draft - university of toronto t-space · draft 2 1 abstract2 3 in-situ rare earth element...

Draft

Palaeoenvironments revealed from rare earth element

systematics in vertebrate bioapatite from the Lower

Devonian of Svalbard

Journal: Canadian Journal of Earth Sciences

Manuscript ID cjes-2015-0206.R1

Manuscript Type: Article

Date Submitted by the Author: 27-Feb-2016

Complete List of Authors: Zigaite, Zivile; Uppsala University, Evolution and Development

Fadel, Alexandre; Universite Lille 1 Perez-Huerta, Alberto; University of Alabama Jeffries, Teresa; Natural History Museum Goujet, Daniel; Museum national d'histoire naturelle Ahlberg, Per; Uppsala University

Keyword: Rare earth elements, palaeoenvironment, vertebrate microfossils, Devonian, Svalbard

https://mc06.manuscriptcentral.com/cjes-pubs

Canadian Journal of Earth Sciences

Draft

1

Zigaite et al. 1

Canadian Journal of Earth Sciences IGCP 591 special issue 2

3

4

Palaeoenvironments revealed by rare earth element 5

systematics in vertebrate bioapatite from the Lower Devonian 6

of Svalbard 7

8

9

Živilė ŽIGAITĖ1*, Alexandre FADEL1,2, Alberto PÉREZ-HUERTA3, Teresa 10

JEFFRIES4, Daniel GOUJET5, Per Erik AHLBERG1 11

12

1 Subdepartment of Evolution and Development, Department of Organismal Biology, Uppsala 13

University, Norbyvägen 18A, 752 36 Uppsala, Sweden 14

2 Laboratoire Géosystèmes, Université Lille 1, CNRS UMR8217, Villeneuve d’Ascq 59655, 15

France 16

3 Department of Geological Sciences, University of Alabama, 2018 Bevill Building, 17

Tuscaloosa, AL 35487, USA 18

4 Earth Science Department, The Natural History Museum, Cromwell Road, SW7 5BD 19

London, UK 20

5 Muséum national d’Histoire naturelle, Département Histoire de la Terre, Laboratoire de 21

Plaéontologie, UMR CNRS 7207, 57 rue Cuvier, 75231 Paris cedex 5, France 22

* coresponding author: [email protected], +467814716481 23

24

Page 1 of 27



Draft

2

1

Abstract 2

In-situ rare earth element compositions have been measured in early 3

vertebrate microremains from the Lower Devonian basin of Andrée Land (Svalbard), 4

with the aim of obtaining information about their early depositional environment and 5

potential reworking. Vertebrate microremains with different histology were used for 6

the analyses, sourced from two different localities of marginal marine to freshwater 7

sediments from geographically distant parts of the Grey Hœk Formation (Skamdalen 8

and Tavlefjellet members). We selected thelodont and undescribed ?chondrichthyan 9

scales, which allowed us to define potential taxonomic, histological, and taphonomic 10

variables of the REE uptake. Results showed REE concentrations to be relatively 11

uniform within the scales of each taxon, but apparent discrepancies were visible 12

between the studied localities and separate taxa. The compilation of REE abundance 13

patterns as well as REE ratios have revealed that thelodont and ?chondrichthyan 14

originating from the same locality, must have had different burial and early diagenetic 15

histories. The shapes of the REE profiles, together with the presence and absence of 16

the Eu and Ce anomalies, equally suggested different depositional and diagenetic 17

environments for these two sympatric taxa resulting from either stratigraphical or 18

long-distance reworking. The REE concentrations appear to have visible differences 19

between separate dental tissues, particularly between enameloid and dentine of 20

thelodonts, emphasising the importance of in-situ measurements in microfossil 21

biomineral geochemistry. 22

Key Words:Rare earth elements, paleoenvironment, vertebrate microfossils, 23

24

Devonian, Svalbard 25

26

Page 2 of 27



Draft

3

1

Introduction 2

3

Rare earth element (REE) compositions from fossil vertebrate bioapatite are 4

known to be diagnostic of fossil provenance and reworking (e.g. Trueman 1999; 5

Suarez et al. 2010), and have been used as proxies for palaeoseawater chemistry 6

and redox conditions helping to accomplish palaeoenvironmental and 7

palaeogeographic interpretations (e.g., Wright et al. 1987; Reynard et al. 1999; 8

Trueman & Tuross 2002; Kemp & Trueman 2003; Lécuyer et al. 2004; Patrick et al. 9

2004; Ounis et al. 2008; Žigaitė et al. 2015). REE concentrations in living fish bones 10

and teeth are very low, since these are not important elements for the metabolic 11

processes of vertebrates (e.g. Elderfield and Pagett 1986; Vennemann et al. 2001). 12

Fossil bioapatites in contrast, display REE concentrations with an enrichment factor 13

of about 106 to 107, (see e.g. Bernat 1975; Wright et al. 1984; Trueman and Tuross 14

2002). REE patterns are usually strictly similar to the seawater patterns, this is 15

achieved through quantitative intake of REE with little or no fractionation, as is the 16

case for all biogenic phosphates at the seawater-sediment interface (see Reynard et 17

al. 1999). Therefore, fossil bioapatite REE and trace element concentrations are 18

exclusively diagenetic and get incorporated post-mortem usually within a few 19

thousand years during early stages of diagenesis (see e.g. Elderfield and Paget 20

1986; Patrick et al. 2004; Tütken et al. 2008, Trueman et al. 2011). Because of this, 21

their compositions reflect bottom and pore-water chemistry and can thus be used as 22

proxies for palaeoenvironment (see Wright et al. 1987; Kemp and Trueman 2003; 23

Lécuyer et al. 2004). 24

The preservation of a palaeoceanographic signal in fossil biominerals 25

depends on the extent of diagenetic alteration and fractionation of REEs during 26

recrystallization or ‘‘extensive diagenesis’’ (Reynard et al. 1999; Lécuyer et al. 2004). 27

Extensive re-crystallization is often characterized by a strong middle REE (MREE) 28

