist in the catchment of Grimsdalen and central/upper Folldal. Note the calcareous rocks surrounding upper Grimsdalen, which are particularly interesting for this study.

Located within the ‘Rondane dry valleys’ north of the Rondane mountains, the climate in middle/upper Folldal and tributary valleys is continental. The mean annual air temperature of 0.7°C reflects relatively warm summers and frigid winters (Station no. 0871 Sørnesset, alt. 739 m) (Aune, 1993), and most of the mean annual precipitation of 364 mm occurs during summer (Station no. 910, Foldal, alt. 709 m) (Førland, 1993). The valley bottom of upper Grimsdalen, at an altitude of about 960–880 m, has a periglacial environment. The mean annual air temperature for the standard period 1991–2020 is between −2°C and 0°C (senorge.no, 2024a), and annual precipitation is less than 500 mm (senorge.no, 2024b). Hence, the climate reflects the location of Folldal/Grimsdalen in the rainshadow for the prevailing westerlies in Scandinavia and is representative of the regional climate in northern Innlandet. However, the leeward position north of the Rondane mountains reinforces the local rainshadow in Folldal/Grimsdalen for much of the occasional precipitation from S-SE. The present predominant wintertime wind direction in Rondane is from SW (Bjørbæk, 1993).

Methods

The core Stormoen Rus 15-01 was retrieved using a Russian corer and was stored dark and chilly at 4°C. The core was first visually inspected, photographed and logged in the sediment laboratory. After this, geochemical analysis was performed using an Itrax XRF (X-ray fluorescence) core scanner (Croudace et al., 2006). First, the measurement was run using a Chrome (Cr) X-ray tube every 200 μm (n = 2895) with power set to 30 kV and 40 mA and with an exposure time of 10 se. Then, following the procedure of Kylander et al. (2011), the XRF results were normalised by incoherent and coherent scattering (inc+coh) to account for density changes and water content of the sediments. Next, the core was scanned every 0.2 cm with a CoreSusc MK III magnetic susceptibility corelogger (n = 301).

Organic- and inorganic content, as well as dry bulk density (DBD), was measured every 0.5 cm (1 cm³) throughout the core by the loss-on-ignition (LOI) method (Dean, 1974; Heiri et al., 2001). The samples were first weighed for wet bulk density and dried at 105°C for 12 h before being weighed for dry bulk density (g/cm³) and water content (%). Next, the samples were ignited at 550°C for 1 h, cooled in a desiccator, and weighed. LOI⁵⁵⁰ was calculated based on the percentage change between the ignition residue and dry weight. Afterwards, the inorganic carbonate (LOI⁹⁵⁰) was determined every 5 cm by heating the samples a second time to 950°C for 1 h and multiplying the ash residue by a factor of 1.36 (Bengtsson and Enell, 1986; Heiri et al., 2001).

Grain size analyses were examined continuously between 439 and 404 cm in Stormoen Rus 15-01, as well as four samples from Myrlitjønne and Lomslåtjønne, using a Malvern Mastersizer 3000 laser diffraction particle size analyser with a Hydro LV unit (Sperazza et al., 2004). Most samples (408–439 cm, n = 32, from Stormoen Rus 15-01, and n = 2 from Myrlitjønne) were run without pre-treatment due to low organic content (2–20%). However, because of the relatively high organic content (20–60%), the interval 404–407 cm (n = 4) in Stormoen Rus 15-01 and the samples from Lomslåtjønne (n = 2) was digested in a 30% H₂O₂ aqueous solution before the analysis. The grain size statistics are calculated using Gradistat v.8 (Blott and Pye, 2001).

The mineralogy in Stormoen Rus 15-01 was identified and compared with deposits from Lomslåtjønne and Myrlitjønne by visual description, grain size analyses and content of LOI⁹⁵⁰ inorganic carbonate. Finally, the geochemical composition was analysed by use of qualitative x-ray diffraction (XRD) (Epp, 2016. Selected samples for XRD analyses were initially dried and ground to a fine powder using a mortar and pestle. A Bruker D8 Advance Eco X-ray diffractometer with a copper (Cu) anode (1.54 Å) was used, and individual samples were measured at 40 kV and 25 mA using a Bragg-Brentano geometry theta/2 with a beam angle setting of 3°–90° and exposure time of 7 min 42 s. The results were analysed using DIFFRAC.SUITE EVA.

Nine samples of organic material from Stormoen Rus 15-01 were submitted for ¹⁴C-dating. First, slices of sediment (0.5–2 cm thick) from the core were sieved through 125 μm using distilled water. Next, terrestrial plant remains were identified and isolated under the microscope before being dried overnight (50°C) and stored dark and chilly in sealed vials. Subsequently, samples were dated by accelerator mass spectrometry (AMS) radiocarbon dating at the Poznań Radiocarbon Laboratory in Poland before using the IntCal20-calibration curve to calibrate the dates to calendar years (Reimer et al., 2020) using Calib 8.2 (Stuiver et al., 2021). Finally, the age-depth modelling of the dates followed the routine in Clam 2.4 (Blaauw, 2010) using the open-source software R.

The age-depth model was calculated based on 6 of the radiocarbon-dated samples using smooth-spline interpolation (smoothing 0.2), calibration curve ‘IntCal.20C14’ for the Northern Hemisphere and 0.95 probability. The radiocarbon dates at 404, 416 and 420 cm depth are older than underlying ages and are therefore marked as outliers in the model. Outliers in aeolian erosional and depositional environments like this are expected because of shifting strength and direction of the wind.

Bacon (Blaauw and Christen, 2011) was evaluated for the age-depth modelling. There are, however, features like marked shifts in the accumulation rates in the investigated core Rus 15-01 that represent substantial challenges for this type of Bayesian analysis. You need to define the expected mean accumulation rate and depth resolution as priors in Bacon, and changing these values will effectively decide the dates which are treated as outliers. Because of this it is not possible to construct complete and reasonable age-depth models for Rus 15-01 in Bacon. Clam, however, uses a less advanced statistical approach, and it allows us to use the assumption of gradually changes in the sedimentation rate (through a smooth spline). Hence, and even if Clam may underestimate when there are large gaps between dates (Trachsel and Telford, 2017), this is the preferred choice for the age-depth modelling.

A field investigation of the relict valley sandur and modern flood plain in upper Grimsdalen is an element of this study. As part of this investigation, a compilation of superficial sediments and available data linked to the extent of calcareous rocks and related flora for central/upper Folldal and the entire catchment of Grimsdalen has been performed (Figure 2).

Results

Lithostratigraphy and chronology of Stormoen Rus 15-01

A core from the early deglaciation (Stormoen Rus 15-01) was retrieved by a Russian corer from a peat-covered kettle hole at Stormoen (Figures 2 and 3a). The core has been visually described and analysed for various sediment parameters and geochemical composition (Table 1). The chronology depends on 9 AMS radiocarbon-dated samples, eight using terrestrial plant macrofossils and one bulk sample (Table 2). The lithostratigraphy and key sediment parameters of Stormoen Rus 15-01 are shown in Figure 4, and grain size distributions of individual samples are displayed in Figure 5. An example of qualitative XRD analysis is presented in Figure 6, and age-depth modelling, including sedimentation rates, appears in Figure 7.

Table 1.

Description of the loess/marl localities Lomslåtjønne, Myrlitjønne and Stormoen Rus 15-01.

Locality	Altitude (m)/Size Past => Present (ha)/Water depth (m)	Geographical coordinates	Bedrock	^cDistance (max/min) to suggested source of allochthonous CaCO₃	Qualitative XRD identification of calcite	Thickness of calcareous sediment (m)	Max. % CaCO₃ (LOI⁹⁵⁰)	Mean grain size (µm)	Remarks
Lomslåtjønne	750 m/1.115 => 0.588 ha/⩾ 1.0 m	62°6′52.97″N 10°0′23.80″E	Tuffitic greenschist^a	19.2–11.3 km	Yes	>0.5 m^d	44.94^d	24.34 (2)^f	Extant, clastic marl pond(?)^g
Myrlitjønne	775 m/0.303 => 0.143 ha 0.303 ha ⩾0.6 m	62°6′54.18″N 9°59′55.35″E	Tuffitic greenschist^a	18.8–10.9 km	Yes	>0.5 m^e	52.17^e	21.73 (2)^f	Extant, clastic marl pond^g
Stormoen Rus 15–01	666 m //0.161 ha/4.37–4.02 m	62°8′18.70″N 10°10′11.50″E	Greenschist/Amphibolite^b	28.9–20.1 km	Yes	∼0.35 m	51.68	22.4 (27)^f	Peat covered, relict, extant, clastic marl pond^g

Bjerkgård et al. (2002).

