Appearance of De Geer moraines in southern and western Finland — Implications for reconstructing glacier retreat dynamics

    Appearance of De Geer moraines in southern and western Finland — Implications for reconstructing glacier retreat dynamics Antti E.K. Ojala PII: DOI: Reference:

S0169-555X(15)30225-7 doi: 10.1016/j.geomorph.2015.12.005 GEOMOR 5458

To appear in:

Geomorphology

Received date: Revised date: Accepted date:

5 October 2015 11 December 2015 14 December 2015

Please cite this article as: Ojala, Antti E.K., Appearance of De Geer moraines in southern and western Finland — Implications for reconstructing glacier retreat dynamics, Geomorphology (2015), doi: 10.1016/j.geomorph.2015.12.005

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Appearance of De Geer moraines in southern and western Finland – implications for reconstructing glacier retreat dynamics Antti E.K. Ojala

RI P

T

Geological Survey of Finland P.O. Box 96, 02150 Espoo Finland E-mail: [email protected]

SC

Abstract

LiDAR digital elevation models (DEMs) from southern and western Finland were investigated to map

NU

and discriminate features of De Geer moraines, sparser and more scattered end moraines, and larger

MA

end moraine features (i.e., ice-marginal complexes). The results indicate that the occurrence and distribution of De Geer moraines and scattered end moraine ridges in Finland are more widespread

ED

than previously suggested. This is probably attributed to the ease of detecting and mapping these features with high-resolution DEMs, indicating the efficiency of LiDAR applications in geological and

PT

geomorphological studies. The variable appearance and distribution of moraine ridges in Finland

AC CE

support previous interpretations that no single model is likely to be appropriate for the genesis of De Geer moraines at all localities and for all types of end moraines. De Geer moraine appearances and interdistances probably result from a combination of the general rapidity of ice-margin recession during deglaciation, the proglacial water depth in which they were formed, and local glacier dynamics related to climate and terrain topography. The correlation between the varved clay-based rate of deglaciation and interdistances of distinct and regularly spaced De Geer moraine ridges indicates that the rate of deglaciation is probably involved in the De Geer ridge-forming process, but more thorough comparisons are needed to understand the extent to which De Geer interdistances represent an annual rate of ice-margin decay and the rapidity of regional deglaciation.

Keywords: De Geer moraine; end moraine; LiDAR; digital elevation model; Quaternary mapping; deglaciation; Quaternary; Finland 1

ACCEPTED MANUSCRIPT

1. Introduction

RI P

T

Reconstruction of the dynamic behavior of the Fennoscandian Ice Sheet (FIS) through the last glacial cycle is based on geological evidence such as drumlins, striations, till fabric and stratigraphy, eskers,

SC

and prominent end moraines and end moraine belts (e.g., Larsen et al., 1999; Boulton et al., 2001; Lunkka et al., 2001, 2004; Bennett and Glasser, 2009), as well as numerical ice sheet modeling (e.g.,

NU

Siegert et al., 2001; Siegert and Dowdeswell, 2002). Most studies agree that De Geer moraines are

MA

important pieces of evidence indicating the direction of deglaciation as well as the curvature of the ice margin during the retreat (e.g., Boulton et al., 2001; Lindén and Möller, 2005; Todd et al., 2007).

ED

Even though controversial theories of their formation have been presented (e.g., Zilliacus, 1987, 1989; Lundqvist, 1988), the majority of studies also agree that distinct and regularly spaced De Geer

PT

moraines have a common origin at the grounding line of calving glaciers (Aartolahti et al., 1995;

2015).

AC CE

Larsen et al., 1999; Blake, 2000; Lindén and Möller, 2005; Golledge and Phillips, 2008; Bouvier et al.,

The rapid decay of the FIS before and after the Younger Dryas (YD) cold event was probably connected to a strong negative mass balance caused by increased insolation and climate warming (e.g., Siegert et al., 2001; Siegert and Dowdeswell, 2002). There is also considerable evidence of areal and local variability in the rate of FIS decay, and it was influenced by factors such as local climate, glacier dynamics, terrain topography, and the proglacial water depth at the retreating ice sheet margin (e.g., Boulton, 1986; Boulton et al., 2001; Ojala et al., 2015). Glaciers terminating in the sea or a lake play an important role in the ice sheet mass balance because, in addition to melting from runoff, evaporation, and sublimation, iceberg calving is a rapid way of losing mass at the margin of glaciers (e.g., Bennett and Glasser, 2009). This applies particularly well to the area that covers the Gulf of Bothnia and southern and western Finland, where fast ice flow areas existed and the margin 2

ACCEPTED MANUSCRIPT

of the retreating FIS terminated at the Baltic Sea basin (BSB) with water up to 300 m deep (e.g., Donner, 1995; Boulton et al., 2001). In such a calving environment, margins of ice sheet lobes were susceptible to cyclic fluctuations at their grounding line, which, in specific conditions, gave rise to the formation of distinctive De Geer moraines (e.g., De Geer, 1889, 1940; Hoppe, 1957, 1959; Golledge

RI P

T

and Phillips, 2008). Their appearance and features thus reflect the behavioral dynamics of the former ice masses during FIS deglaciation.

SC

The present study is a continuation of the LiDAR-based digital elevation model (DEM) characterization of De Geer moraines in Finland by Ojala et al. (2015). Ojala et al. (2015) in Finland

NU

and Bouvier et al. (2015) in Sweden demonstrated that LiDAR-based DEMs capture the dimensions

MA

and characteristics of De Geer moraines efficiently and accurately and provide an opportunity to examine their appearance rather rapidly and at an increasingly detailed level, allowing their

ED

integration with various digital mapping elements using geographical information systems (GIS). Ojala et al. (2015) used a methodological approach and LiDAR DEM testing by selecting only 10 previously

PT

known sites with De Geer moraines in Finland, whereas Bouvier et al. (2015) already mapped all the

AC CE

De Geer fields appearing in Sweden. In this study, an attempt was made to establish a database with an extensive collection of different types of end moraines, including De Geer moraines, from southern and western Finland where LiDAR DEMs are available. The present focus is on their spatial distribution and appearance (maturity, orientation, number of consecutive ridges, mean interdistances), but the proglacial water depth in which they were formed is also recorded. The second aim was to test if the proposed relationship between proglacial water depth and interdistance suggested by Ojala et al. (2015) exists for a comprehensive data set and to investigate how corresponding De Geer moraine interdistances are with regional varved clay-based ice margin retreat rates in different parts of Finland. The work is also part of a larger ongoing research project with the main overall aim of reconstructing the glacier dynamic maps of Quaternary geology in Finland.