Page 3 of 27



Draft

4

enrichment relative to light REE (LREE) and heavy REE (HREE), leading to ‘‘bell-1

shaped’’ patterns (Lécuyer et al. 1998; 2004). Recently, Herwartz et al. (2011) have 2

demonstrated that the bioapatite of vertebrate long bones, such as femora, humeri 3

and ribs, composed of fine, nanometer-size plate-like crystallites with a large 4

surface/volume ratio, shows open system behavior in respect to REE and Hf uptake, 5

and therefore should not be used for REE analyses aiming at palaeoenvironmental 6

reconstructions, and equally not for 176Lu-176Hf radiometric dating (Trueman 1999; 7

Herwartz et al. 2013a). However, fossil dental tissues such as enamel and enameloid 8

with compact and large crystallites and as little as 1 wt% of organic matter, 9

demonstrate high potential to preserve an early REE signal (Trueman and Palmer 10

1997; Kohn et al. 1999; Kemp and Trueman 2003; Trotter and Eggins 2006; Tütken 11

et al. 2008; Sire et al. 2009; Trueman et al. 2011; Enax et al. 2012) and can be used 12

as proxies for early depositional environment. For example, the redox conditions of 13

marine palaeobasins can be determined from the presence or absence of a cerium 14

(Ce) anomaly. Its occurrence in REE patterns of sedimentary rocks has been 15

identified as an indicator of anoxia (or absence of anoxia) in the bottom seawater or 16

pore waters of the basin sediment (e.g. Wright et al. 1987; Bau and Dulski, 1996; 17

Reynard et al. 1999). Nevertheless, it must be used with caution in 18

paleoenvironmental interpretations: a negative Ce anomaly is a strong indicator of an 19

oxic depositional environment, but the lack of a Ce anomaly or a positive Ce anomaly 20

indicate reductive, anoxic conditions, which also may reflect solely the pore water 21

anoxia (Kemp and Trueman, 2003; Johannesson et al. 2006; Planavsky et al. 2009). 22

REE patterns can also serve as indicators of water salinity. Modern seawater REE 23

patterns are known to display negative Ce anomalies, and enrichment in HREE (see 24

de Baar et al., 1991), while flat REE profiles suggest fresh groundwater (cold) or 25

riverine environments (Yuan et al 2014), and positive La anomalies have been 26

reported in rivers (Kulaksiz and Bau, 2011). However, it is important to emphasize 27

that a fossil dental tissue REE record can be interpreted as a proxy for the 28

Page 4 of 27



Draft

5

palaeoenvironment only if it can be demonstrated that the late diagenetic contribution 1

of REE is relatively minor (e.g. Reynard et al. 1999; Trueman 1999; Patrick et al. 2

2004; Fadel et al., 2015; Žigaitė et al. 2013; 2015). 3

In this work we analysed dental tissues of early vertebrate 4

microremains from the Lower Devonian basin of Andrée Land (Svalbard), in an 5

attempt to retrieve geochemical information about their early depositional 6

environment and potential reworking. Two different localities of marginal marine to 7

freshwater sediments from geographically distant parts of the same Grey Hœk 8

Formation were sampled. To be able to identify potential taxonomic, histological, and 9

taphonomic variables of the REE uptake, we have chosen vertebrate microfossils 10

with two contrasting types of histology: thelodont scales and probable 11

?chondrichthyan scales (see also Žigaitė 2013a,b). REE analyses have been 12

performed in-situ, using laser ablation inductively coupled plasma mass spectrometry 13

(LA-ICP-MS), in order to obtain intra-tissue compositions, which required 14

micrometer-scale spatial resolution. The measured REE values are explored below 15

using basic geochemical calculations and quantifications, with the aim of identifying 16

the presence of reworking and the nature of early-burial palaeoenvrionmental 17

conditions. 18

19

20

Geological setting and material 21

22

The thelodont and ?chondrichthyan scales used in this study come from the 23

Andrée Land territory in the northern part of Spitsbergen Island, Svalbard 24

archipelago. Stratigraphically the material originates from the Lower Devonian Old 25

Red Sandstone succession referred to as the Andrée Land Group (Blomeier et al. 26

2003), and represents deposition in a continental rift basin along the northern margin 27

of the Old Red Sandstone (ORS) landmass. The succession is essentially confined 28

Page 5 of 27



Draft

6

to a major graben with a unique depositional history, involving a shift from coarse 1

clastic red-beds, mainly of alluvial fan and fluvial origin, to a series of more greyish 2

fluvial and possibly deltaic sediments illustrating the transition from the southern arid 3

zone to the equatorial tropics. The nature of the basin and the palaeoenvironmental 4

conditions are as yet poorly understood, although it plays an important role as a 5

regional niche and separate biogeographical province in the Early Devonian. 6

Vertebrate microfossils are quite common in the Andrée Land deposits, and 7

include isolated micromeric elements of the dermal exoskeleton (dental scales or 8

dermal denticles, see Žigaitė et al. 2013b) of acanthodians, chondrichthyans, and 9

thelodonts (Ørvig 1969; Blieck et al. 1987; Blom and Goujet 2002; Žigaitė et al. 10

2013a). Scales analysed comprise two taxa, the thelodont Talivalia svalbardia and 11

an undescribed probable ?chondrichthyan, both of which come from the Grey Hœk 12

Formation in the upper part of the Andrée Land Group succession. The Formation is 13

Early to Middle Devonian in age, and consists mainly of grey and black sandstones 14

and siltstones, interbedding with calcareous claystones and siltstones with carbonate 15

nodules (Blomeier et al. 2003). It is subdivided into three lithographical units: the 16

Verdalen, Skamdalen and Tavlefjellet members (Blomeier et al. 2003; Volohonsky et 17

al. 2008). Scales of T. svalbardia come from the two latter members: the 18

Gråkammen locality, Skamdalen member, which represents the Lower Grey Hœk 19

Formation, and the Tavlefjellet locality, Tavlefjellet member, which represents the 20

Middle Grey Hœk Formation (Žigaitė et al. 2013a). The ?chondrichthyan scales 21

originate from the Gråkammen locality (Žigaitė et al. 2013a). Material used for this 22

study was obtained from the palaeontological collections of the Paris National 23