Nilsen and Wolff (1989).

Distance from upstream start of calcareous relict sandur along valley bottom towards E-NE in upper Grimsdalen.

Measured at 70 cm depth next to margin – most likely more in central parts of ponds.

Measured on surface sediments close to margin. Most likely higher values deeper in the stratigraphy and in central parts of lake (Pentecost, 2009, and references therein).

Number of samples.

Suggested to be extant/relict clastic marl ponds according to definition by Pentecost (2009). See text for further discussion.

Table 2.

AMS radiocarbon dates from Stormoen Rus 15-01 from Stormoen, Folldal.

Depth from surface (cm)	Lab. no.	Sample type	AMS¹⁴C age	^aCal. yr BP±2 σ	^aCal. yr BP±1 σ	^aMedian probability cal. yr BP	^aMedian of most probable 2 σ (in brackets)cal. yr BP	Remarks
401.5–402.5	Poz-135575	Needles of Pinus sylvestris, plant remains.	8770 ± 50	10,116–9552	9896–9681	9773	9735 ± 183 (0.939)	End of loess deposition
404.0	Poz-81343	Bulk sediments	10,480 ± 60^b	12,670–12,100	12,620–12,200	12,480	12,550 ± 120 (0.618)	0.5 mg C. Not reliable – suggested hard water effect.
411.5	Poz-81344	Catcin scales from Betula sp.	8820 ± 50	10,156–9679	10,115–9716	9871	9825 ± 146 (0.71)	Main loess deposition culminates and is gradually reduced
416.0	Poz-81345	Small piece of twig, plant remains	9300 ± 50	10,652–10,292	10,576–10,412	10,495	10,463 ± 132 (0.909)	Somewhat reduced loess deposition. Re-sedimentation?
416.5–418.5	Poz-81346	Twig w/bark	8890 ± 50	10,191–9776	10,154–9909	10,019	10,038 ± 153 (0.915)	Somewhat reduced loess deposition
420.0	Poz-81348	Seeds from Betula sp., undetermined plant remains	9190 ± 60^b	10,500–10,230	10,480–10,250	10,360	10,370 ± 130 (1.0)	0.7 mg C. Loess deposition. ¹⁴C plateau/Re-sedimentation?
426.5	Poz-81349	Twigs w/bark	9150 ± 50	10,490–10,226	10,375–10,237	10,319	10,330 ± 104 (0.938)	Start of main period of loess deposition
436.5	Poz-81350	Dryas octopetala leaves, twigs	9190 ± 60^b	10,500–10,230	10,480–10,250	10,360	10,370 ± 130 (1.0)	0.3 mg C. Start of loess deposition
440.5	Poz-81351	Dryas octopetala leaves	9320 ± 70^b	10,700–10,280	10,650–10,410	10,520	10,490 ± 210 (1.0)	0.2 mg C. No loess deposition

Calibrated by use of IntCal20/Calib 8.2 (Stuiver et al., 2021).

Calibation is rounded to nearest 10 year for samples with standard deviation greater than 50 years.

Figure 4.

Lithostratigraphy of Stormoen Rus 15-01 with organic and inorganic content based on loss-onignition (LOI) 550°C and 950°C, respectively, dry bulk density (DBD), magnetic susceptibility (MS), XRF analysis represented by Ca/Inc+coh and Ti/Inc+coh adjusted for density changes and water content of the sediments (Kylander et al., 2011) and grain size analysis (the interval between 404 and 407 were digested in a 30% H₂O₂ aqueous solution prior to the analysis because of high organic content). Radiocarbon dates and units based on sediment parameters are shown. Note the yellowish-brownish colour of units C-F and partly unit G suggested to be representative of loess. The lithostratigraphy is further discussed in the text.

Figure 5.

Continuous grain size analysis with 1 cm resolution between 439 and 404 cm depth in Stormoen Rus 15-01. All samples between 439 and 405 cm depth were run without pre-treatment, whereas the uppermost 4 samples (407–404 cm depth) were digested in a 30% H2O2 aqueous solution prior to the analysis because of high organic content. Unit A consists of very fine to fine sand and represents the early deglaciation, unit B consists of very coarse silt and mark the transition from early deglaciation to an environment dominated by loess deposition, units C, D, E and F consist of coarse silt (~22.4 μm) and are dominated by loess deposition, and unit G consists of coarse silt which is influenced by increased organic production.

Figure 6.

Qualitative XRD analysis showing the identification of calcite (Ca) as part of CaCO₃ from Stormoen Rus 15-01 and surface sediments from Myrlitjønne.

Figure 7.

Age-depth model based on 6 AMS radiocarbon dates (marked with blue) using terrestrial plant macrofossils from Stormoen Rus 15-01. Two inverted dates (marked with red) based on terrestrial plant macrofossils are not used in the age-depth modelling. One date on bulk sediment is not shown (see Table 2 for overview of radiocarbon dates). The samples are dated at Poznań Radiocarbon Laboratory. The IntCal20 calibration curve was used for calibrating the samples to calendar years (Reimer et al., 2020) using Calib 8.2 (Stuiver et al., 2021). The dates were then applied for age-depth modelling based on Clam 2.4 (Blaauw, 2010) using the open-source software R. Sedimentation rates for the different units as defined in Figure 4 show a slow sedimentation rate (0.035 cm/yr) during the final deglaciation (unit A), and an increase (0.062 cm/yr) during the transition (unit B) towards an environment dominated by loess deposition. A high sedimentation rate of loess (0.161 cm/yr) is recorded in unit C. This is followed by a marked decrease (0.030 cm/yr) in unit D, before the sedimentation rate of loess again increases (0.055 cm/yr) in unit E. The very high sedimentation rate (1.00 cm/yr) in unit F is attributed to a marked increase in organic production. See text for further discussion.

Unit A, 444.5–437.0 cm depth. As reflected by LOI⁵⁵⁰, the organic content is very low (1–3%), DBD is between 1.2 and 1.0 g/cm³ and MS (10⁻⁶) is from 1.1 to 0.4. Calcium (Ca) adjusted for density changes and water content (Ca/inc+coh) is very low, as is the inorganic content based on LOI⁹⁵⁰. Titanium (Ti) adjusted for density changes and water content (Ti/inc+coh) has mean values of about 0.3 but is weakly declining towards the top of the unit. The colour of unit A is dark brown to more blackish towards the top, and grain sizes are from fine to very fine sand. Based on the age-depth modelling, unit A lasts from ca. 10,500 to 10,390 cal yr BP (~110 years). The mean sedimentation rate of the unit is 0.036 cm/yr (Figures 4, 5 and 7).

Unit B, 437.0–435.5 cm depth. LOI⁵⁵⁰ values are low (2–4%), DBD is about 1.1 g/cm³, and MS (10^-6) drops from about 0.4 to near zero. Calcium (Ca/inc+coh) increases from near zero to about 12. The calcium increase is reflected in the inorganic content, with values of LOI⁹⁵⁰ rising from close to zero to ca. 20%. Titanium (Ti/inc+coh) drops from about 0.3 to 0.1, and the mean grain size drops from very fine sand to very coarse/coarse silt at the top. The colour of unit B is black. Age-depth modelling indicates a duration of the unit from ca. 10,390 to 10,365 cal yr BP (~ 25 years), and it has a mean sedimentation rate of 0.055 cm/yr (Figures 4, 5 and 7).