3

ACCEPTED MANUSCRIPT 2. Study area and methods

Much of northern Eurasia, including Finland, was covered by an enormous ice sheet complex during

RI P

T

the Last Glacial Maximum (LGM) (e.g., Grosswald, 1980, 1998; Siegert et al., 2001; Svendsen et al., 2004). The Fennoscandian Ice Sheet (FIS) reached its maximum extension in the Vologda area in the

SC

northwestern part of the Russian plain at about 18 ka ago and started to retreat some 14.8 ka ago (Larsen et al., 1999; Lunkka et al., 2001). By 14,250 cal BP, the ice margin had retreated to the

NU

southeast sector of Lake Onega, and the deglaciation of the Onega basin took about 1500 years

MA

(Saarnisto and Saarinen, 2001). According to the chronology by Saarnisto and Saarinen (2001), the formation of the Younger Dryas end moraines, the Finnish Salpausselkä I (Ss I) and Salpausselkä II (Ss

ED

II) in southern Finland, took place between 12,250 and 11,590 cal BP, with about 200 varve years for the FIS margin to retreat from Ss I to the Ss II, which are located ca. 40 km apart (Sauramo, 1923,

PT

1929). According to Sauramo's (1923, 1929) varve clay chronology, the SIS margin then retreated to

AC CE

the Vaasa area 1300 years later at about 10,300 cal BP. Earlier mapping and characterization of De Geer moraines in Finland has been conducted by Virkkala (1963), Aartolahti (1972), Aartolahti et al. (1995), Zilliacus (1987, 1989), Breilin et al. (2005), and Mäkinen et al. (2007), among others. As recently reviewed by Ojala et al. (2015) and Bouvier et al. (2015), their appearance in Fennoscandia is related to the history of the BSB (e.g., Björck, 1995). In Finland, De Geer moraines are mainly clustered in the Salpausselkä region, the foreland of the Salpausselkäs, the Mynämäki-Pyhämaa area in SW Finland, and in the Vaasa archipelago in the west. De Geer moraine fields are very scarce in supra-aquatic areas and only sporadically appear in northern Finland at locations where short-lived, ice-dammed lakes existed (Johansson and Kujansuu, 2005). Zilliacus (1989) identified 17 variable-sized De Geer moraine areas in Finland; whereas Mäkinen et al. (2007) divided them into eight significant fields. The original observations by Zilliacus (1989) were mainly based on interpretations of topographic maps (1:20000) and field investigations. 4

ACCEPTED MANUSCRIPT

2.1. LiDAR DEM mapping and moraine feature classification

RI P

T

In this study, LiDAR cloud point data provided by the National Land Survey of Finland were processed at the Geological Survey of Finland (GTK) with ArcGis (©ESRI) software to construct a 2-m grid DEM

SC

for the entire study area. This was based on a feature data set created from cloud point data, which was then used to construct a terrain model (DTM) and finally processed to a DEM of the bedrock and

NU

superficial deposits lying upon it. The final DEM was then processed with a multidirectional, oblique-

MA

weighted hillshade effect (primary illumination direction 325) and optional vertical exaggeration of six times (Jenness, 2013).

ED

The moraine feature classification used in the present study is presented in Table 1. The term 'end moraine' is usually divided into two groups: 'terminal moraines' referring to the farthest

PT

extension of an ice sheet and 'recessional moraines' related to the temporal standstills and/or

AC CE

readvance of a generally retreating glacier (e.g., Benn and Evans, 1998; Bennett and Glasser, 2009). Recessional moraines, which are considered in the present study, are further divided into (i) 'De Geer moraines' and (ii) 'end moraines' with the definitions below. A De Geer moraine is a type of moraine comprised of successions of small parallel to subparallel, subequally spaced, often sharp-crested, narrow, subangular surface boulders, sandy to silty till ridges up to 12 m high. They may be either straight to broadly curvilinear in plan and are interpreted to have formed underwater where the glacier terminated in a former deep lake and subsequently retreated. The individual ridges are often covered with large subangular boulders and separated by varved silt and clay (e.g., Smith, 1987; Benn and Evans, 1998). In the present study, moraine features were classified as De Geer moraines if they included more than five consecutive ridges. An end moraine, on the other hand, is defined as ridge or ridge-like accumulation of till that marks a stillstand position of a present or past glacier front. End moraines that typically form a series of subparallel ridges are sometimes termed ridged end moraines 5

ACCEPTED MANUSCRIPT

(e.g., Whittow, 1984; Bennett and Glasser, 2009). In the present study, moraine features were classified as end moraines if they included 1−5 consecutive ridges. Furthermore, large end moraines (i.e., ice-marginal complexes) – namely Ss I, Ss II, Ss III and the central Finland end moraine (CFEM) (e.g., Lunkka et al., 2004) – were put into an additional category as presented in Table 1.

RI P

T

In the present study, different types of end moraine fields were identified from processed LiDAR DEMs and separated in the database as polygon features. Subsequently, De Geer moraines

SC

were further classified according to their distinction and regularity (see Table 1). This classification did not follow any formal published system but rather was developed for this study in order to

NU

separate different qualities of De Geer fields that typically appear in the studied area in

MA

Fennoscandia. Examples of different maturity De Geer moraine fields are presented in Fig. 1. A key point in establishing maturity classes for De Geer fields was to facilitate more objective discussion of

ED

their appearance, sedimentary environment, and the possible annual cycle involved in the ridgeforming process (see Bouvier et al., 2015). In a wider perspective, such categorization also allows us

PT

to select information only from the better quality De Geer fields, if necessary, for an overall

AC CE

reconstruction of the glacier dynamics of the FIS in Finland during deglaciation. As suggested by Ojala et al. (2015), the dimensions of De Geer moraines, such as their length and maximum height, are less interesting from the perspective of the depositional environment and glacier dynamics than morphological regularity, orientation, and interdistances of ridges. For this reason, the parameters recorded for each end moraine polygon in the present data set included (i) the type of end moraine as given in Table 1, (ii) the number of consecutive ridges, (iii) the area, (iv) interdistances, (v) orientation, (vi) proglacial water depth at the time of their formation, and (vii) maturity classification for De Geer moraine fields. The proglacial water depth was based on the modeled highest shoreline of the BSB by Ojala et al. (2013). Interdistances and the orientation for each field were calculated from polylines drawn perpendicular to moraine ridges and as an average of at least twice as many locations (lines) as there were consecutive ridges in the field.