Natural History Museum (Museum national d’Histoire naturelle), France. 24

25

Sampling and analytical techniques 26

27

Page 6 of 27



Draft

7

Considering the nature of in-situ microanalyses, we have chosen to analyse 1

only post-pectoral scales of the thelodonts, and ‘body’ scales of chondrichthyans, 2

which are the most common in the dermal exoskeleton of these taxa, and as a result, 3

the most abundant microremains in the fossil record. Post-pectoral scales are also 4

less fragile than other types of thelodont scales, such as cephalo-pectoral, precaudal 5

and pinnal scales (e.g. Žigaitė and Goujet, 2012; Žigaitė 2013), which facilitate the 6

process of grinding and polishing. Preparation of the samples was conducted in the 7

Imaging and Analysis Center (Natural History Museum of London), apart from the 8

extraction of the microremains by mechanical and acetic acid preparation of 9

calcareous sandstones, and transmitted light microscopy, which were performed at 10

the Paris National Natural History Museum (Museum national d’Histoire naturelle). 11

Prior to analyses, two-dimentional (2-D) sections of the scales were obtained from 12

vertical anterior – posterior cuts along the longitudinal axis of dental scales (Fig. 2). 13

The scales were embedded in araldite epoxy resin (Buehler epoxy resin with a mix of 14

50g of resin to 1 g of hardener), and prepared through a series of grinding and 15

polishing steps similar to the protocol described in Pérez-Huerta and Cusack (2009), 16

with a final 5 min treatment of 0.06 µm colloidal silica. Finally, samples were cleaned 17

in an ultrasonic bath and dried at room temperature. 18

Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS) 19

The compositions of trace and rare earth elements (REE) were obtained by 20

LA-ICPMS at the Imaging and Analysis Center of Science Facilities Department, the 21

Natural History Museum, London (UK). Analyses were performed using a New Wave 22

Research NWR193 193nm excimer laser ablation accessory coupled to an Agilent 23

Technologies 7500cs ICP-MS. Data were acquired for 120s at each analysis site, 24

taking individual points in different scale tissue regions (dentine or enameloid). 25

Background signals were collected for the first ca 60s and the laser fired at the 26

sample to collect sample signals for the remaining acquisition time. Data were 27

collected using the time resolved method and were processed offline using 28

Page 7 of 27



Draft

8

LAMTRACE software (Simon Jackson, Natural Resources, Ottawa, Canada). 1

Elemental concentrations were calculated using the National Institute of Standards 2

and Technology (NIST) standard reference material 612 for calibration and calcium 3

was used for internal standardization. The limit of detection was taken as 1σ of the 4

mean background count, and the data filtered at twice this limit (2σ). Calculated 5

precision was better than 3% RSD (at 1σ error) when using 43Ca as internal 6

standard. The REE concentrations were measured in parts per million (ppm), see 7

Table 1, and were normalized to both Orgueil (CI chondrite) concentrations (Palme 8

and Jones 2005), and to Post-Archean Australian Shale (PAAS) concentrations 9

(McLennan 1989). For the calculations in this work we used shale-normalised 10

(PAAS) values, since as a sedimentary rock it represents closer overall REE 11

concentrations, and is commonly applied in fossil bio-mineral studies (see Kemp and 12

Trueman 2003; Pucéat et al. 2004; Trotter and Eggins 2006; Tütken et al. 2008; 13

Kocsis et al. 2010; Herwartz et al. 2013b). 14

15

Results 16

17

REE compositions were quite distinct between the two localities and different 18

taxa, but showed uniform values within each taxon and within each dental tissue 19

analysed (Fig 3, A-D). Offsets in absolute REE concentrations were visible between 20

thelodont dentine and enameloid, dentine REE values being distinctively higher 21

(Fig.3,A,C). Unlike thelodonts, the chondrichthyan scales did not display any 22

significant inter-tissue variability of the REE abundances, except for a variable 23

degree of Eu depletion correlated with different types of tubular dentine (see Žigaitė 24

et al., 2013b) (Fig.2; Fig.3,B). In addition, the overall REE concentrations of the 25

chondrichthyan scales were greater than those of the T. svalbardia scales from the 26

same locality (Gråkammen, Fig.3,D). Only the Lanthanum (La) concentrations in both 27

T. svalbardia and chondrichthyan scales were similar, ranging from 100 to 180 ppm, 28

Page 8 of 27



Draft

9

while all the other REE concentrations and normalised abundances were 3 to 10 1

times greater in the chondrichthyan scales, if compared to the thelodonts 2

(Fig.3,A,B,D). 3

REE abundance patterns 4

Addressing the aforementioned inter-tissue discrepancies in absolute 5

concentrations of REE, we have used shale-normalized REE values to produce 6

tissue-selective plots for each studied scale. The resulting REE abundance patterns 7

revealed significant differences between the two localities, Tavlefjellet and 8

Gråkammen, and between the two different taxa from the Gråkammen locality. 9

?Chondrichthyan REE patterns were “bell-shaped”, displaying the enrichment in 10

medium rare earths (MREE) (Fig.2,A,B), similar to more moderate “bell-shaped” 11

patterns of T. svalbardia from Tavlefjellet. T. svalbardia from Gråkammen had flat 12

profiles, and much lower overall REE concentrations (Fig.2,A,C). At the same time, 13

REE patterns of T. svalbardia from both Tavlefjellet and Gråkammen displayed 14

significant positive Europium (Eu) anomalies (Fig.2,A,C), compared to the 15

?chondrichthyans from Gråkammen, the REE patterns of which did not show any 16

positive Eu anomaly, but on the contrary a negative one of a lesser degree (Fig.2,B). 17

REE ratios 18

We have calculated ratios for the Cerium (Ce) anomaly 19

([Ce/Ce*=Ce/(0.5La+0.5Pr)]SN where SN refers to Shale Normalised) and the 20

Praseodymium (Pr) anomaly ([Pr/Pr*=Pr/(0.5Ce+0.5Nd)]SN) in an attempt to reveal 21

the redox potential of the bottom waters and sediment pore waters (after Bau and 22