Unit C, 435.5–422.0 cm depth. Values of LOI⁵⁵⁰ are rising upwards from about 3 to ca. 4–8%, DBD drops from ca. 0.8 to 0.4 g/cm³, and MS (10⁻⁶) is close to zero. Calcium (Ca/inc+coh) increases from 12 to stabilising between 14 and 15, but with a maximum at the upper end of the unit. The same general pattern for this unit is observed for the inorganic content as reflected by values of LOI⁹⁵⁰, whereas titanium (Ti/inc+coh) drops to values close to zero. Grain size distribution remains stable within coarse silt. The colour of unit C is yellowish brown with some sporadic speckles/thin layers, which are more brownish. Based on age-depth modelling, unit C lasts from about 10,365–10,190 cal yr BP (~ 175 years), with a mean sedimentation rate of 0.158 cm/yr (Figures 4, 5 and 7).

Unit D, 422.0–417.0 cm depth. LOI⁵⁵⁰ is between ca. 6 and 15%, DBD drops from ca. 0.6 to 0.4 g/cm³ and MS (10⁻⁶) is close to zero. Calcium (Ca/inc+coh) drops from a value of ca. 15 to a level between 10 and 11, as also recorded in LOI⁹⁵⁰. Titanium (Ti/inc+coh) is close to zero, whereas mean grain size is stable within coarse silt. The colour of unit D is yellowish brown, but with some thin brownish layers towards the top. Age-depth modelling indicates that unit D lasts from ca. 10,190–10,020 cal yr BP (~170 years) with a mean sedimentation rate of 0.030 cm/yr (Figures 4, 5 and 7).

Unit E, 417.0–414.5 cm depth. Values of LOI⁵⁵⁰ vary between ca. 8.6 and 22% but is lowest in the upper part. DBD varies from 0.3 to 0.5 g/cm³ but is highest in the upper end, and MS (10⁻⁶) is close to zero. Calcium (Ca/inc+coh) increases from ca. 11 to 14, and a similar pattern is observed in values of LOI⁹⁵⁰, whereas Titanium (Ti/inc+coh) is close to zero. Mean grain size has only minor fluctuations within coarse silt. Unit E is yellowish brown, but brownish layers become more abundant towards the top. The age-depth modelling indicates a duration of unit E from ca. 10,020 to 9950 cal yr BP (~70 years) with a mean sedimentation rate of 0.036 cm/yr (Figures 4, 5 and 7).

Unit F, 414.5–409.0 cm depth. LOI⁵⁵⁰ rises from ca. 7.5% to 13.6%, DBD drops from ca. 0.5 to 0.45 g/cm³, and MS (10⁻⁶) is close to zero. Calcium (Ca/inc+coh) gradually drops from ca. 14 to 7. This decrease is also observed in inorganic carbon based on LOI⁹⁵⁰, whereas Titanium (Ti/inc+coh) remains close to zero. Mean grain size remains stable within coarse silt, and the unit’s colour transitions gradually from yellowish brown to more brownish at the top. Based on age-depth modelling, the unit lasts from ca. 9955 to 9850 cal yr BP (~105 years), with a mean sedimentation rate of 0.055 cm/yr (Figures 4, 5 and 7).

Unit G, 409.0–402.0 cm depth. LOI⁵⁵⁰ values rise quickly from 13.6% to 85.7%, DBD values drop from ca. 0.45 to 0.1 g/cm³, and MS (10⁻⁶) is close to zero. Calcium (Ca/inc+coh), inorganic carbon based on LOI⁹⁵⁰ and Titanium (Ti/inc+coh) are all close to zero. Grain size analyses are not available (not possible to perform). The colour of unit G is black. Based on age-depth modelling, the unit lasts from ca. 9850 to 9780 cal yr BP (~70 years) with a mean sedimentation rate of 0.095 cm/yr (Figures 4, 5 and 7).

Unit H, from 402.0 cm depth and upwards. LOI⁵⁵⁰ values are stable above 85%, DBD is close to 0.1 g/cm³, and MS (10⁻⁶) is close to zero. Calcium (Ca/inc+coh), inorganic carbon based on LOI⁹⁵⁰ and Titanium (Ti/inc+coh) are all close to zero. Grain size analyses are not available (not possible to perform). The colour of unit G is black. Age-depth modelling is unavailable, but the unit starts at ca. 9780 cal. yr BP (Figure 4).

Qualitative XRD analysis, inorganic carbonate and grain size

Samples from units C and E in Stormoen Rus 15-01, sub-surface sediments from Lomslåtjønne and surface sediments from Myrlitjønne were analysed using qualitative XRD, and the amount (%) of inorganic carbonate (CaCO₃) in the same samples were obtained using LOI⁹⁵⁰. Table 1 shows the results of the analyses and thickness estimates of calcareous sediments. Calcite (CaCO₃) is identified at all three sites using XRD (see Figure 6). Based on LOI⁹⁵⁰, units C and E in Stormoen Rus 15-01 have maximum values of inorganic carbonate of 50.72% and 51.68%, respectively. Along the margin of Lomslåtjønne and beneath 55 cm of peat, analysis shows a maximum content of inorganic carbonate at 70 cm depth of 44.94%. From surface sediments near the margin of Myrlitjønne, the amount of inorganic carbonate based on LOI⁹⁵⁰ is 52.17–51.99%.

The thickness estimate from Stormoen Rus 15-01 is reliable. In contrast, figures from Lomslåtjønne and Myrlitjønne are only minimum estimates based on observations from the surrounding bog and the margin of the small ponds, respectively.

The investigated kettle hole at Stormoen (core Rus 15-01) has a peat cover (Figure 3a). It differs markedly from the small, open, shallow ponds Lomslåtjønne (Figure 3b) and Myrlitjønne (Figure 3c and d), which, seen from above, has a brownish-yellowish and whitish-yellowish colour, respectively. The calcareous sediments at all three sites are well sorted, contain significant amounts of quartz, and have mean grain size within coarse silt (~22.4 µm) (Figures 5 and 6). Based on Nilsen and Wolff (1989), the bedrock at Lomslåtjønne, Myrlitjønne and Stormoen Rus 15-01 consists of greenschist and amphibolite, and according to Bjerkgård et al. (2002), the bedrock at Lomslåtjønne and Myrlitjønne is tuffitic greenschist. The bedrock at the Stormoen site is hard to map, as it has a thick cover of glaciofluvial deposits dominated by feldspar and quartz originating south of Folldal.

No survey for charophytes to evaluate if the ponds are calcareous exists for Lomslåtjønne and Myrlitjønne. However, charophytes like delicate stonewort ( Chara virgata Kützing) are detected in the larger kettle-hole ponds Butjønna, Kroktønna and Røtjønna next to Stormoen Rus 15-01 (Klepsland, 2007; Larsen et al., 2014; Miljødirektoratet Naturbase Faktaark, 2013). Among these, Kroktjønna and Røtjønna are groundwater-fed ponds in thick glaciofluvial sediments dominated by quartz and feldspar. In contrast, Butjønna may have some input of calcareous material by a local stream passing through phyllite/calcareous schist before entering the pond (Figure 2).

Relict valley sandur and modern floodplain in upper Grimsdalen

Calcareous ponds/peat-filled kettle holes in non-calcareous superficial sediments/rocks need a source. The geographical distribution led to an investigation of a relict valley sandur/modern floodplain in upper Grimsdalen towards W-SW (Figure 2). The valley sandur was built out in an ice-dammed lake with erosion basis at an altitude of ca. 880 m. It covered the central and upper parts of Grimsdalen, and based on optically stimulated luminescence (OSL) dates adjusted for mean water content and depth from Svartdalsbekken (Figure 2), the formation of the valley sandur took place between ca. 17.2 and 13.8 ka (modified from Bøe et al., 2007). Quartz/feldspar-rich sediments deposited from S-SE dominate the downstream central part of the ice-dammed lake. In contrast, calcareous material of both local and more distant origin strongly influences the relict valley sandur/modern floodplain in upper Grimsdalen. Distant input of calcareous material originates from meltwater streams through overflow gaps from W-SW and N-NW during the deglaciation (Figure 2) Mountain aven ( Dryas octopetala L.) is a calciphilous plant (Mossberg and Stenberg, 2003: 291). Hence, the widespread occurrence of this plant demonstrates the presence of calcareous material on the relict valley sandur/modern floodplain and related rock outcrops in upper Grimsdalen (Artsdatabanken, 2020) (Figure 2).