6

ACCEPTED MANUSCRIPT 3. Results and discussion

RI P

T

3.1. Regional distribution of De Geer moraines

A total of 711 end moraines and 811 De Geer moraine fields were recorded in the database in the

SC

present study. Their regional occurrence is presented and compared with earlier studies by Zilliacus (1989) and Mäkinen et al. (2007) in Fig. 2. Even though many areas that are characterized by the

NU

most distinctive and well-known De Geer moraine fields in Finland have been identified in all these

MA

studies, substantial differences exists between the earlier reviews of their distribution and the findings of the present investigation. The present LiDAR DEM studies indicate that the appearance of

ED

De Geer and more scattered end moraine features is more widespread in Finland than previously presented. From north to south, De Geer fields in the Vaasa area in western Finland have been clearly

PT

identified in all studies (M7 and Z1, here and hereafter referred to as M for Mäkinen et al. (2007) and

AC CE

Z for Zilliacus (1989)). Where Mäkinen et al. (2007) also took into account the De Geer moraine features in a marine area in the Vaasa archipelago (see, e.g., Breilin et al, 2005), the present analysis of high-resolution DEMs indicates that De Geer ridges are more widely distributed to the west and south of Vaasa than suggested earlier. Zilliacus (1989) identified a small area south of Vaasa (Z16), but LiDAR DEMs showed no clear indications of moraine ridges inside this area. However, LiDAR DEMs showed pronounced De Geer fields appearing all around Z16, extending to a larger field some 70 km southwest of Z16. The area Z5 of Zilliacus (1989) compares well with the present observations of De Geer moraine fields, but this was not noted by Mäkinen et al. (2007). A peculiar observation by Zilliacus (1989) is the presence of De Geer moraines in areas Z6, Z7, and Z8. No indication of distinct and regularly spaced ridge features or more scattered end moraines in LiDAR DEMs in these areas was found in the present study. Instead, the present analysis identified a substantial number of De Geer moraine fields 7

ACCEPTED MANUSCRIPT

and scattered end moraines some 20 km northeast of Z7, in a location of some kilometers inside the CFEM. A few distinct De Geer fields were also identified inside the northernmost extension of the CFEM between Jyväskylä and Äänekoski in central Finland. North of this, LiDAR DEMs indicated a number of scattered continuations of 1 to 5 consecutive ridges and ridge-like features that were

RI P

T

classified as end moraines in the present study.

The classical areas with distinctive De Geer moraines in Eura-Luvia (M2, Z3) and Mynämäki-

SC

Laitila (M1, Z2) were almost identically identified in the present study compared to how they were proposed by Zilliacus (1989) and Mäkinen et al. (2007). The same applies to area M3 by Mäkinen et

NU

al. (2007), which was for some reason not acknowledged by Zilliacus (1989). According to present

were classified as De Geer moraines.

MA

LiDAR DEM studies, some scattered end moraines appear between M1 and M3, but only a few fields

ED

The area of Åland Island in the southwestern archipelago is also characterized by De Geer moraine ridges as already suggested, for example, by Zilliacus (1989) (Z15). According to the present

AC CE

occurrence.

PT

study, they cover a more extensive area than previously suggested but are mainly rather scattered in

De Geer moraines clustered in the Salpausselkä region and in the foreland of the Salpausselkäs are more common and widespread than previously presented (Zilliacus, 1989; Mäkinen et al., 2007; Fig. 2). Their main occurrence areas (M4, M5 and Z4, Z9, Z10, Z12) are well known and confined, such as a classic field in the Hyvinkää-Mäntsälä area (Z4), which was also easily identified from LiDAR DEMs in the present study. However, many of the ridges within the area M4 of Mäkinen et al. (2007) were classified as end moraines in the present study rather than De Geer moraines, which best agrees with the interpretation by Zilliacus (1989). This particularly applies to most ridge features situated between Ss I and Ss II on the northwest side of Z4. Further east, and south of Z11, the ridge formations between Ss I and Ss II are again very well pronounced and equally spaced, showing the characteristics of typical De Geer moraines. Their appearance was not detected in previous studies by Zilliacus (1989) or Mäkinen et al. (2007). Neither had they discovered the abundance of distinct De 8

ACCEPTED MANUSCRIPT

Geer moraine features between Z13 and M6, which are located directly in the foreland of Ss I and continue inside and south of M6 of Mäkinen et al. (2007). The present screening of LiDAR DEMs indicate that the southern coastline of Finland is divided into two parts in terms of the appearance of De Geer moraines. In the western sector, between

RI P

T

Porvoo and Hanko, no De Geer fields are visible in the geomorphology close to the coastline; whereas in the eastern sector (east of Porvoo), they appear frequently and continue along the Finnish-Russian

SC

border. The current coverage of LiDAR DEMs in Finland does not extend to the Ilomantsi De Geer

MA

NU

field (M8) on the southeast side of the Koitere ice-marginal complex.

ED

3.2. Deposition environment of De Geer moraines

A number of investigations have aimed at revealing a unifying theory for the genesis of De Geer

PT

moraines with varying sedimentological composition and morphological features (e.g., De Geer,

AC CE

1889; Hoppe, 1959; Zilliacus, 1987; Larsen et al., 1999; Blake, 2000; Lindén and Möller, 2005). However, as suggested by Bennett and Glasser (2009) and Bouvier et al. (2015), De Geer moraines are equifinal landforms and no single model is likely to be appropriate for their genesis at all localities and for all types of De Geer ridges. This is probably because of variations in the geological environment as well as in the dynamic nature of a glacier grounding line and its seasonal stability, which are a function of ice sheet velocity and the calving rate at the margin. This interpretation is supported by studies on different types of moraine ridges currently produced by modern retreating glaciers in Iceland, Spitsbergen, and Baffin Island (e.g., Boulton, 1986; Ottesen and Dowdeswell, 2006; Flink et al., 2015). Numerous studies indicate that detailed analysis of the internal sediment architecture of De Geer and end moraine ridges is in each case needed in order to reveal and understand the site-specific processes responsible for their formation (e.g., Aartolahti et al., 1995; Lindén and Möller, 2005; Golledge and Phillips, 2008). Such analyses are practically impossible with 9

ACCEPTED MANUSCRIPT

the large geomorphological data set presented in this study, and consequently the main aim here is to provide a general characterization of the environment where De Geer moraine fields of differing maturity appear in Finland. The present mapping of the regional distribution of De Geer moraines using LiDAR DEMs

RI P

T

analysis supports the interpretation by Aartolahti et al. (1995), who suggested that De Geer fields were deposited during practically all temporal stages of deglaciation and the history of the BSB in

SC

Finland. Their deposition history represents a period of 3000 years from ca. 13,000 on the southern coast of Finland to 10,000 cal BP in the Vaasa area near the Gulf of Bothnia (Fig. 2), indicating that