Dulski 1996). The ratios of T. svalbardia thelodont scales from the Tavlefjellet locality 23

lacked both Ce and La anomalies (Fig.4, field I), whereas T. svalbardia from 24

Gråkammen locality displayed a positive La anomaly (Fig.4, field IIa). The ratios of 25

the ?chondrichthyan scales from Gråkammen were very uniform, plotting outside the 26

fields of positive Ce or positive La anomalies (Fig. 4). 27

28

Page 9 of 27



Draft

10

Discussion 1

Our results suggest that the REE concentrations and normalized abundance 2

patterns obtained from early vertebrate microremains do reflect the taphonomic 3

conditions and early diagenetic history. The REE compositions do not show any 4

specific taxonomic behavior, but have visible inter-tissue differences reflecting the 5

biomineral structure and compactness of the studied hard tissues. In the Gråkammen 6

locality, two different diagenetic settings for the thelodonts and chondrichthyans can 7

be suggested from the presence of a positive Eu anomaly in T. svalbardia but not in 8

the ?chondrichthyan scales. The reduced Eu2+ of the pore waters is known to 9

partition strongly into calcite producing a positive Eu anomaly, which leaves pore 10

waters depleted in Eu leading to negative Eu anomalies (Trueman et al., 2003). 11

Positive Eu anomalies are quite unusual in sediments, however, its presence in fossil 12

bioapatite would suggest locally reducing conditions, most likely caused by the 13

microbial decomposition of organic matter (Trueman et al., 2003; Fadel et al., 2015). 14

In oxidized environments, the Eu is present as a trivalent cation (Eu3+) non soluble in 15

water, while in a reductive environment, the bivalent form (Eu2+) is mobile and can be 16

transported by diagenetic pore waters. The pore water enriched in Eu2+ can 17

reprecipitate under low oxygen content and create a positive Eu anomaly. This 18

process is rare but has been reported in Pleistocene muds of the Amazon deep-sea 19

fans and in a few other similar depositional environments (MacRae et al. 1992). The 20

rarity of these processes of dissolution and reprecipitation, however, suggests that 21

Eu anomalies cannot alone be used to assess the reductive or oxidizing conditions 22

(Martinez-Ruiz et al. 1999), but can indicate the difference in diagenetic histories, as 23

in our case between the chondrichthyan and thelodont microremains, originating 24

from the same locality (Fig.2,A,B). 25

The shapes of the REE patterns are known to reflect certain 26

palaeoenvironmental conditions (e.g. Reynard et al. 1999; Trueman & Turros, 2002; 27

Patrick et al., 2004; Ounis et al., 2008; Bright et al., 2009). The MREE enriched, or 28

Page 10 of 27



Draft

11

“bell-shaped” REE patterns are most common and typical for Palaeozoic ichthyoliths, 1

such as conodonts and fish remains (e.g. Wright et al., 1987; Trotter & Eggins, 2006; 2

Bright et al., 2009), and are suggested to arise from the equilibrium fractionation 3

between seawater and biogenic apatite, serving as an indicator of marine 4

environment in Palaeozoic (e.g. Wright et al., 1987; Trueman & Tuross, 2002). This 5

assumption is made in spite of the fact that MREE enrichment is not observed in 6

modern oceanic waters (e.g. Byrne & Sholkovitz, 1996), a discrepancy that is 7

attributed to major differences in oceanic water chemistry before and after the 8

Cretaceous (e.g. Picard et al., 2002; Bright et al., 2009). On the other hand, MREE 9

enriched “bell-shaped” patterns of fossil bioapatites have been often explained as a 10

result of extensive, late diagenetic REE uptake, predominately via preferential 11

substitution of Ca2+ with MREEs (e.g. Raynard et al., 1999; Lécuyer et al., 1998; 12

2004). However, it has been shown in a number of studies, that the authigenic 13

phosphate precipitates become enriched in MREE already during the early 14

diagenesis (Bryne et al., 1996; Rasmussen et al., 1998). 15

The REE patterns of the T. svalbardia from Tavlefjellet, as well as of the 16

?chondrichthyan from the Gråkammen locality (Fig.2,C,B) are both MREE and 17

slightly HREE enriched, showing clear “bell-shaped” REE patterns. This is however 18

not the case for the T. svalbardia from Gråkammen, which shows predominantly flat 19

REE profiles, suggesting different taphonomic conditions, and – as long as the 20

marine signature of the Palaeozoic “bell-shaped” patterns is accepted - potentially 21

different habitats of these two different taxa from Gråkammen. However, in that case 22

one would expect certain similarities between the patterns of T. svalbardia in both 23

localities, which, apart of the common pronounced positive Eu anomaly, are 24

otherwise quite different (Fig. 2A,C). This gives evidence either for a geological age 25

difference between the two taxa, implying stratigraphical reworking, or otherwise for 26

long-distance reworking between distinct environments. The Gråkammen material of 27

T. svalbardia shows low absolute REE concentrations with flattened REE profiles 28

suggesting strong freshwater influence (see Kulaksis and Bau, 201; Yuan et al., 2014). 29

This matches the non-marine origin of the sedimentary sequence at the locality 30

(Blomeier et al., 2003). By contrast, the ?chondrichthyan scales from Gråkammen 31

Page 11 of 27



Draft

12

have higher overall REE levels and an REE profile that suggests deposition in a 1

marine environment. This could be interpreted as reflecting reworking of the 2

?chondrichthyan scales from an original marine depositional environment followed 3

by redeposition of these scales in the non-marine sequence of Gråkammen. 4

However, the above interpretation can be justified only if we consider early uptake of 5

the REE for both taxa, and therefore their profiles to reflect the early depositional 6

palaeoenvironment. 7

The (Ce/Ce*)SN ratio calculations revealed no presence of a Ce anomaly in 8

any of the studied scales (Fig.4), which might be indicative of slightly anoxic 9

conditions during the REE uptake, and accords with the above discussed Eu 10

anomaly. The T. svalbardia from Gråkammen locality displayed positive La anomaly, 11

which together with flattened REE profiles suggest a freshwater, possibly riverine 12

environment (e.g. Kulaksis and Bau, 2011). This, in adition to the difference in the 13

shapes of the REE patterns, as well as Eu anomaly (Fig. 3,D), would suggest 14

reworking and certainly separate diagenetic histories for the two taxa from 15

Gråkammen. 16

To conclude, the REE geochemistry in studied Lower Devonian early 17

vertebrate microremains proves to be a useful proxy for determining reworking in 18

vertebrate microfossil assemblage as deep in time as the Early Devonian, and to a 19

certain extent is sugestive of palaeosalinity and palaeoredox conditions of the bottom 20

water. It is important to add that separate hard tissues, particularly thelodont dentine 21

and enameloid, show notable quantitative offsets in overall REE concentrations, and, 22

in the case of the ?chondrichthyans, even minor qualitative inter-tissue differences. 23