Discussion

Occurrence, origin and timing of loess deposition in Folldal

Yellowish brown clastic sediments in Folldal, consisting of homogenous coarse silt (~22.4 µm) and some clay and fine sand, are very well sorted and contain high amounts of carbonate and quartz. (Figures 4 –6). As the geographical distribution can only be explained by aeolian processes (Figures 2 and 8), these sediments are suggested to be loess ( sensu stricto ) deposited in a periglacial environment. Similar sediments from the Swedish inland dunes Bonäsheden in Dalarne and Brattforsheden in Värmland (Figure 1) are described by Stevens et al. (2022) as loess. The aeolian dust at Bonäsheden and Brattforsheden lack the characteristic carbonate content, however, and the silt from the latter area has previously been described by Alexanderson and Fabel (2015) as loess-like deposits (Pécsi, 1990; Seppälä, 2004). Reworking or decalcifying the aeolian silt (Lehmkuhl et al., 2021; Schaetzl and Loope, 2008) at Bonäsheden and Brattforsheden may explain the missing carbonate content at these inland dunes. The described clastic aeolian sediment from Folldal, however, fulfil the physical and geochemical criteria for loess (Muhs, 2013; Pécsi, 1990; Pye, 1995; Seppälä, 2004; Smalley et al., 2011; Sprafke and Obreht, 2016). Hence, loess ( sensu stricto ) has been documented in Norway and possibly the Scandinavian Peninsula for the first time.

Figure 8.

Conceptual model for loess accumulation from the source area in upper Grimsdalen to the accumulation area (sink) in Folldal.

The present distribution of loess in Folldal is very scattered. Primarily the windblown dust occurs in sediment traps like peat-covered kettle holes with a bottom below the local groundwater table or in small and shallow ponds. The present distribution and the need for a calcareous source area strongly indicate a W-SW origin for the loess. Hence, the provenance of the loess is suggested to be a relict valley sandur/existing flood plain in upper Grimsdalen just north of the Rondane mountain massif (Figures 2 and 8). About 25–30 km downwind from the suggested source area, an about 35 cm long core section of loess from a kettle hole at Stormoen in Folldal (Stormoen Rus 15-01) is investigated. Based on age-depth modelling using 9 AMS radiocarbon dates (Table 2, Figure 7), loess deposition took place between ca. 10,390 and 9780 cal. yr BP (ca. 610 years). The highest loess accumulation appears to have occurred in two main periods from ca. 10,390 to 10,190 (200 years) and from ca. 10,020 to 9950 (70 years) cal. yr BP. A reduced deposition rate separates the two periods from 10,190 to 10,020 cal. yr BP (170 years) (Figures 7 and 9).

Figure 9.

Deposition of loess during the early Holocene in Folldal. Note the somewhat reduced accumulation of loess in between the two deposition maxima of loess during Folldal Loess Event (FLE) 1 and 2.

Considering a suitable source area and available winds, the response time to erode, transport and accumulate loess in Folldal is suggested to be short, season to 1 year.

The mean grain size, however, is in the finer range of average loess (Muhs, 2013; Pye, 1995; Vandenberghe et al., 2018). The ca. 28.9–20.1 km long transportation and approximately 250–300 m airborne uphill move across Streitkampen from the suggested source area to the accumulation site (trap) at Stormoen may explain why the windblown dust is in the finer range of average loess (Table 1, Figure 8).

Marl lakes/ponds

Marl (mergel in German) lakes/ponds can be defined based on the composition of the lacustrine sediments and the amount of dissolved carbonate in the water. Morgan and Britton (1977) defined marl lakes as those lakes whose waters, on average, contained ⩾100 mg/L dissolved calcium carbonate (CaCO₃), whereas Kelts and Hsu (1978) defined them as lakes whose lacustrine sediments contained >60 % CaCO₃. Pentecost (2009), however, argued for a change in the definition of marl lakes as lakes with calcium carbonate as the main component (>50 % CaCO₃). Despite Schnurrenberger et al. (2003) and others advocating to replace the term «marl» with «carbonate mud», lacustrine environments with sediments containing more than 50 % CaCO₃ are defined in this paper according to Pentecost (2009) as marl lakes/ponds.

The amount of CaCO₃ in Lomslåtjønne based on LOI⁹⁵⁰ analysis on a sediment sample beneath (70 cm depth) the present peaty margin of the pond is 44.94% (Table 1). In contrast, the maximum amount of CaCO₃ based on LOI⁹⁵⁰ recorded in surface samples of lacustrine sediments near the margin of Myrlitjønne and beneath the peat at Stormoen Rus 15-01, are 52.17% and 51.68%, respectively. The figure from Lomslåtjønne, however, is suggested to underestimate the actual values with depth and in central parts of the pond because of more interbedding of organic material along the margins (Pentecost, 2009, and references therein). Hence, according to Pentecost (2009), Lomslåtjønne and Myrlitjønne are suggested to be extant marl ponds, and the kettle hole at Stormoen Rus 15-01 is a relict marl pond.

Based on the occurrence of charophytes like delicate stonewort ( Chara virgata Kützing) (skjørkrans in Norwegian), the ponds Kroktjønna and Røtjønna next to Stormoen Rus 15-01 are calcareous (Klepsland, 2007; Larsen et al., 2014; Miljødirektoratet Naturbase Faktaark, 2013). The presence of charophytes indicates that modern biological marl production may contribute to, or partially replace, the original clastic marl from the last deglaciation/early Holocene in these ponds (Pentecost, 2005: 302, and references therein). However, surveying for charophytes as an indicator of modern biological marl production in Lomslåtjønne and Myrlitjønne has not been performed.

The composition of the local bedrock at the investigated sites cannot explain the high content of CaCO₃ in the lacustrine sediments. Hence, these sediments must have an allochthonous origin and are likely to be clastic marl due to the input of suspended carbonate (Pentecost, 2005: 302). The accumulation of allochthonous carbonate may result from aeolian, colluvial or (glacio-) fluvial activity (Pentecost, 2009; Schnurrenberger et al., 2003). However, only input of suspended carbonate as loess by aeolian processes can explain the geographical distribution of the extant small and shallow marl ponds Lomslåtjønne and Myrlitjønne, the relict peat-filled marl pond Stormoen Rus 15-01 and the calcareous ponds Kroktjønna and Røtjønna (Figures 2 and 8).

Climate during loess deposition in Folldal

Strong wind is most likely to occur in Folldal during autumn and winter. As reflected in the present spatial distribution of leeward accumulation of snow, the predominant snow-bearing wind direction in Rondane is from SW (Bjørbæk, 1993). Exposed towards W-SW, the broad watershed (wind gap) area between Grimsdalen and Gudbrandsdalen may have played an essential role in directing and strengthening wintertime W-SW winds towards Folldal. Hence, tracking the loess in Folldal to its source area indicates a provenance and erosion/transport by strong W-SW winds from the calcareous relict sandur/existing flood plain in upper Grimsdalen. The occurrence of plant remains (e.g. leaves) from mountain aven ( Dryas octopetala ) in loess at Stormoen Rus 15-01 supports this wind direction. This plant is not growing at Stormoen, but the plant remains may have a likely origin from the sizeable present population at the calcareous areas of upper Grimsdalen (Figures 2 and 8). Birch seeds ( Betula sp.) found in the loess support the assumption of erosion and deposition of the loess during autumn and winter.

The present dry wintertime climate with slight snow was likely even more pronounced during loess deposition in Folldal. Gradually becoming more distinct upwards, however, thin organic laminae recorded in the loess sequence at Stormoen in Folldal may indicate relatively warm but likely very dry (semi-arid) summers with access to water as the limiting factor for organic production. Loess accumulation ended ca. 9780 cal. yr BP. After that, a rapid increase (a few decades?) in organic production which quickly reached stable LOI values of about 85% (Figure 4), took place. The probable result of a shift to a warmer and wetter climate, this transition is suggested to represent the initiation of the local Holocene climatic optimum. An AMS radiocarbon date on needles from scots pine ( Pinus sylvestris ) from the transition demonstrates that this conifer tree species had colonised the area (Table 2). Recording of the cooling event dated to 9700 cal. yr BP at Flåfattjørna (alt. 1110 m) N-NE of Folldal (Paus et al., 2019) is not identified in Stormoen Rus 15-01.