NU

neither the salinity of the proglacial sea/lake basin nor the mean annual temperature are sole or even

MA

substantially important forcing factors in their formation. However, the deposition of De Geer moraines in Finland appears to be particularly well connected to the YD and post-YD climate

ED

oscillations as they abundantly appear in the zones of Salpausselkäs and CFEM. This is in very good agreement with their regional distributions in Sweden (e.g., De Geer, 1940; Bouvier et al., 2015),

PT

indicating that their formation is typical for climatic periods that experienced major fluctuations and

AC CE

the related places where substantial glacier oscillations took place during the FIS deglaciation. Another observation from Fig. 3 is that the best maturity (classes 1a and 1b) De Geer fields are evenly distributed all around southern and western Finland and that no evident geographical pattern exists in the appearance of more scattered De Geer features either. This indicates that the local environment and glacier dynamics determine their maturity and distinctiveness rather than solely more regional climate oscillations. Of the 811 De Geer fields recorded in the present study, about one-fourth were classified as distinct and regular (class 1a) or moderately regular (class 1b) (Table 2). The remaining observations of De Geer ridges were classified as more scattered, very scattered, or less distinct in their appearance. The most distinctive and regularly spaced De Geer fields appear in locations where they have typically been described and studied earlier by Virkkala (1963), Zilliacus (1987, 1989), and Aartolahti et al. (1995), among others. The substantial variability in their geomorphological character, 10

ACCEPTED MANUSCRIPT

even within a limited region in Finland, probably indicates the diversity of their genesis at different locations or at least meaningful variability in their formation processes in different environments. According to the statistics (Table 2), the regular and distinctive De Geer moraine fields typically are somewhat larger in area and contain a larger number of consecutive and evenly spaced ridges than

RI P

T

classes 1c and 1d. This suggests that where conditions for the dynamic oscillations of the ice margin were met, a larger number and more contiguous area of distinct and regularly spaced De Geer

SC

moraines were regionally formed. The distinct and regular De Geer moraine fields often contain 20−30 consecutive ridges and cover an area of 50 to 100 ha, whereas the more scattered ones

NU

typically are about 25−30 ha or less and contain 10−15 consecutive ridges. The fact that De Geer

MA

moraine fields of classes 1a and 1b have been most frequently investigated and referred to in previous studies (e.g., Zilliacus, 1989; Aartolahti et al., 1995; Mäkinen et al., 2007) indicates that their

ED

size and interdistance interpretations, among other characteristics, have mostly been based on the best quality and most distinct De Geer moraine features. This enhances the need for and importance

PT

of more thorough sedimentological studies of the less distinct and scattered De Geer ridges in order

AC CE

to better understand the processes related to their formation. The relative water depth in which the De Geer moraines were formed during the deglaciation of the FIS is also presented in Fig. 3. According to Zilliacus (1989), they occur in areas with proglacial water depths from 15 to 270 m at deglaciation, although preferentially in excess of 150 m according to Lindén and Möller (2005). The present results concerning De Geer field occurrences support the interpretation by Zilliacus (1989), as earlier pointed out by Ojala et al. (2015) with only 10 sites from southern and western Finland. Furthermore, as seen from Fig. 3, no clear pattern exists in the occurrence of the regular and well-pronounced (classes 1a and 1b) De Geer moraine fields in a certain proglacial depth zone in the course of deglaciation. The same applies to less regular and indistinct (classes 1c, 1d, and 1e) De Geer ridges. This observation suggests that the proglacial water depth cannot be solely responsible for their formation, the number of consecutive ridges, or their morphological characteristics. 11

ACCEPTED MANUSCRIPT

The present data set also indicates the absence of De Geer moraines closer to supra-aquatic areas, for example in the northeastern sector of the presently screened area in central Finland (Fig. 2). Based on LiDAR DEMs, scattered end moraines are abundant in these areas, some of which extend many meters above the highest shoreline. Here, the ridges and ridge-like features probably represent

RI P

T

frontal push-moraines formed on land during annual winter readvances, as described by Boulton (1986) in Breidamerkurjokull, Iceland. They were formed as a result of seasonal changes in the

SC

behavior of the glacier terminus, where terminal advances took place when the small winter flow velocity exceeded the horizontal component of ablation (Boulton, 1986).

NU

As typical De Geer ridges in Finland often consist of till from basal melt-out (e.g., Virkkala, 1963;

MA

Aartolahti et al., 1995), a sufficient thickness of Quaternary deposits needed to provide source material for ridges is probably another important prerequisite for their formation and appearance.

ED

The Åland Island in southwestern Finland (Z15) is an example of an area where only scattered De Geer moraines exist, probably because of bedrock dominance and thin, <1-m-thick superficial

PT

Quaternary coverage (till and sand, gravel) (GTK, 2015). In general, the typical thickness of superficial

AC CE

deposits where De Geer and end moraines appear in Finland varies between 5 and 30 m, but no clear relationship exists between the appearance of different maturity De Geer moraine ridges and the thickness of superficial deposit.

De Geer moraines typically appear close to the main esker systems in recently deglaciated terrains (e.g., Virkkala, 1963; Elverhøi et al., 1980; Sollid and Carlson, 1984). This also seems to be the case with the presently recorded ridge distributions, as seen from Fig. 4A, where De Geer moraines and end moraines are examined together with the main glaciofluvial deposits in Finland. The best known and most distinct De Geer fields in Eura-Luvia (M2, Z3) and Mynämäki-Laitila (M1, Z2) are respectively associated with the Mynämäki-Laitila-Pyhäranta and Säkylä-Eura-Luvia esker systems in southwestern Finland. Farther north, a group of De Geer moraine fields is related to the Hämeenkangas-Pohjankangas glaciofluvial system (Z5 and its vicinity), as well as to its northern extension some 50 km southeast of Vaasa. In such places, the orientation of De Geer moraine ridges 12

ACCEPTED MANUSCRIPT

is more or less perpendicular to the direction of eskers as shown in an example from Tarinmaa, Janakkala in Fig. 4B. In many places, the orientation of De Geer moraine ridges even follows a more complicated and branching glaciofluvial system, thus indicating an undulating concave or convex icemargin configuration during deglaciation. In such cases, De Geer moraines provide an enhanced and

RI P

T

substantially more detailed configuration of the retreating ice margin than the direction of eskers. Altogether, 1191 polygonal mean directions of deglaciation and interdistances were recorded in the

SC

present data set. These were based on 14,052 measurements of polylines recorded for 811 De Geer fields and 380 end moraine fields where more than two consecutive ridges appeared.