This emphasises the importance of selected in-situ measurements when analysing 24

fossil biominerals, if we want to obtain sound geochemical proxies. 25

26

27

Acknowledgments 28

Page 12 of 27



Draft

13

We would like to thank Prof Gaël Clément and Prof Philippe Janvier (Muséum 1

national d'Histoire naturelle, Paris) for providing fossil material, equipment and 2

curatorial support for this project. We are grateful to Dr Kévin Lepot (Université Lille-3

1) for valuable discussion and advices. This work is also a contribution to the 4

UNESCO and IUGS International Geosciences Programme project IGCP 591 “Early 5

to Middle Palaeozoic Revolution”. The research has been funded by a Young 6

Researcher Grant from the Swedish Research Council, awarded to Dr Ž. Žigaitė, and 7

a Wallenberg Scholarship from the Knut and Alice Wallenberg Foundation awarded 8

to Prof P.E. Ahlberg. 9

10

Page 13 of 27



Draft

14

1

2

References 3

4

Bau, M. and Dulski, P. 1996. Distribution of yttrium and rare-earth elements in the Penge and 5

Kuruman iron-formations, Transvaal Supergroup, South Africa. Precambrian Research, 79: 6

37–55. 7

8 Bernat, M. 1975. Les isotopes de l’uranium et du thorium et les terres rares dans 9

l’environment marin. Cah. OSTROM, Serie Geologique, 7, 65-83. 10

11

Blieck, A., Goujet, D. and Janvier, P. 1987. The vertebrate stratigraphy of the Lower 12

Devonian (Red Bay Group and Wood Bay Formation) of Spitsbergen. Modern Geology, 11: 13

197-2017. 14

15

Blom, H., Goujet, D. 2002. Thelodont scales from the Lower Devonian Red Bay Group. 16

Spitsbergen, Palaeontology 45, 795-820. 17

18

Blomeier, D., Wisshak, M., Dallmann, W., Volohonsky, E. and Freiwald, A. 2003. Facies 19

analsysis of the Old Red Sandstone of Spitsbergen (Wood Bay Formation): Reconstruction of 20

the depositional environments and implications of basin development. Facies, 49: 151–174. 21

22 Byrne, R.H.,, Scholkovitz, E.R. 1996. Marine chemistry and geochemistry of the lanthanides. 23

In : K.A. Geschneider Jr. and L. Eyring, The Handbook on the Physics and Geochemistry of 24

the Rare Earths. Elsevier, Amsterdam, The Netherlands, pp. 497-593. 25

26

Byrne, R.H., Liu, X., and Schijf, J. 1996. The influence of phosphate coprecipitation on rare 27

earth distribution in natural waters. Geochimica Cosmochimica Acta 60, 3341-3346. 28

29

Page 14 of 27



Draft

15

Bright, C.A., Cruse, A.M., Lyons, T.W., Mac Leod, K.G., Glascock, M.D., Ethington, R.L. 1

2009. Seawater rare-earth element patterns preserved in apatite of Pennsylvanian 2

conodonts? Geochimica Cosmochimica Acta 73, 1609–1624. 3

Elderfield, H., Pagett, R. 1986. Rare earth elements in ichthyoliths: Variations with redox 4

conditions and depositional environment. Science of The Total Environment, 49: 175-197. 5

6

Enax, J., Prymak, O., Raabe, D. and Epple, M. 2012. Structure, composition, and mechanical 7

properties of shark teeth. Journal of Structural Biology, 178: 290–299. 8

9 Fadel, A., Žigaitė, Ž., Blom, H., Pérez-Huerta, A., Jeffries, T., Märss, T., Ahlberg, P.E. 2015. 10

Palaeoenvironmental signatures revealed from the rare earth element (REE) compositions in 11

vertebrate microremains of Vesiku bone bed (Homerian, Wenlock), Saaremaa island, 12

Estonia. In Histon, K., Žigaitė, Ž. (eds.) Estonian Journal of Earth Sciences, 64(1), 36-41. 13

14

Herwartz, D., Tütken, T., Münker, C., Jochum, K. P., Stoll, B., Sander, P. M. 2011. Rare earth 15

element systematics of fossil bone revealed by LA-ICPMS analysis. Geochimica 16

Cosmochimica Acta, 75: 82–105. 17

18 Herwartz D., Münker, C., Tütken T., Hoffmann, J.E., Wittke, A., Barbier, B. 2013a. Lu-Hf 19

isotope systematics of fossil biogenic apatite and their effects on geochronology. Geochimica 20

Cosmochimica Acta, 101: 328–343. 21

22 Herwartz, D., Münker, C., Tütken, T., Hoffmann, J.E., Wittke, A., Barbier, B. Jochum, K. P. 23

and Sander, P. M. 2013b. Rare earth element systematics of fossil bone revealed by LA-24

ICPMS analysis. Geochimica Cosmochimica Acta, 103: 161–183. 25

26 Johannesson, K. H., Hawkins, Jr. D. L. and Cortés, A. 2006. Do Archean chemical sediments 27

record ancient seawater rare earth element patterns? Geochimica Cosmochimica Acta, 70: 28

871–890. 29

30 Kemp, R. A. and Trueman, C. N. 2003. Rare earth elements in Solnhofen biogenic apatite: 31

geochemical clues to the palaeoenvironment. Sedimentary Geology ,155: 109–127. 32