Predominantly W-NW winds linked to the deposition of aeolian sand and related landforms along a S-N transect in south-central Sweden have recently been reviewed by Bernhardson et al. (2019). Primarily based on OSL dating, most dune fields stabilised ca. 10.5–9.0 ka despite the local deglaciation along the transect taking place over almost 4000 years (Stroeven et al., 2016). Hence, regional environmental conditions with an unstable and dry early Holocene climate likely explain the prolonged dune activity followed by a nearly synchronous stabilisation. The lowest Holocene lake-level stands recorded between 10,500 and 9500 cal. yr BP in lake Bysjön (Digerfeldt, 1988) and lake Igelsjön (Digerfeldt et al., 2013) in south-central Sweden (Figure 1) are supporting dry conditions in Folldal. Probably due to a low precipitation/evaporation (P/E) ratio, the low groundwater level caused delayed plant colonisation of the dune fields before a vegetation cover made them more resilient to climate change.

OSL dates reflect the timing for the last aeolian sand transport prior to deposition/stabilisation. However, Bernhardson et al. (2019) found a peak in OSL ages around 10.0 ± 0.5 ka at sites along the S-N transect in south-central Sweden. At Bonäsheden, the largest continuous dune field in Sweden (Figure 1), Bernhardson and Alexanderson (2017) identified an apparent shift in the prevailing wind pattern from NW to W c. 10 ka. Local topography appears to have had little influence on the dune-forming regional winds. However, NW katabatic winds from the shrinking Scandinavian Ice Sheet likely impacted dune formation soon after the local deglaciation. In a later investigation of small dune fields in central Sweden, Bernhardson and Alexanderson (2018) could neither confirm nor reject the hypothesis. However, the loess deposition in Folldal by W-SW winds centred around 10,085 ± 305 cal yr BP (Figure 9) supports the hypothesis by Bernhardson and Alexanderson (2017). Strong W-SW wintertime winds are in southern Norway strongly linked to a positive mode of the North Atlantic Oscillation (NAO) (Nesje et al., 2000, and references therein).

Impact on permafrost and last remnants of the Scandinavian Ice sheet

Central parts of the Scandinavian Ice Sheet, including the Folldal area, are suggested to have been cold based during the Last Glacial Maximum (LGM), Younger Dryas and the following deglaciation (see Figure 4 in Stroeven et al., 2016, and references therein). Cold-based ice implies that the underlying soft sediments had sub-zero temperatures until the last ice remnants melted away. The present climate in Folldal is continental, with frigid and dry winters (Aune, 1993; Førland, 1993). However, the local loess deposition indicates a periglacial environment with even drier and windier conditions than at present. As a result, most permafrost in the predominantly soft-sediment-covered Folldal area (Sollid and Carlson, 1980) is likely to have survived until Late Preboreal/Early Boreal.

Summer (July) temperatures close to present values first took place in western Norway at about 10,000 cal. yr BP (Birks and Ammann, 2000; Nesje et al., 1991; Nesje and Kvamme, 1991), and coincides with loess deposition in Folldal centred about 10,085 ± 305 cal. yr BP (Figure 8). During and immediately after loess deposition in Folldal, a transition with widespread permafrost melting probably occurred before establishing an early Holocene steady-state landscape. Melting of the permafrost implies that most landforms with incorporated ice or frost-sensitive sediments experienced slope instability (Etzelmüller et al., 2020; Miesen et al., 2021) and most likely first stabilised about 10,000 cal yr BP. This steady-state landscape is, however, still influenced by stochastic episodes of intense snow melt or heavy spring/summer precipitation, which may cause major local and regional river floods and related colluvial processes (Bøe et al., 2006; Nesje et al., 2001).

The dry regional wintertime climate, as indicated by the loess deposition in Folldal between 10,390 and 9780 cal. yr BP probably had a marked impact on the winter balance (accumulation season) of the remaining Scandinavian Ice Sheet. Hence, the final drainage of the sizeable ice-dammed lake Nedre Glåmsjø in southeastern Norway, about 10,200 ± 200 cal yr BP (Høgaas and Longva, 2016), may have been influenced by this period of precipitation starvation.

Conclusions

Based on the presented data and discussion, the following main conclusions of local and regional importance are suggested:

In Folldal, east central southern Norway (Figure 1), windblown yellowish-brown dust fulfilling physical and geochemical criteria of loess, including a high carbonate content, is identified (Figure 4). The presence of sensu stricto loess is the first documented in Norway, and possibly the Scandinavian Peninsula.

The two extant marl ponds Lomslåtjønne and Myrlitjønne, and the relict marl pond Stormoen Rus 15-01 are investigated. The marl ponds are located on low-carbonate bedrock, and an allochthonous origin for the marl is suggested. The most likely mechanism explaining the marl ponds is the input of suspended clastic carbonate as loess by aeolian processes (Figures 2 and 8, Table 1).

The scattered geographical distribution of loess in Folldal and the need for a calcareous source area suggest an upper Grimsdalen provenance from a relict valley sandur/existing flood plain (Figures 2 and 8).

The mean grain size (~ 22.4 µm) (Figures 5 and 6) of the aeolian silt is in the finer range of average loess. The ca. 25–30 km long transport distance and approximately 250–300 m airborne uphill move across Streitkampen from the suggested source area to the sediment trap at Stormoen Rus 15-01 may explain the finer grain size (Figure 8).

Having a suitable source area and available winds, the response time to erode, transport and accumulate loess in Folldal is likely to be seasonal to 1 year.

Based on age-depth modelling using 9 AMS radiocarbon dates from Stormoen Rus 15-01 (Table 2, Figures 4 and 7), loess deposition took place between ca. 10,390 and 9780 cal. yr BP (ca. 610 years). The highest loess accumulation appears to have occurred in two main periods from ca. 10,390 to 10,190 (200 years) and from ca. 10,020 to 9950 (70 years) cal. yr BP. A reduced deposition rate of loess separates the two periods from 10,190 to 10,020 cal. yr BP (170 years) (Figures 7 and 9).

Loess deposition in Folldal took place during late autumn/wintertime. Together with minimal summertime organic production in kettle holes beneath the groundwater table, this is suggested to reflect very dry to semi-arid climatic conditions with access to water as the limiting factor. The dry climate is in accordance with the lowest Holocene lake-level stands recorded between 10,500 and 9500 cal. yr BP in south-central Sweden (Figure 1) (Digerfeldt, 1988; Digerfeldt et al., 2013).

Loess deposition in Folldal by W-SW winds about 10,085 ± 305 cal. yr BP agrees with an apparent shift from prevailing NW to W winds at Bonäsheden in central Sweden ca. 10 ka (Figure 1) (Bernhardson and Alexanderson, 2017).

The loess deposition in Folldal is related to prevailing W-SW winds during late autumn and wintertime. This wind direction is closely associated with a positive mode of the North Atlantic Oscillation (NAO) in southern Norway.

The Scandinavian Ice Sheet was probably cold-based in Folldal during the last deglaciation (Stroeven et al., 2016), implying subglacial frozen ground. During and just after loess deposition in Folldal, widespread permafrost melting probably occurred, followed by an immediate increase in organic production as early initiation of a warmer and wetter Holocene climatic optimum (Figure 4). The ice remnants damming the sizeable ice-dammed lake Nedre Glåmsjø may have been affected by precipitation starvation during loess deposition prior to the final drainage 10,200 ± 200 cal yr BP.

Footnotes

Acknowledgements

Rannveig Øvrevik Skoglund, Erlend Sporstøl Vikestrand and Matthew James Jenkin for assistance with Russian coring. Eirik Skjerdal Lysne, Matthew James Jenkin and Joseph Michael Buckby for assistance with laboratory analyses. Laboratory analyses were performed at EarthLab (226171/F50), Department of Earth Science, University of Bergen. Anne Bjune at Department of Biological Sciences, University of Bergen, quality assured terrestrial plant macrofossils used for AMS radiocarbon dating. Thanks to Carin Andersson Dahl for useful comments on an early draft of the paper and to two anonymous reviewers for their suggestions on the first submitted version.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: We thank The Meltzer Research Fund, project 102783117, for contributions to Svein Olaf Dahl for fieldwork and laboratory analyses. The authors thank the University of Bergen for financial support for publication of this article.