NU

Another typical location for well-pronounced and distinct De Geer moraine ridges is on the

MA

proximal side of Ss II and the CFEM, as well in the foreland of Ss I. Here, they display a parallel orientation with large end moraines, indicating the lobate nature of the FIS during deglaciation. This

ED

is shown in an example from the distal side of Ss I in Kankaanranta, Lappeenranta (Fig. 4C). As seen from Fig. 4A, the regional distribution of different types of end moraines is not solely limited to the

PT

vicinity of major esker systems and large end moraine features, but many orientation and

AC CE

interdistance measurements recorded in the present data set are also located between major branching esker systems and in areas with either weak or irregular esker features. In such places, De Geer moraine fields typically are smaller in size and their ridges are often not so well pronounced, but they nevertheless provide an important element for the future construction of FIS glacier dynamic maps of Quaternary geology in Finland.

3.3. Interdistances and the rate of deglaciation

Since De Geer (1889) first presented evidence for the processes and preservation of De Geer moraines, a number of opinions and interpretations have been reported concerning whether they are formed annually and thus directly reflect the rate of regional deglaciation (e.g., Hoppe, 1959; 13

ACCEPTED MANUSCRIPT

Strömberg, 1965; Zilliacus, 1987; Larssen et al., 1999; Lindén and Möller, 2005; Ottesen and Dowdeswell, 2006; Bouvier et al., 2015; Ojala et al., 2015). The current understanding is that because of their heterogeneous appearance (equifinal nature of formation) and inaccurate means of identification and classification worldwide, not all till, gravel, and sand ridges determined as De Geer

RI P

T

moraines can be taken as reflecting the annual rhythm of deglaciation sensu stricto (e.g., Sollid and Carlson, 1984; Aartolahti et al., 1995; Blake, 2000; Lindén and Möller, 2005; Todd et al., 2007;

SC

Bouvier et al., 2015; Ojala et al., 2015). Sedimentological studies with modern retreating glaciers have demonstrated that several ridges may be annually formed because of the subannual cyclicity of

NU

meltwater production or calving dynamics, while in other cases some ridges may take more than one

MA

year to form because of the local characteristics of glacier dynamics and mass balance or the lack of basal material (e.g., Boulton, 1986; Tikkanen, 1989; Blake, 2000; Lindén and Möller, 2005).

ED

Sometimes, the truncation of previously formed ridges is also apparent because of the readvance of the glacier margin (Larsen et al., 1999). Nevertheless, using high-resolution LiDAR DEM

PT

interpretations and supported by the findings of Ottesen and Dowdeswell (2006) and Flink et al.

AC CE

(2015), Bouvier et al. (2015) suggested that the interdistances of late-glacial/early Holocene regular and evenly spaced De Geer moraines in Sweden probably represent the annual rate of local icemargin retreat. They based their findings on comparisons between ridge interdistances and independently determined ice-margin rates that were based on Swedish varve chronology (De Geer, 1940). The regional distribution of interdistance measurements from De Geer moraine ridges carried out in the present study is presented in Fig. 5. They are based on the same measurements of polylines that were used in direction determinations and recorded for 811 De Geer fields. A general trend is that De Geer interdistances substantially increase in the direction of deglaciation, which is in accordance with the rate of deglaciation by Sauramo (1929) and Boulton et al. (2001), the latter being mainly based on Sauramo's varved clay chronology. The area between Ss I and Ss II and sporadic places in the immediate foreland of Ss I clearly stand out in the map as having shorter 14

ACCEPTED MANUSCRIPT

(below 60 m) interdistances than anywhere else in Finland. South of the Salpausselkäs, the interdistances typically are between 60 and 100 m and then 100−140 m closer to the coastline and on the western sector outside Ss I. North of the Salpausselkä zone, interdistances increase to more than 100 m, as seen in De Geer Fields at locations M3 and Z11. The same applies immediately north of the

RI P

T

CFEM zone, southwest Finland, between Ss II and Ss III and inside Ss III (M3), where the interdistances are generally 100−160 m. In the Mynämäki-Laitila (M1, Z2) and Eura-Luvia (M2, Z3) De

SC

Geer moraine interdistances mostly vary between 160−220 m, but some fields also indicate 220−300 m distances between adjacent ridges. About 100 km south of the Vaasa archipelago, interdistances of

NU

moraine ridges are constantly over 300 m and exceed 500 m in the archipelago. With the exception

MA

of Åland Island, De Geer interdistances are regionally coherent, suggesting that the ridge-forming process is driven by regional forcing in glacial dynamics and marginal zone sedimentation rather than

ED

random local variables. This suggests that De Geer moraine ridge interdistances, and thus their formation process, are evidently related to the rate of deglaciation. It also indicates that the method

PT

applied in the present study works efficiently and provides rational data.

AC CE

Ojala et al. (2015) earlier showed that in several of the measured test localities in Finland the interdistances of De Geer ridges agreed to a substantial extent with the rate of regional deglaciation by Sauramo (1929). To further investigate this potential relationship, mean interdistances of 50 randomly selected De Geer polygons from the present data set were compared with Sauramo's (1929) deglaciation chronology. The basis for the selection of these 50 sites was that they represent regularly spaced and distinct De Geer moraines (mostly from classes 1a and 1b), cover the entire study area, and are situated as centrally as possible in Sauramo's deglaciation isochrones presented in Fig. 5. The comparison between their ridge interdistances and Sauramo's varve-based rate of deglaciation is given in Fig. 6. It shows that there is a clear linear relationship between these two variables (r2 = 0.800), which is rather close to a linear 1:1 correspondence. Interestingly, the equation of the linear trend (Y = 1.0226X + 13.05) is very similar to that suggested by Bouvier et al. (2015) in a similar type of comparison from Sweden. In the lower end of the diagram, mainly represented by 15

ACCEPTED MANUSCRIPT

data points from the Ss I and Ss II zones and their immediate foreland (i.e., -1000 to ±0 according to Sauramo's chronology), the measured interdistances apparently are generally somewhat longer than the annual rate of deglaciation suggested by Sauramo (1929). This may be related to a substantial oscillation of the ice-margin around the Younger Dryas (YD) cold period, probably characterized by

RI P

T

standstills, retardation, and readvances of separate ice lobes (e.g., Boulton et al., 2001). Regarding the middle and upper parts of the diagram, where the measured interdistances fluctuate above and

SC

below Sauramo's values, it is impossible to estimate to what extent this is caused by inaccuracies in determining Sauramo's rate of deglaciation from isochrones, by site-specific variations in local

NU

climate and glacial dynamics, or because interdistances simply do not always follow the annual

MA

pattern of deposition. In general, these results support the conclusion of Bouvier et al. (2015) that an annual cycle is involved in the De Geer ridge-forming process and that regular and evenly spaced De

ED

Geer ridges probably represent the local rate of ice-margin retreat, or at least are very close to that (e.g., Ottesen and Dowdeswell, 2006; Bouvier et al., 2015; Flink et al., 2015). The key question here is

PT

to what extent and quality these De Geer moraine ridges can be interpreted as annual formations.