Page 15 of 27



Draft

16

1

Kocsis L., Trueman C. N. and Palmer M. R. (2010) Protracted diagenetic alteration of REE 2

contents in fossil bioapatites: Direct evidence from Lu–Hf isotope systematics. Geochimica 3

Cosmochimica Acta 74, 6077–6092. 4

5

Kohn, M. J., Schoeninger, M. J. and Barker, W. W. 1999. Altered States: Effects of 6

diagenesis on fossil tooth chemistry. Geochimica Cosmochimica Acta, 63: 2737–2747. 7

8

Kulaksiz, S. and Bau, M. 2011. Rare earth elelements in the Rhine River, Germany: First 9

case of anthropogenic lanthanum as a dissolved microcontaminant in the hydrosphere. 10

Environment International, 37, 973-979. 11

12

Lécuyer, C., Grandjean P., Barrat, J.-A., Nolvak, J., Emig, C.C., Paris, F., Robardet, M. 1998. 13

D18O and REE contents of phosphatic brachiopods: a comparison between modern and 14

lower Palaeozoic populations. Geochimica Cosmochimica Acta 62, 2429–2436. 15

16

Lécuyer, C., Reynard, B. and Grandjean, P. 2004. Rare earth element evolution of 17

Phanerozoic seawater recorded in biogenic apatites. Chemical Geology, 204: 64–102. 18

19

MacRae, N. D., Nesbitt, H. W. and Kronberg, B. I. 1992. Development of a positive Eu 20

anomaly during diagenesis. Earth ans Planetarty Science Letters, 109: 585–591. 21

22

Martinez-Ruiz, F., Ortega-Huertas, M. and Palomo, I. 1999. Positive Eu anomaly 23

development during diagenesis of the K/T boundary ejecta layer in the Agost section (SE 24

Spain): implications for trace-element remobilization. Terra Nova, 11: 290–296. 25

26

McLennan, S. M. 1989. Rare earth elements in sedimentary rocks; influence of provenance 27

and sedimentary processes. Reviews in Mineralogy and Geochemistry, 21: 169–200. 28

29

Page 16 of 27



Draft

17

Ørvig, T. 1969. Vertebrates from the Wood Bay Group and the position of the Emsian-Eifelian 1

boundary in the Devonian of Vestspitsbergen. Lethaia, 2: 273–328. 2

3 Ounis, A., Kocsis, L., Chaabani, F., Pfeifer, H-R. 2008. Rare earth elements and stable 4

isotope geochemistry (d13C and d18O) of phosphorite deposits in the Gafsa Basin, Tunisia. 5

Palaeogeography, Palaeoclimatology, Palaeoecology, 268, 1-18. 6

7

Palme, H. and Jones, A. 2005. Solar System Abundances of the Elements. In Meteorites, 8

Comets, and Planets, Treatise on Geochemistry, v1. Edited by A.M. Davis, Elsevier, 9

Amsterdam. pp. 41–62. 10

11 Patrick, D., Martin, J.E., Parris, D.C., Grandstaff, D.E. 2004. Palaeoenvironmental 12

interpretations of rare earth element signatures in mosasaurs (reptilian) from the upper 13

Cretaceous Pierre Shale, central South Dakota, USA. Palaeogeography, Palaeoclimatology, 14

Palaeoecology, 212, 277-294. 15

16

Pérez-Huerta, A. and Cusack, M. 2009. Optimising electron backscatter diffraction (EBSD) of 17

carbonate biominerals – resin type and carbon thickness. Microscopy and Microanalysis, 15: 18

197–203. 19

20 Picard, S., Lécuyer, C.,Barrat, J.-A., Garcia, J.-P., Dromart, G., Sheppard, S.M.F. 2002. Rare 21

earth element contents of Jurassic fish and reptile teeth and their potential relation to 22

seawater composition (Anglo-Paris basin, France and England). Chemical Geology, 186, 1-23

16. 24

25

Planavsky, N., Rouxel, O., Bekker, A., Shapiro, R., Fralick, P. and Knudsen, A. 2009. Iron-26

oxidizing microbial ecosystems thrived in late Paleoproterozoic redox-stratified oceans. Earth 27

and Planetary Science Letters, 286: 230–242. 28

29

Pucéat, E., Reynard, B. and Lécuyer, C. 2004. Can crystallinity be used to determine the 30

degree of chemical alteration of biogenic apatites? Chemical Geology, 205: 83–97. 31

Page 17 of 27



Draft

18

1

Rassmussen, B., Buick, R., Taylor, W.R. 1998. Removal of oceanic REE by authigenic 2

precipitation of phosphatic minerals. Earth and Planetary Science Letters, 164: 135-149. 3

4

5

Reynard, B., Lécuyer, C. and Grandjean, P. 1999. Crystal-chemical controls on rare-earth 6

element concentrations in fossil biogenic apatites and implications for paleoenvironmental 7

reconstructions. Chemical Geology, 155: 233–241. 8

9

Sire, L.-Y., Donoghue, P.C.J. and Vickaryous, M.K. 2009. Origin and evolution of the 10

integumentary skeleton in non-tetrapod vertebrates. Journal of Anatomy, 214: 409-440. 11

12

Suarez, C.A., Macpherson, G.L., Gonzalez, L.A., Grandstaff, D.E. 2010. Heterogeneous rare 13

earth element (REE) patterns and concentrations in a fossil bone: Implications for the use of 14

REE in vertebrate taphonomy and fossilization history. Geochimica et Cosmochimica Acta 74: 15

2970-2988. 16

17

Trotter, J. A. and Eggins, S. M. 2006. Chemical systematic of conodont apatite determined by 18

laser ablation. Chemical Geology, 233: 196–216. 19

20

Trueman, C. N. 1999. Rare Earth Element Geochemistry and Taphonomy of Terrestrial 21

Vertebrate Assemblages. Palaios, 14: 555–568. 22

23

Trueman, C. N. and Palmer, M. R. 1997. Diagenetic Origin of REE in Vertebrate Apatite: A 24

Reconsideration of Samoilov and Benjamini, 1996. Palaios, 12: 495–497. 25

26

Trueman, C.N., Tuross, N. 2002. Trace elements in recent and fossil bone apatite. 27