ORCID iDs

Svein Olaf Dahl

Pål Ringkjøb Nielsen

References

Aas

Faarlund

(1988) Postglacial forest limits in central south Norwegian mountains. Radiocarbon datings of subfossil pine and birch specimens. Norsk Geografisk Tidsskrift 42(1): 25–61.

Agrell

Hultman

(1971) Deposits of windblown silt in the northeastern part of the southern Swedish Uplands. Geologiska Föreningen i Stockholm Förhandlingar 93(1): 231–235.

Albani

Mahowald

Winckler

, et al. (2015) Twelve thousand years of dust: The Holocene global dust cycle constrained by natural archives. Climate of the Past 11(6): 869–903.

Alexanderson

Bernhardson

(2016) OSL dating and luminescence characteristics of aeolian deposits and their source material in Dalarna, central Sweden. Boreas 45(4): 876–893.

Alexanderson

Bernhardson

(2019) Late Glacial and Holocene sand drift in northern Götaland and Värmland, Sweden: Sediments and ages. GFF 141(2): 84–105.

Alexanderson

Bernhardson

Laliska-Nartiša

(2016) Aeolian activity in Sweden: An unexplored environmental archive. LUNDQUA Report 42: 1–53. Lund: Department of Geology, Lund University.

Alexanderson

Fabel

(2015) Holocene chronology of the Brattforsheden delta and inland dune field, SW Sweden. Geochronometria 42(1): 1–16.

Alexanderson

Henriksen

(2015) A short-lived aeolian event during the early Holocene in southeastern Norway. Quaternary Geochronology 30(part B): 175–180.

Artsdatabanken (2020) Reinrose Dryas octopetala L. Available at: https://www.artsdatabanken.no/Taxon/Dryas%20octopetala/103303 (accessed 12 November 2020). In Norwegian.

10.

Aune

(1993) Air temperature normals, normal period 1961-1990. DNMI-Rapport Nr. 02/93 Klima: 1–63. Oslo: Det Norske Meteorologiske Institutt.

11.

Bengtsson

Enell

(1986) Chemical analysis. In: Berglund

(ed.) Handbook of Holocene Palaeoecology and Palaeohydrology. Chichester: John Wiley & Sons Ltd, pp.423–451.

12.

Bergqvist

(1981) Svenska inlandsdyner. Översikt och förslag till dynreservat. /Ancient inland dunes in Sweden. Survey and proposals for dune reserves. Statens Naturvårdsverk. Rapport PM 1412: 1–111. In Swedish.

13.

Bernhardson

Alexanderson

(2017) Early Holocene dune field development in Dalarna, central Sweden: A geomorphological and geophysical case study. Earth Surface Processes and Landforms 42(12): 1847–1859.

14.

Bernhardson

Alexanderson

(2018) Early Holocene NW-W winds reconstructed from small dune fields, central Sweden. Boreas 47(3): 869–883.

15.

Bernhardson

Alexanderson

Björck

, et al. (2019) Sand drift events and surface winds in south-central Sweden: From the deglaciation to the present. Quaternary Science Reviews 209: 13–22.

16.

Birks

Ammann

(2000) Two terrestrial records of rapid climatic change during the glacial–Holocene transition (14,000– 9,000 calendar years BP) from Europe. Proceedings of the National Academy of Sciences 97(4): 1390–1394.

17.

Birks

Battarbee

Birks

HJB

(2000) The development of the aquatic ecosystem at Kråkenes Lake, western Norway, during the late glacial and early Holocene – A synthesis. Journal of Paleolimnology 23(1): 91–114.

18.

Bjerkgård

Schack Pedersen

Siedlecka

(2002) Berggrunnskart Folldal 1519 II, M 1:50.000, foreløpig utgave. Norges geologiske undersøkelse.

19.

Björck

Clemmensen

(2004) Aeolian sediment in raised bog deposits, Halland, SW Sweden: A new proxy record of Holocene winter storminess variation in southern scandinavia? Holocene 14(5): 677–688.

20.

Björlykke

(1916) Jordbunden paa Romerike. Tekst til jordbundskart over øvre og midtre Romerike (kartbladet Nannestad). Det Kongelige Selskap for Norges Vels Jordbundsutvalg. Jordbundsbeskrivelse nr. 14. Kristiania.

21.

Björlykke

(1918) Løssjord i Norge? Forhandlinger ved de skandinaviske naturforskeres 16. møte i Kristiania den 10.-15. juli : 511–517.

22.

Bjørbæk

(1993) Nasjonalatlas for Norge, Vind, Kartblad 3.1.8, M 1:3,5 mill. Statens Kartverk.

23.

Blaauw

(2010) Methods and code for ‘classical’ age-modelling of radiocarbon sequences. Quaternary Geochronology 5(5): 512–518.

24.

Blaauw

Christen

(2011) Flexible paleoclimate age-depth models using an autoregressive gamma process. Bayesian Analysis 6(3): 457–474.

25.

Blott

Pye

(2001) GRADISTAT: A grain size distribution and statistics package for the analysis of unconsolidated sediments. Earth Surface Processes and Landforms 26(11): 1237–1248.

26.

Bøe

A-G

Dahl

Lie

, et al. (2006) Holocene river floods in the upper Glomma catchment, southern Norway: A high-resolution multiproxy record from lacustrine sediments. Holocene 16(3): 445–455.

27.

Bøe

A-G

Murray

Dahl

(2007) Resetting of sediments mobilised by the LGM- ice-sheet in southern Norway. Quaternary Geochronology 2(1-4): 222–228.

28.

Clarke

Käyhkö

(1997) Evidence of Holocene aeolian activity in sand dunes from Lapland. Quaternary Science Reviews 16(3-5): 341–348.

29.

Croudace

Rindby

Rothwell

(2006) ITRAX: Description and evaluation of a new multi-function X-ray core scanner. Society Special Publication 267(1): 51–63.

30.

Dahl

Nesje

(1996) A new approach to calculating Holocene winter precipitation by combining glacier equilibrium-line altitudes and pine-tree limits: A case study from Hardangerjokulen, central southern Norway. Holocene 6(4): 381–398.

31.

Dean

(1974) Determination of carbonate and organic matter in calcareous sediments and sedimentary rocks by loss on ignition: Comparison with other methods. Journal of Sedimentary Petrology 44(1): 242–248.

32.

de Jong

Björck

Björkman

, et al. (2006) Storminess variation during the last 6500 years as reconstructed from an ombrotrophic peat bog in Halland, southwest Sweden. Journal of Quaternary Science 21(8): 905–919.

33.

de Jong

Schoning

Björck

(2007) Increased aeolian activity during humidity shifts as recorded in a raised bog in southwest Sweden during the past 1700 years. Climate of the Past 3(3): 411–422.

34.

Digerfeldt

(1988) Reconstruction and regional correlation of Holocene lake-level fluctuations in lake Bysjön, south Sweden. Boreas 17(2): 165–182.

35.

Digerfeldt

Björck

Hammarlund

, et al. (2013) Reconstruction of Holocene lake-level changes in lake Igelsjön, Southern Sweden. GFF 135(2): 162–170.

36.

Editors of Encyclopaedia Britannica (2010) Loess. Website name: Encyclopaedia Britannica. Encyclopaedia Britannica, Inc. Available at: https://www.britannica.com/science/loess (accessed 29 September 2020).

37.

Epp

(2016) X-ray diffraction (XRD) techniques for materials characterisation. In: Hübschen

Altpeter

Tschuncky

, et al. (eds) Materials Characterization Using Nondestructive Evaluation (NDE) Methods. Sawston: Woodhead Publishing, pp.81–124.

38.

Etzelmüller

Patton

Schomacker

, et al. (2020) Icelandic permafrost dynamics since the Last Glacial Maximum – Model results and geomorphological implications. Quaternary Science Reviews 233(106236): 1–15.

39.

Frechen

(2011) Loess in Europe: Guest Editorial. E&G Quaternary Science Journal 60(1): 3–5.

40.

Førland

(1993) Precipitation normals, normal period 1961-1990. DNMI-Rapport Nr. 39/93 Klima: 1–63. Oslo: Det Norske Meteorologiske Institutt.

41.