AC CE

Answering this would require more detailed analysis and comparison between all different De Geer field classes (1a to 1e) stored in the present data set and the original site-specific clay varve observations by Sauramo (1923, 1929) instead of interpolated isochrones. The final observation from the present data set relates to the prominent relationship between De Geer ridge interdistances and the proglacial water depth in which they were formed, as presented earlier by Ojala et al. (2015). This relationship also relates to the discussion on how well De Geer moraine ridges represent the rate of deglaciation because interdistances are likely reflecting a combination of the rapidity of ice-margin recession, proglacial water depth, and terrain topography, which are probably all interacting (e.g., Boulton et al., 2001; Ojala et al., 2015). With the present data set, this relationship is illustrated in Fig. 7. By selecting classes 1a and 1b De Geer moraine fields, a very prominent correlation (r2 = 0.828) exists between these variables where the interdistance appears to be about two times higher than the proglacial water depth earlier suggested by Ojala et al. 16

ACCEPTED MANUSCRIPT

(2015). This correlation coefficient is also very close to that of r2 = 0.841 provided by Ojala et al. (2015) using only 10 selected sites. If all De Geer maturity classes in the present data set are included, then their correlation is slightly lower (r2 = 0.718) but still statistically significant. One interesting exception to this is that De Geer moraine fields that are located more than 15 km south of the

RI P

T

Salpausselkäs, i.e., deposited before the YD cold period and substantial climate oscillation, are characterized by a more or less 1:1 linear relationship between interdistances and relative water

SC

depth. Connected to the generally recognized slower rate of FIS decay (average of 60 m/y) (e.g., Sauramo, 1929; Lunkka et al., 2004), the dynamic behavior of the FIS was perhaps different prior to

NU

the Younger Dryas than after it. One possible explanation could be that melting from runoff was a

MA

more important factor in the decay of the FIS earlier, whereas after the Younger Dryas calving became the dominant factor in SIS mass loss at the margin. However, a deficiency of reliable data on

AC CE

4. Conclusions

PT

ED

the BSB water levels for south of the Salpausselkä area makes this inconclusive.

 LiDAR DEM investigation provides a revolutionary tool for interpreting geological and geomorphological characteristics related to former ice sheet advances and decay, for example with De Geer and scattered end moraines. By using GIS interpretations, these features can be mapped efficiently and at an increasingly detailed level for large areas, such as southern and western Finland, allowing their integration with various digital mapping elements.  LiDAR DEM-based interpretations revealed that the regional distribution of De Geer moraines and scattered end moraine ridges is more widespread in Finland than previously presented.  A total of 711 end moraine and 811 De Geer moraine fields were recorded in the database in the present study. Of these, mean interdistances and orientations were recorded for all De

17

ACCEPTED MANUSCRIPT

Geer fields and for 380 end moraine fields, which will provide information for reconstructing glacier dynamic maps of Quaternary geology in southern and western Finland.  Evident indications exists that De Geer ridge interdistances reflect the rate of deglaciation but are probably also affected by the proglacial water depth in which they were formed and by

RI P

T

local glacier dynamics related to climate and terrain topography, which all interact.  A clear relationship exists between proglacial water depth and De Geer moraine interdistance

SC

where longer interdistances are related to deeper water in the proglacial basins.

NU

Acknowledgements

MA

Jukka-Pekka Palmu and two anonymous reviewers are thanked for their valuable comments.

ED

References

PT

Aartolahti, T., 1972. On deglaciation in southern and western Finland. Fennia 114, 1 –84.

AC CE

Aartolahti, T., Koivisto, M., Nenonen, K., 1995. De Geer moraines in Finland. Geological Survey of Finland, Special Paper 20, 67–74. Benn, D.I., Evans, D.J.A., 1998. Glaciers and Glaciation. Arnold, London, 734 pp. Bennett, M., Glasser, N., 2009. Glacial geology – ice sheets and landforms. Wiley-Blackwell, Chichester, 385 pp. Björck, S., 1995. A review of the history of the Baltic Sea, 13.0–8.0 ka BP. Quaternary International, 27, 19–40. Blake, K.P., 2000. Common origin for De Geer moraines of variable composition in Raudvassdalen, northern Norway. Journal of Quaternary Science 15, 633– 644. Boulton, G.S., 1986. Push-moraines and glacier-contact fans in marine and terrestrial environments. Sedimentology 33, 677–698.

18

ACCEPTED MANUSCRIPT

Boulton, G.S., Dongelmans, P., Punkari, M., Broadgate, M., 2001. Palaeoglaciology of an ice sheet through a glacial cycle: the European ice sheet through the Weichselian. Quaternary Science Reviews 20, 591–625. Bouvier, V., Johnson, M., Påsse, T., 2015. Distribution, genesis, and annual-origin of De Geer moraines

RI P

T

in Sweden: insights revealed by LiDAR. GFF, in press (DOI: 10.1080/11035897.2015.1089933). Breilin, O., Kotilainen, A., Nenonen, K., Räsänen, M., 2005. The unique moraine morphology,

SC

stratotypes and ongoing geological processes at the Kvarken Archipelago on the land uplift area in the western coast of Finland. In: Ojala, A.E.K. (ed), Quaternary studies in the northern and Arctic

NU

regions of Finland: proceedings of the workshop organized within the Finnish National Committee

MA

for Quaternary Research (INQUA). Geological Survey of Finland, Special Paper 40, 97–111. De Geer, G., 1889. Ändmoräner i trakten mellan Spånga och Sundbyberg. Geologiska Föreningens i

ED

Stockholm Foörhandlingar 11, 395–397.

Handlingar III, 367 pp.

