Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/233944401

Whatdoestheoxygenisotopecompositionofrodentteethrecord?

ArticleinEarthandPlanetaryScienceLetters·January2013

DOI:10.1016/j.epsl.2012.09.058

CITATIONS

12

READS

182

8authors,including:

Someoftheauthorsofthispublicationarealsoworkingontheserelatedprojects:

MIS3/2paleoenvironmentalcontextforhumanoccupationsinnortheasternIberia.Paleontology,

taphonomyandoxygenisotopeanalysisappliedtosmallmammals.Viewproject

AragonQuaternaryFaunaViewproject

ChristopheLécuyer

ClaudeBernardUniversityLyon1

255PUBLICATIONS5,698CITATIONS

SEEPROFILE

SophieMontuire

EcolePratiquedesHautesEtudes

85PUBLICATIONS1,017CITATIONS

SEEPROFILE

SergeLegendre

FrenchNationalCentreforScientificResearch

145PUBLICATIONS2,171CITATIONS

SEEPROFILE

GloriaCuenca-Bescós

UniversityofZaragoza

308PUBLICATIONS4,386CITATIONS

SEEPROFILE

AllcontentfollowingthispagewasuploadedbyAurélienRoyeron03December2016.

Theuserhasrequestedenhancementofthedownloadedfile.Allin-textreferencesunderlinedinblueareaddedtotheoriginaldocument

andarelinkedtopublicationsonResearchGate,lettingyouaccessandreadthemimmediately.

Earth and Planetary Science Letters 361 (2013) 258–271

Contents lists available at SciVerse ScienceDirect

Earth and Planetary Science Letters

0012-82

http://d

n Corr

Univers

France.

E-m

journal homepage: www.elsevier.com/locate/epsl

What does the oxygen isotope composition of rodent teeth record?

Aurelien Royer a,e, Christophe Lecuyer a,f,n, Sophie Montuire b,e, Romain Amiot a, Serge Legendre a,Gloria Cuenca-Bescos c, Marcel Jeannet d, Franc-ois Martineau a

a Laboratoire de Geologie de Lyon, UMR CNRS 5276, Universite Lyon 1 et Ecole Normale Superieure de Lyon, 69622 Villeurbanne, Franceb Biogeosciences, UMR CNRS 5561, Universite de Bourgogne, 6 Boulevard Gabriel, 21000 Dijon, Francec Departamento de Ciencias de la Tierra, Area de Paleontologıa, Edificio de Geologicas, 50009 Zaragoza, Spaind LAMPEA, UMR 6636, MMSH, 5 rue du Chateau de l’Horloge, BP 647, 13094 Aix-en-Provence Cedex 2, Francee Laboratoire Paleobiodiversite et Evolution, EPHE-Ecole Pratique des Hautes Etudes, 21000 Dijon, Francef Institut Universitaire de France, Paris, France

a r t i c l e i n f o

Article history:

Received 6 December 2011

Received in revised form

25 September 2012

Accepted 28 September 2012

Editor: J. Lynch-Stieglitzinter-species scale within pellets from a given location. This isotopic variability is partly explained by

Available online 24 November 2012

Keywords:

rodent

fractionation equation

oxygen isotope

tooth phosphate

palaeoclimate

1X/$ - see front matter & 2012 Elsevier B.V.

x.doi.org/10.1016/j.epsl.2012.09.058

esponding author at: Laboratoire de Geologi

ite Lyon 1 et Ecole Normale Superieure de Ly

ail address: clecuyer@univ-lyon1.fr (C. Lecuye

a b s t r a c t

Oxygen isotope compositions of tooth phosphate (d18Op) were measured in 107 samples defined on the

basis of teeth obtained from 375 specimens of extant rodents. These rodents were sampled from pellets

collected in Europe from 381N (Portugal) to 651N (Finland) with most samples coming from sites

located in France and Spain. Large oxygen isotopic variability in d18Op is observed both at the intra- and

heterochrony in tooth formation related to the short time of mineralization for all rodent species as well

as the duration of mineralization that is species-dependent. Consequently, tooth phosphate of rodents records

a short seasonal interval in the oxygen isotope compositions of meteoric waters (d18Omw). In addition, inter-

species isotopic variability observed in the same pellets suggests behavioural differences implying distinct

isotopic compositions for species living in the same location. At the scale of Europe, a robust linear oxygen

isotope fractionation equation was determined for Muroidea between the midrange d18Op values and d18Omw

values: d18Op¼1.21(70.20)d18Omwþ24.76(72.70) with R2¼0.79 (n¼9; po0.0001).