Gulliksen

Birks

Possnert

, et al. (1998) A calendar age estimate of the Younger Dryas-Holocene boundary at Kråkenes, Western Norway. Holocene 8(3): 249–259.

42.

Hansen

Tassis

Høgaas

(2020) Sand dunes and valley fills from preboreal glacial-lake outburst floods in south-eastern Norway – Beyond the aeolian paradigm. Sedimentology 67(2): 810–848.

43.

Heiri

Lotter

Lemcke

(2001) Loss on ignition as a method for estimating organic and carbonate content in sediments: Reproducibility and comparability of results. Journal of Paleolimnology 25(1): 101–110.

44.

Hjulström

(1955) Chapter I. The problem of the geographical distribution of windblown silt. An attempt of explanation. Geografiska Annaler 37(1-2): 86–93.

45.

Högbom

(1899) Om Ragundadalens geologi. Sveriges Geologiska Undersökning Serie C 182: 123 p. 2 maps. Summary in German.

46.

Högbom

(1913) Finiglaziale Flugsandfelder in Dalarne. Geologiska Föreningen i Stockholm Förhandlingar 35(7): 484–500.

47.

Högbom

(1923) Ancient inland dunes of northern and middle Europe. Geografiska Annaler 5: 113–243.

48.

Hörner

(1927) Brattforsheden ett värmlandskt randdeltakomplex och dess dyner. Geological Survey of Sweden. SGU C 342: 206 p. In Swedish.

49.

Høgaas

Longva

(2016) Mega deposits and erosive features related to the glacial lake Nedre Glomsjø outburst flood, southeastern Norway. Quaternary Science Reviews 151: 273–291.

50.

Høyland

(1974) Sand-shores, dunes and dune vegetation with examples from Lista, Vest- Agder County, SW Norway. Blyttia 32: 103–118.

51.

Isarin

RFB

Renssen

Koster

(1997) Surface wind climate during the Younger Dryas in Europe as inferred from aeolian records and model simulations. Palaeogeography Palaeoclimatology Palaeoecology 134(1-4): 127–148.

52.

Kasse

(1997) Cold-climate aeolian sand-sheet formation in north-Western Europe (c. 14–12.4 ka); a response to permafrost degradation and increased aridity. Permafrost and Periglacial Processes 8(3): 295–311.

53.

Kasse

(2002) Sandy aeolian deposits and environments and their relation to climate during the Last Glacial Maximum and lateglacial in northwest and central Europe. Progress in Physical Geography 26(4): 507–532.

54.

Käyhkö

Worsley

Pye

, et al. (1999) A revised chronology for Aeolian activity in subarctic Fennoscandia during the Holocene. Holocene 9(2): 195–205.

55.

Kelts

Hsu

(1978) Freshwater carbonate sedimentation. In: Lerman

(ed.) Lakes: Chemistry, Geology, Physics. New York: Springer-Verlag, pp.295–323.

56.

Klemsdal

(1969) Eolian forms in parts of Norway. Norsk Geografisk Tidsskrift 23(2): 49–66.

57.

Klemsdal

(2010) The eolian landforms and sediment in the valley of River Jømna, east of Elverum, south-east Norway. Norwegian Journal of Geography 64(2): 94–104.

58.

Klepsland

(2007) Kartlegging og vurdering av naturtyper og biologisk mangfold i Folldal kommune. BioFokus-rapport 2007-5: 1–42. In Norwegian.

59.

Koster

(1988) Ancient and modern cold-climate aeolian sand deposition: A review. Journal of Quaternary Science 3(1): 69–83.

60.

Kullman

(1980) Radiocarbon dating of subfossil Scots pine (Pinus sylvestris L.) in the southern Swedish Scandes. Boreas 9(2): 101–106.

61.

Kullman

(1995) Holocene Tree-limit and climate history from the Scandes Mountains, Sweden. Ecology 76(8): 2490–2502.

62.

Kylander

Ampel

Wohlfarth

, et al. (2011) High-resolution X-ray fluorescence core scanning analysis of Les echets (France) sedimentary sequence: New insights from chemical proxies. Journal of Quaternary Science 26(1): 109–117.

63.

Kylander

Bindler

Cortizas

, et al. (2013) A novel geochemical approach to paleorecords of dust deposition and effective humidity: 8500 years of peat accumulation at store Mosse (the “Great Bog”), Sweden. Quaternary Science Reviews 69: 69–82.

64.

Kylander

Martínez-Cortizas

Bindler

, et al. (2016) Potentials and problems of building detailed dust records using peat archives: An example from store Mosse (the “Great Bog”), Sweden. Geochimica et Cosmochimica Acta 190: 156–174.

65.

Kylander

Martínez-Cortizas

Bindler

, et al. (2018) Mineral dust as a driver of carbon accumulation in northern latitudes. Scientific Reports 8(1): 6876–6910.

66.

Lancaster

(1990) Palaeoclimatic evidence from sand seas. Palaeogeography Palaeoclimatology Palaeoecology 76(3-4): 279–290.

67.

Lancaster

Wolfe

Thomas

, et al. (2016) The INQUA Dunes Atlas chronologic database. Quaternary International 410: 3–10.

68.

Larsen

Fjeldstad

Gaarder

, et al. (2014) Kvalitetssikring og nykartlegging av naturtyper i Folldal kommune i 2013. Miljøfaglig utredning, rapport 2014-13: 1–51. In Norwegian.

69.

Lehmkuhl

Nett

Pötter

, et al. (2021) Loess landscapes of Europe – Mapping, geomorphology, and zonal differentiation. Earth-Science Reviews 215(103496): 1–42.

70.

Leonhard K. C.

von

(1824) Charakteristik der Felsarten – Dritte Abteilung. Heidelberg: Joseph Engelmann Verlag.

71.

Longva

Bakkejord

(1990) Iceberg deformation and erosion in soft sediments, southeast Norway. Marine Geology 92(1-2): 87–104.

72.

Longva

Thoresen

(1991) Iceberg scours, iceberg gravity craters and current erosion marks from a gigantic preboreal flood in southeastern Norway. Boreas 20(1): 47–62.

73.

Matthews

Seppälä

(2014) Holocene environmental change in subarctic aeolian dune fields: The chronology of sand dune re-activation events in relation to forest fires, palaeosol development and climatic variations in Finnish Lapland. Holocene 24(2): 149–164.

74.

Miesen

Dahl

Schrott

(2021) Evidence of glacier-permafrost interactions associated with hydro-geomorphological processes and landforms at Snøhetta, Dovrefjell, Norway. Geografiska Annaler Series A Physical Geography 103(3): 273–302.

75.

Miljødirektoratet Naturbase Faktaark (2013) Kroktønna-Røtjønna. Available at: https://faktaark.naturbase.no/?id=BN00104987 (accessed 23 February 2022). In Norwegian.

76.

Moe

(1979) Tregrense-fluktuasjoner på Hardangervidda etter siste istid. In: Nydal

Westin

Hafsten

, et al. (eds) Fortiden i Søkelyset. 14C Datering Gjennom 25 År. Trondheim: Laboratoriet for radiologisk datering, pp.199–208.

77.

Morgan

Britton

(1977) Open Waters. In: Ratcliffe

(ed) A Nature Conservation Review, Vol. 1. Cambridge: Cambridge University Press, pp.200–248.

78.

Mossberg

Stenberg

(eds) (2003) Den Nya Nordiska Floran/Gyldendals Store Nordiske Flora, Norwegian 2012/2014 edn. Oslo: Gyldendal.

79.

Muhs

(2013) Loess deposits: Origins and properties. In: Elias SA and Mock CJ (eds) Encyclopaedia of Quaternary Science, 2nd edition. Amsterdam: Elsevier, pp.573–584.

80.

Nesje

Dahl

Matthews

, et al. (2001) A c. 4500-yr record of river floods obtained from a sediment core in lake Atnsjøen, eastern Norway. Journal of Paleolimnology 25(3): 329–342.

81.

Nesje

Kvamme

(1991) Holocene glacier and climatic variations in westernNorway: Evidence for early Holocene glacier demise and multiple Neoglacial events. Geology 19(6): 610–612.

82.