PT

De Geer, G., 1940. Geochronologia Suecica principles. 18, Kungliga Svenska Vetenskapsakademiens

AC CE

Donner, J., 1995. The Quaternary history of Scandinavia. Cambridge University Press, Cambridge. Elverhøi, A., Liestøl, O., Nagy, J., 1980. Glacial erosion, sedimentation and microfauna in the inner part of Kongsfjorden, Spiztsbergen. Nor. Polarinst. Skr. 172, 33–62. Flink, A.E., Noormets, R., Kirchner, N., Benn, D., Luckman, A., Lovell, H., 2015. The evolution of a submarine landform record following recent and multiple surges of Tunabreen glacier, Svalbard. Quaternary Science Reviews 108, 37–50. Golledge, N.R., Phillips, E., 2008. Sedimentology and architecture of De Geer moraines in the western Scottish Highlands, and implications for groundingline glacier dynamics.Sedimentary Geology 208, 1–14. Grosswald, M.G., 1980. Late Weichselian ice sheet of northern Eurasia. Quaternary Research 13, 132.

19

ACCEPTED MANUSCRIPT

Grosswald, M.G., 1998. Late-Weichselian ice sheets in Arctic and Pacific Siberia. Quaternary International 45/46, 3–18. GTK, 2015. Superficial deposit thickness 1:1 000 000, Geological Survey of Finland. URL: http://hakku.gtk.fi/

RI P

T

Hoppe, G., 1957. Problems of glacial morphology and the ice age. Geografiska Annaler 39, 1–18. Hoppe, G., 1959. Glacial morphology and inland ice recession in northern Sweden. Geografiska

SC

Annaler 41, 193–212.

Jenness, J., 2013. DEM surface tools for ArcGIS (surface_area.exe). Jenness Enterprises. Available at:

NU

http://www.jennessent.com/arcgis/surface_area.htm

MA

Johansson, P., Kujansuu, R., 2005. Quaternary deposits of Northern Finland. Geological Survey of Finland, Espoo.

ED

Larsen, E., Lyså, A., Demidov, I., Funder, S., Houmark-Nielsen, M., Kjaer, K., Murray, A.S., 1999. Age and extent of the Scandinavian ice sheet in northwest Russia. Boreas 28, 115–132.

PT

Lindén, M., Möller, P., 2005. Marginal formation of De Geer moraines and their implications to the

AC CE

dynamics of grounding-line recession. Journal of Quaternary Science 20, 113–133. Lundqvist, J., 1988. Late glacial ice lobes and glacial land forms in Scandinavia. In: Goldthwait, R.P., Matsch, C.L. (eds.), Genetic classification of glacigenic deposits, A.A. Balkema, Rotterdam, 217– 225.

Lunkka, J.P., Saarnisto, M., Gey, V., Demidov, I., Kiselova, V., 2001. Extent and age of the Last Glacial Maximum in the southeastern sector of the Scandinavian Ice Sheet. Global and Planetary Change 31, 407–425. Lunkka, J.P., Johansson, P., Saarnisto, M., Sallasmaa, O., 2004. Glaciation in Finland. In: Ehlers, J., Gibbard, P.L. (eds.), Quaternary glaciations – Extent and chronology, Elsevier, Amsterdam, 93–100. Mäkinen, K., Palmu, J.-P., Teeriaho, J., Rönty, H., Rauhaniemi, T., Jarva, J., 2007. Nationally valuable moraine formations. The Finnish Environment 14/2007 Ministry of the Environment.

20

ACCEPTED MANUSCRIPT

Ojala, A.E.K., Palmu, J.-P, Åberg, A., Åberg, S., Virkki, H., 2013. Development of an ancient shoreline database to reconstruct the Litorina Sea maximum extension and the highest shoreline of the Baltic Sea basin in Finland. Bulletin of the Geological Society of Finland 85, 127–144. Ojala, A.E.K., Putkinen, N., Palmu, J.-P., Nenonen, K., 2015. Characterization of De Geer moraines in

RI P

T

Finland based on LiDAR DEM mapping. GFF, in press (DOI: 10.1080/11035897.2015.1050449). Ottesen, D., Dowdeswell, J.A., 2006. Assemblages of submarine landforms produced by tidewater

SC

glaciers in Svalbard. Journal of Geophysical Research 111, F1.

Saarnisto, M., Saarinen, T., 2001. Deglaciation chronology of the Scandinavian Ice Sheet from the

NU

Lake Onega Basin to the Salpausselkä end moraines. Global and Planetary Change 31, 387–405.

MA

Sauramo, M., 1923. Studies on the Quaternary varve sediments in southern Finland. Bulletin de la Commission Géologique de Finlande 60, 164 pp.

ED

Sauramo, M., 1929. The Quaternary geology of Finland. Bulletin de la Commission géologique de Finlande 86, 110 pp.

PT

Siegert, M.J., Dowdeswell, J.A., 2002. Late Weichselian iceberg, surface melting and sediment

AC CE

production from the last Eurasian Ice Sheet: results from numerical ice-sheet modelling. Marine Geology 188, 109–127.

Siegert, M.J., Dowdeswell, J.A., Hald, M., Svendsen, J.I., 2001. Modelling the Eurasian Ice Sheet through a full (Weichselian) glacial cycle. Global and Planetary Change 31, 367–385. Smith, D.G., 1987. Landforms of Alberta. Alberta Remote Sensing Center, Alberta Environment, Edmonton. Publication 87-1. Sollid, J.L., Carlson, A.B., 1984. De Geer moraines and eskers in Pasvik, Northe Norway. In: Königsson, L.-K. (ed) Ten years of Nordic till research. Striae 20, 55–61. Strömberg, B., 1965. Mappings and geochronological investigations in some moraine areas of southcentral Sweden. Geografiska Annaler 47A, 73–82. Svendsen, J.I., Alexanderson, H., Astakhov, V.I., Demidov, I., Dowdeswell, J.A., Funder, S., Gataullin, V., Henriksen, M., Hjort, C., Houmark-Nielsen, M., Hubberten, H.W., Ingólfsson, Ó., Jakobsson, M., 21

ACCEPTED MANUSCRIPT

Kjær, K.H., Larsen, E., Lokrantz, H., Lunkka, J.P., Lyså, A., Mangerud, M., Matiouchkov, A., Murray, A., Möller, P., Niessen, F., Nikolskaya, O., Polyak, L., Saarnisto, M., Siegert, C., Siegert, M.J., Spielhagen, R., Stein, R., 2004. Late Quaternary ice sheet history of northern Eurasia. Quaternary Science Reviews 23, 1229–1271.

RI P

T

Tikkanen, M., 1989. Geomorphology of the Vantaanjoki drainage basin, southern Finland. Fennia 167, 19–72.

SC

Todd, B.J., Valentine, P.C., Longva, O., Shaw, J., 2007. Glacial landforms on German Bank, Scotian Shelf: evidence for Late Wisconsinan ice-sheet dynamics and implications for the formation of De

NU

Geer moraines. Boreas 36, 148–169.

géologique de Finlande 210, 1–76.