Phosphates: geochemical, geobiological, and materials importance. Reviews in Mineralogy 28

and Geochemistry. Americal Mineralogical Society, Washington D.C., vol. 48, 742 pp. 29

Page 18 of 27



Draft

19

1

Trueman, C.N., Benton, M.J., Palmer, M.R. 2003. Geochemical taphonomy of shallow marine 2

vertebrate assemblages. Palaeogeography, Palaeoclimatology, Palaeoecology, 197(3-4): 3

151–169. 4

5

Trueman, C. N., Kocsis, L., Palmer, M. R. and Dewdney, C. 2011. Fractionation of rare earth 6

elements within bone mineral: A natural cation exchange system. Palaeogeography, 7

Palaeoclimatology, Palaeoecology, 310: 124–132. 8

9 Tütken, T., Vennemann T. W. and Pfretzschner, H.-U. 2008. Early diagenesis of bone and 10

tooth apatite in fluvial and marine settings: Constraints from combined oxygen isotope, 11

nitrogen and REE analysis. Palaeogeography, Palaeoclimatology, Palaeoecology, 266: 254–12

268. 13

14 Vennemann, T.W., Hegner, E., Cliff, G., & Benz, G.W., 2001: Isotopic composition of recent 15

shark teeth as a proxy for environmental conditions. Geochimica Cosmochimica Acta, 65: 16

1583–1599. 17

18

Volohonsky, E., Wisshak, M., Blomeier, D., Seilacher, A. and Snigirevsky, S. 2008. A new 19

helical trace fossil from the Lower Devonian of Spitsbergen (Svalbard) and its 20

palaeoenvironmental significance. Palaeogeography, Palaeoclimatology, Palaeoecology, 267: 21

17–20. 22

23 Wright, J., Seymour, R.S., Shaw, H.F. 1984. REE and Nd isotopes in conodont apatite: 24

variations with geological age and depositional environment. In: Clarck, D.L. (Ed.), Conodont 25

Biofacies and Provincialism. Geological Society of America Special Paper, 196, 325-340. 26

27

Wright, J., Schrader, H., Holster, W.T. 1987. Paleoredox variations in ancient oceans 28

recorded by rare earth elements in fossil apatite. Geochimica Cosmichimica Acta 51, 631-29

644. 30

31

Page 19 of 27



Draft

20

Žigaitė, Ž., Goujet, D. 2012. New observations on the squamation patterns of articulated 1

exoskeletons of Loganellia scotica (Vertebrata: Thelodonti) from the Lower Silurian of 2

Scotland. Geodiversitas, 34(2), 253-270. 3

4

Žigaitė, Ž. 2013. Endemic thelodonts (Vertebrata: Thelodonti) from the Lower Silurian of 5

central Asia and southern Siberia. Earth and Environmental Science Transactions of the 6

Royal Society of Edinburgh, 104, 1-21. 7

8

Žigaitė, Ž., Karatajūtė-Talimaa, V., Goujet, D. and Blom, H. 2013a. Thelodont scales from the 9

Lower and Middle Devonian Andrée Land Group, Spitsbergen. GFF, 135, 57–73. 10

11 Žigaitė, Ž., Richter, M., Karatajutrė-Talimaa, V. and Smith, M. M. 2013b. Tissue diversity and 12

evolutionary trends of the dermal skeleton of Silurian thelodonts. Historical Biology, 25, 143–13

154. 14

15 Žigaitė, Ž., Fadel, A., Blom, H., Perez-Huerta, A., Jeffries, T., Märss, T., and Ahlberg, P.E. 16

2015. Rare earth elements (REEs) in vertebrate microremains from the upper Pridoli 17

Ohesaare beds of Saaremaa Island, Estonia: geochemical clues to palaeoenvironmnet. 18

Estonian Journal of Earth Sciences, 64, 115–120. 19

20

Page 20 of 27



Draft

21

1

Figure Captions: 2

Table 1: Shale normalised (PAAS) REE values and Ce/Ce* and Pr/Pr* ratios for the 3

studied scales. PAAS values were taken from McLennan (1989): 38.2 (La), 79.6 4

(Ce), 8.83 (Pr), 33.9 (Nd), 5.55 (Sm), 1.08 (Eu), 4.66 (Gd), 0.774 (Tb), 4.68 (Dy), 5

0.991 (Ho), 2.85 (Er), 0.405 (Tm), 2.82 (Yb), 0.433 (Lu). Equations used for the 6

calculation of ratios: Ce/Ce* = [Ce/(0.5La + 0.5Pr)]SN. Pr/Pr* = [Pr/(0.5Ce + 7

0.5Nd)]SN (after Bau and Dulski, 1996). 8

9

Fig. 1: Map and stratigraphical framework of Andrée Land, Spitsbergen indicating the 10

localization of the different samples. 11

12

Fig 2: Laser ablation ICP-MS analysis spots: analysed dermal scale sections of (A) 13

Talivalia svalbardia from Gråkammen locality, four (=4) datapoints; (B) Talivalia 14

svalbardia from Tavlefjellet locality, seven (=7) datapoints, and (C) the 15

?chondrichthyan from Gråkammen locality, eight (=8) datapoints. Total of nineteen 16

(=19) datapoints, for histological definition and numbering see Table 1. Scale bars = 17

0.2 mm. 18

19

Fig. 3: Shale normalised (PAAS) REE abundance patterns of (A) Talivalia svalbardia 20

scales from Gråkammen locality, (B) ?chondrichthyan scales from Gråkammen 21

locality, (C) Talivalia svalbardia scales from Tavlefjellet locality. For A and C, solid 22

lines represent dentine datapoints, dashed lines – enameloid; for B, solid lines 23

represent thick tubule dentine datapoints, dashed lines – thin tubule dentine. (D) 24

Comparison of the three taxa, enameloid and thin tubule dentine datapoint only. For 25

clarity, solid lines in part D correspond to the T.svalbardia profiles (enameloid only), 26

and dashed lines to the profiles of ?chondrichthyan (thin tubule dentine only). 27

28

Page 21 of 27



Draft

22

Fig. 4: (Ce/Ce*)SN vs. (Pr/Pr*)SN diagram, where (Ce/Ce*)SN = [Ce/(0.5La + 0.5Pr)]SN, 1

and (Pr/Pr*)SN = [Pr/(0.5Ce + 0.5Nd)]SN, ratio anomaly calculations for the two taxa, 2

and two localities, using enameloid REE concentrations only. TAV- Tavlefjellet; GRÅ- 3

Gråkammen. Field I: neither CeSN nor LaSN anomaly; field IIa: a positive LaSN 4

anomaly, no CeSN anomaly; field IIb: a negative LaSN, no CeSN anomaly;; IIIa: a 5

positive CeSN anomaly; field IIIb: a negative CeSN anomaly. After Bau and Dulski, 6

1996. 7

8

9

10

Page 22 of 27



Draft

Fig. 1: Map and stratigraphical framework of Andrée Land, Spitsbergen indicating the localization of the different samples.