Nesje

Kvamme

Rye

, et al. (1991) Holocene glacial and climate history of the Jostedalsbreen region, western Norway; evidence from lake sediments and terrestrial deposits. Quaternary Science Reviews 10(1): 87–114.

83.

Nesje

Dahl

(2000) Is the North Atlantic oscillation reflected in Scandinavian glacier mass balance records? Journal of Quaternary Science 15(6): 587–601.

84.

Nielsen

Dahl

Jansen

(2016a) Mid- to late Holocene aeolian activity recorded in a coastal dunefield and lacustrine sediments on Andøya, northern Norway. Holocene 26(9): 1486–1501.

85.

Nielsen

Dahl

Jansen

, et al. (2016b) Holocene aeolian sedimentation and episodic mass-wasting events recorded in lacustrine sediments on Langøya in Vesterålen, northern Norway. Quaternary Science Reviews 148: 146–162.

86.

Nilsen

Wolff

(1989) Geologisk kart over Norge, bergrunnskart over Røros og Sveg – 1:250000. Norges geologiske undersøkelse.

87.

Norges Geologiske Undersøkelse (2021) Bedrock Map of Norway 1:1 350000. Norwegian Geological Survey. Available at: https://www.ngu.no/upload/Publikasjoner/Kart/BGNorge_1350_000.pdf (accessed 10 December 2021).

88.

Paus

Haflidason

Routh

, et al. (2019) Environmental responses to the 9.7 and 8.2 cold events at two ecotonal sites in the Dovre mountains, mid-Norway. Quaternary Science Reviews 205: 45–61.

89.

Paus

Haugland

(2017) Early- to mid-Holocene forest-line and climate dynamics in southern Scandes mountains inferred from contrasting megafossil and pollen data. Holocene 27(3): 361–383.

90.

Pécsi

(1990) Loess is not just the accumulation of dust. Quaternary International 7-8: 1–21.

91.

Pentecost

(2005) Travertine. Berlin: Springer-Verlag.

92.

Pentecost

(2009) The marl lakes of the British Isles. Freshwater Reviews 2(2): 167–197.

93.

Prøsch-Danielsen

Selsing

(2009) Aeolian activity during the last 9200 calendar years BP along the southwestern coastal rim of Norway. AmS-Skrifter 21: 1–96.

94.

Pye

(1995) The nature, origin and accumulation of loess. Quaternary Science Reviews 14(7-8): 653–667.

95.

Reimer

Austin

WEN

Bard

, et al. (2020) The IntCal20 Northern Hemisphere radiocarbon age calibration curve (0–55 cal kBP). Radiocarbon 62(4): 725–757.

96.

Renssen

Kasse

Vandenberghe

, et al. (2007) Weichselian late pleniglacial surface winds over northwest and central Europe: A model–data comparison. Journal of Quaternary Science 22(3): 281–293.

97.

Samuelsen

(1935) Flyvesand og vinderosjon i Tanadalen og tilgrensende fjellstrøk. Norsk Geografisk Tidsskrift 5(3): 182–190.

98.

Schaetzl

Loope

(2008) Evidence for an eolian origin for the silt-enriched soil mantles on the glaciated uplands of eastern Upper Michigan, USA. Geomorphology 100(3-4): 285–295.

99.

Schnurrenberger

Russell

Kerry

(2003) Classification of lacustrine sediments based on sedimentary components. Journal of Paleolimnology 29(2): 141–154.

100.

Selsing

Mejdahl

(1994) Aeolian stratigraphy and thermoluminescence dating of sediments of late Holocene age from Sola, southwest Norway. Boreas 23(1): 92–104.

101.

Senorge.no (2024a) Temperature 1991-2020. Available at: https://www.senorge.no. Daily maps, 30yr, Temperature (accessed 13 April 2024).

102.

Senorge.no (2024b) Annual precipitation 1991-2020. Available at: https://www.senorge.no. Daily maps, 30yr, Precipitation (accessed 13 April 2024).

103.

Seppälä

(1971) Evolution of eolian relief of the Kaamasjoki-Kiellajoki river basin in Finnish Lapland. Fennia 104(1): 1–88.

104.

Seppälä

(1972) Location, morphology and orientation of inland dunes in northern Sweden. Geografiska Annaler Series A Physical Geography 54(2): 85–104.

105.

Seppälä

(1981) Forest fires as activator of geomorphic processes in Kuttanen esker-dune region, northernmost Finland. Fennia 159: 221–228.

106.

Seppälä

(1995) Deflation and redeposition of sand dunes in Finnish Lapland. Quaternary Science Reviews 14(7-8): 799–809.

107.

Seppälä

(2004) Wind as a Geomorphologic Agent in Cold Climates. Cambridge: Cambridge University Press.

108.

Sjögren

(2009) Sand mass accumulation rate as a proxy for wind regimes in the SW Barents Sea during the past 3 ka. Holocene 19(4): 591–598.

109.

Sjöström

Bindler

Martínez Cortizas

, et al. (2022) Late Holocene peat paleodust deposition in south-western Sweden - exploring geochemical properties, local mineral sources and regional aeolian activity. Chemical Geology 602(120881): 1–14.

110.

Smalley

Marković

(2019) Controls on the nature of loess particles and the formation of loess deposits. Quaternary International 502: 160–164.

111.

Smalley

Marković

Svirčev

(2011) Loess is [almost totally formed by] the accumulation of dust. Quaternary International 240: 4–11.

112.

Sollid

Carlson

(1980) Folldal. Beskrivelse til kvartærgeologisk kart 1:50 000 (1519 II) folldal. Description of Quarternary map 1:50,000 (1519 II). Norsk Geografisk Tidsskrift 34(4): 191–212.

113.

Sperazza

Moore

Hendrix

(2004) High-resolution particle size analysis of naturally occurring very fine-grained sediment through laser diffractometry: Research methods papers. Journal of Sedimentary Research 74(5): 736–743.

114.

Sprafke

Obreht

(2016) Loess: Rock, sediment or soil – What is missing for its definition? Quaternary International 399: 198–207.

115.

Stevens

Sechi

Tziavaras

, et al. (2022) Age, formation and significance of loess deposits in central Sweden. Earth Surface Processes and Landforms 47(14): 3276–3301.

116.

Stevens

Thomas

Armitage

, et al. (2007) Reinterpreting climate proxy records from late Quaternary Chinese loess: A detailed OSL investigation. Earth-Science Reviews 80(1-2): 111–136.

117.

Stroeven

Hättestrand

Kleman

, et al. (2016) Deglaciation of Scandinavia. Quaternary Science Reviews 147: 91–121.

118.

Stuiver

Reimer

(2021) CALIB 8.2 [WWW program]. Available at: http://calib.org (accessed 26 May 2021).

119.

Sundborg

(1955) Chapter II. Meteorological and climatological conditions for the genesis of aeolian sediments. Geografiska Annaler 37(1-2): 94–111.

120.

Sørbel

Tolgensbakk

(2006) Hjerkinn 1519 III – kvartærgeologisk kart M 1:50,000. Institutt for geofag, Universitetet i Oslo.

121.

Thoresen

Follestad

(1999) Alvdal Kommune, kvartærgeologisk kart i M 1:60000. Norges geologiske undersøkelse.

122.

Tolksdorf

Kaiser

(2012) Holocene aeolian dynamics in the European sand-belt as indicated by geochronological data. Boreas 41(3): 408–421.

123.

Trachsel

Telford

(2017) All age–depth models are wrong, but are getting better. Holocene 27(6): 860–869.

124.

Újvári

Kok

Varga

, et al. (2016) The physics of windblown loess: Implications for grain size proxy interpretations in Quaternary paleoclimate studies. Earth-Science Reviews 154: 247–278.

125.

Vandenberghe

(2013) Grain size of fine-grained windblown sediment: A powerful proxy for process identification. Earth-Science Reviews 121: 18–30.

126.

Vandenberghe

Sun

Wang

, et al. (2018) Grain-size characterization of reworked fine-grained aeolian deposits. Earth-Science Reviews 177: 43–52.

127.

Vliet-Lanoë

Seppälä

Käyhkö

(1993) Dune dynamics and cryoturbation features controlled by Holocene water level change, Hietatievat, Finnish Lapland. Geologie en Mijnbouw 72(2): 211–214.