MA

Virkkala, K., 1963. On ice-marginal features in southwestern Finland. Bulletin de la Commission

ED

Whittow, J.B., 1984. The Penguin Dictionary of Physical Geography. Penguin Books, 591 pp.

239.

PT

Zilliacus, H., 1987. De Geer moraines in Finland and the annual moraine problem. Fennia 165, 145 –

Captions

AC CE

Zilliacus, H., 1989. Genesis of De Geer moraines in Finland. Sedimentary Geology 62, 309–317.

Fig. 1. Hillshaded DEM (LiDAR DEMs) examples of De Geer moraine field maturity classification applied in the present study. Examples are given for classes 1a from Ruokojärvi, Kouvola (A), 1b from Kortistonkulma, Hyvinkää (B), 1c from Etu-Holsti, Porvoo (C), 1d from Kituranta, Jämijärvi (D), and 1e from Husula, Luumäki (E), as described in Table 1. A map showing their locations is provided (F). See Ojala et al. (2015) for additional De Geer moraine LiDAR DEMs from Finland. Basemap and LiDAR © the National Land Survey of Finland and LiDAR processing by GTK. (For interpretation of colours in this figure, the reader is referred to the web version of this article.) 22

ACCEPTED MANUSCRIPT Fig. 2. An index map showing the screened areas and locations of different types of end moraines in southern and western Finland recorded in the present study. The main De Geer moraine areas presented by Zilliacus (1989) are indicated with brown line polygons, where Z1−Z4 represents their

RI P

T

major and Z5−Z17 minor occurrences. Eight significant De Geer moraine fields described by Mäkinen et al. (2007) are indicated with black line polygons and numbered M1−M8. The extent of the Baltic

SC

Ice Lake is indicated with a hatched symbol (Ojala et al., 2013), and north of that the deglaciation occurred in the Yoldia Sea and Ancylus Lake. Locations of large end moraines (i.e., ice-marginal

NU

complexes) of the Salpausselkä's (Ss I, Ss II, Ss III) and the central Finland end moraine (CFEM) are

MA

specified with abbreviations. Basemap © the National Land Survey of Finland. (For interpretation of

ED

colours in this figure, the reader is referred to the web version of this article.)

PT

Fig. 3. A map showing the locations of maturity classified De Geer moraine fields in Finland

AC CE

interpreted in the present study. The proglacial water depth at the time of their deposition is indicated with a color scale according to Ojala et al. (2013), including supra-aquatic areas. The extent of the Baltic Ice Lake is indicated with a hatched symbol (Ojala et al., 2013). Basemap © the National Land Survey of Finland. (For interpretation of colours in this figure, the reader is referred to the web version of this article.)

Fig. 4. (A) Locations of De Geer and end moraine fields in the present data set in relation to eskers and ice-marginal complexes in Finland. (B) An example of a De Geer field situated alongside an esker system in Tarinmaa, Janakkala. (C) An example of a De Geer field situated on the distal side of Salpausselkä I (Ss I) in Kankaanranta, Lappeenranta. Basemaps and LiDAR © the National Land Survey

23

ACCEPTED MANUSCRIPT

of Finland and LiDAR processing and map of Quaternary deposits by GTK. (For interpretation of colours in this figure, the reader is referred to the web version of this article.)

Fig. 5. Measured average interdistances of De Geer moraines for all fields in the present data set

RI P

T

compared with marginal ice recession rates reported by Sauramo (1923, 1929) for southern and western Finland. Solid lines are Sauramo’s varved clay chronology isochrones and mean

SC

interdistances are indicated with different colors. Basemap © the National Land Survey of Finland.

NU

(For interpretation of colours in this figure, the reader is referred to the web version of this article.)

MA

Fig. 6. Comparison between mean ridge interdistances of 50 randomly selected distinct (classes 1a and 1b) De Geer fields around the study area and the rate of deglaciation based on Sauramo's (1929)

ED

varved clay chronology (see Fig. 5). The scattered line indicates 1:1 correspondence, while the solid

PT

line is a linear regression (r2 = 0.800) between these two variables.

AC CE

Fig. 7. Relationship between the interdistance of De Geer moraines of classes 1a and 1b (black dots) and classes 1c, 1d and 1e (gray dots) and proglacial water depth. Their linear regressions are r2 = 0.841 for classes 1a, and 1b (black line) and r2 = 0.718 for the entire data set (gray line), indicating that longer distances between adjacent ridges are associated with deeper water in front of the retreating ice sheet.

Table 1. Classification basis of different types of end moraine features used in the present study on LiDAR DEM interpretations.

Table 2. General statistics for different maturity De Geer moraine fields in the present data set.

24

AC CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

Figure 1

25

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

Figure 2

26

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

Figure 3

27

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

Figure 4a

28

AC CE

Figure 4b

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

29

AC CE

Figure 4c

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

30

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

Figure 5

31

Figure 6

AC CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

32

Figure 7

AC CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

33

Table 1

ACCEPTED MANUSCRIPT

RI P

T

1. De Geer moraines: a succession of more than 5 parallel to subparallel and subequally spaced ridges 1a. Distinct and regular 1b. Moderately regular 1c. Fairly scattered 1d. Very scattered 1e. Regular but less distict (often masked by fine-grained sediments)

AC CE

PT

ED

MA

NU

3. Large end moraines (i.e., ice-marginal complexes)

SC

2. End moraines: a series of 1-5 subparallel ridges or ridge-like accumulations

34

Table 2

ACCEPTED MANUSCRIPT Number of polygons

Percentage of polygons

Mean area (ha)

Mean number of consecutive ridges

Max number of consecutive ridges

1a. Distinct and regular 1b. Moderately regular 1c. Fairly scattered 1d. Very scattered 1e. Regular but less distinct

98 93 230 317 73

12.08 11.47 28.36 39.09 9.00

88.29 55.42 31.73 24.84 19.10

24 19 16 13 11

150 90 80 40 30

AC CE

PT

ED

MA

NU

SC

RI P

T

Maturity of De Geer moraines (n=811)

35

AC CE

Graphical abstract

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

36

Highlights:

T

RI P

SC NU MA ED

 

PT

 

Timely and comprehensive update of the De Geer moraines distribution in southern and central Finland Classification of De Geer moraine appearances according to their maturity and regularity LiDAR DEM investigations revealed that De Geer moraines are more widespread in Finland than previously presented A clear relationship exists between proglacial water depth and De Geer moraine interdistance Interdistances of mature and regularly-spaced De Geer ridges reflect the rate of deglaciation

AC CE



ACCEPTED MANUSCRIPT

37