82x78mm (300 x 300 DPI)

Page 23 of 27



Draft

Fig 2: Laser ablation ICP-MS analyses spots: analysed dermal scale sections of (A) Talivalia svalbardia from Gråkammen locality, four (=4) datapoints; (B) Talivalia svalbardia from Tavlefjellet locality, seven (=7)

datapoints, and (C) the ?chondrichthyan from Gråkammen locality, eight (=8) datapoints. Total of nineteen (=19) datapoints, for histological definition and numbering see Table 1. Scale bars = 0.2 mm.

182x45mm (300 x 300 DPI)

Page 24 of 27



Draft

Fig. 3: Shale normalised (PAAS) REE abundance patterns of (A) Talivalia svalbardia scales from Gråkammen locality, (B) ?chondrichthyan scales from Gråkammen locality, (C) Talivalia svalbardia scales from

Tavlefjellet locality. For A and C, solid lines represent dentine datapoints, dashed lines – enameloid; for B, solid lines represent thick tubule dentine datapoints, dashed lines – thin tubule dentine. (D) Comparison of the three taxa, enameloid and thin tubule dentine datapoint only. For clarity, solid lines in part D correspond to the T.svalbardia profiles (enameloid only), and dashed lines to the profiles of ?chondrichthyan (thin tubule

dentine only). 182x136mm (300 x 300 DPI)

Page 25 of 27



Draft

- - - This is a copy of Figure 4 in greyscale (black & white) for printing 86x69mm (300 x 300 DPI)

Page 26 of 27



Draft

Table 1. Shale normalised (PAAS) REE values and Ce/Ce* and Pr/Pr* ratios for the studied

scales. PAAS values were taken from McLennan (1989): 38.2 (La), 79.6 (Ce), 8.83 (Pr), 33.9

(Nd), 5.55 (Sm), 1.08 (Eu), 4.66 (Gd), 0.774 (Tb), 4.68 (Dy), 0.991 (Ho), 2.85 (Er), 0.405

(Tm), 2.82 (Yb), 0.433 (Lu). Equation used for ratios: Ce/Ce* = [Ce/(0.5La + 0.5Pr)]SN.

Pr/Pr* = [Pr/(0.5Ce + 0.5Nd)]SN. (after Bau and Dulski, 1996).

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Ce/Ce* Pr/Pr*

T.svalbardia - TAV

enameloid 03 8,06 12,22 15,29 18,44 26,85 68,89 36,70 33,07 35,47 31,28 25,23 18,72 14,04 10,25 1,05 1,00

enameloid 05 8,40 12,56 17,55 19,94 28,47 71,67 40,34 35,40 38,25 32,49 25,26 19,48 13,83 9,98 0,97 1,08

enameloid 06 8,74 12,69 16,87 20,53 30,63 73,43 43,56 37,34 37,82 31,99 25,79 20,12 13,37 10,69 0,99 1,02

dentine 04 17,80 24,87 33,64 40,71 58,74 142,59 88,63 77,39 81,41 69,53 56,84 43,70 29,15 24,02 0,97 1,03

dentine 07 8,09 12,55 15,63 18,58 25,59 63,70 36,48 32,56 34,62 28,86 23,58 18,17 11,28 8,57 1,06 1,00

basal dentine 09 16,57 22,36 30,46 40,12 55,32 137,04 84,33 73,00 78,63 69,12 55,44 39,75 28,65 22,54 0,95 0,98

basal dentine 10 17,77 24,62 33,30 43,36 63,06 161,11 97,64 84,24 91,24 80,32 65,26 47,41 35,11 26,10 0,96 0,98

T.svalbardia - GRÅ

enameloid 04 3,17 2,58 2,12 2,35 2,05 6,21 2,47 2,42 2,35 2,35 2,42 2,79 2,21 2,27 0,97 0,86

enameloid 06 2,98 2,39 2,19 2,11 2,05 6,63 2,64 2,17 2,59 2,53 2,35 2,57 2,39 2,11 0,92 0,97

enameloid 07 2,72 1,97 1,82 1,71 1,57 5,09 1,96 1,60 2,09 2,02 1,80 2,16 1,77 1,64 0,87 0,99

dentine 08 4,01 3,43 3,50 3,30 3,77 10,28 4,61 3,66 3,93 3,73 4,07 3,73 2,80 2,47 0,91 1,04

?Chondrichthyan - GRÅ

thin tubule dentine 03 3,53 6,60 12,34 20,47 40,18 29,17 43,99 31,52 25,43 17,76 12,84 9,28 5,89 4,99 0,83 0,92



thick tubule dentine 06 4,48 8,09 15,29 25,99 52,07 42,22 53,65 37,34 30,77 21,59 15,58 10,84 6,95 5,45 0,82 0,90



basal dentine 11 4,82 7,91 15,63 28,32 61,08 57,96 68,03 45,35 36,11 24,72 19,05 12,96 8,87 8,04 0,77 0,86

basal dentine 12 3,98 7,45 14,27 25,93 56,94 55,37 61,59 43,54 33,12 24,22 16,56 12,15 8,48 7,67 0,82 0,86

Page 27 of 27



draft - university of toronto t-space · draft 2 1 abstract2 3 in-situ rare earth element...

Documents