& 2012 Elsevier B.V. All rights reserved.

1. Introduction

The Quaternary is characterized by a succession of alternatingglacial and interglacial stages leading to climatic changes that havedeeply modified the floral and faunal associations (e.g. Andersonet al., 2007). At high-latitude, the oxygen isotope composition of iceprovides a continuous and high-resolution temporal record of thecontinental climate (e.g. Dansgaard et al., 1993). At lower latitudes,continental sedimentary deposits only provide discontinuous tem-poral and spatial records associated with difficulties in the absolutedating of fossils or their host sediments. Most continental climateproxies are based on pollen, insect or mammal associations. Differentmethods are applied to reconstruct climate modes; they are basedeither on the analysis of faunal associations in relation to theirpreferential extant ecological niches (Delpech et al., 1983, 2000;Chaline et al., 1995; Griggo, 1996; Jeannet, 2010), or on faunal com-munities (e.g. Legendre, 1986, 1989; Montuire and Marcolini, 2002), oron diversity and species richness (Montuire et al., 1997, 2006;Hernandez Fernandez, 2001; Legendre et al., 2005; van Dam, 2006;

All rights reserved.

e de Lyon, UMR CNRS 5276,

on, 69622 Villeurbanne,

r).

Hernandez Fernandez et al., 2007; Cuenca-Bescos et al., 2011). Thesedifferent approaches generally involve the study of small mammalsbecause of their high sensitivity to environmental parameters andclimatic changes.

Longinelli’s (1973, 1984) pioneer study demonstrated that thestable oxygen isotope composition of phosphatic tissues is a valuableproxy for investigating terrestrial climatic conditions. The oxygenisotope composition of biogenic phosphate (d18Op) depends on bothbody temperature and oxygen isotope composition of body water(d18Obw), itself related to the composition of ingested water mostcommonly of meteoric origin (Longinelli and Nuti, 1973; Kolodnyet al., 1983; Luz et al., 1984). Indeed, most previous studies assumedthat the oxygen isotope composition of vertebrate’s body waterreflects the oxygen isotope composition of meteoric water (d18Omw)depending on physiological parameters such as the general metabo-lism. These interdependent relationships allow the use of taxon-dependent oxygen isotope fractionation equations relating d18Op tod18Omw. At mid- and high-latitudes, the mean annual d18Omw islinearly related to the mean annual air temperature at the global scale(Dansgaard, 1964; Rozanski et al., 1993). Many oxygen isotopestudies were performed on bones and teeth of large mammal fossils(e.g. Koch et al., 1989; Fricke et al., 1998; Tutken et al., 2007; Bernardet al., 2009; Chritz et al., 2009) as well as to small mammal fossils toestimate successfully past air temperatures (e.g. Lindars et al., 2001;

A. Royer et al. / Earth and Planetary Science Letters 361 (2013) 258–271 259

Grimes et al., 2003, 2004; Navarro et al., 2004; Passey and Cerling,2006; Tutken et al., 2006; Feranec et al., 2010; Heran et al., 2010;Gehler et al., 2011). Compared to large mammals, their smallcounterparts are characterized by a restricted habitat and mostlylack long-distance migratory behaviour, thus incorporating in theirskeletal tissues a reliable record of their local environment. Moreover,the high abundance of small mammal remains in Quaternarysedimentary deposits potentially allows a high-temporal resolutionof climate changes in the continental domain.

Grimes et al. (2008) questioned whether or not ‘‘small isbeautiful?’’ before concluding that the oxygen isotope composi-tion of small mammals constitutes a valuable proxy when it isinterpreted in association with d18Op values obtained from othervertebrate or invertebrate fossils. Grimes et al. (2008) dealtmainly with the use of small mammals for the reconstruction ofTertiary climates, and detailed a list of limitations implying arestricted use of small mammals alone to estimate Quaternaryair temperatures. They emphasized problems such as the18O-enrichment of the ingested water relative to meteoric waters,variation in body temperature and a possible diagenetic altera-tion. In addition, we underline that 4 oxygen isotope fractionationequations relating d18Omw to d18Op of rodents have been pro-posed so far (Luz et al., 1984; Luz and Kolodny, 1985; D’Angelaand Longinelli, 1990; Longinelli et al., 2003; Navarro et al., 2004).These linear equations are characterized by significant differencesin their intercepts and slopes, which can lead to drastic differ-ences in d18Omw estimates and interpretations in terms of pastclimatic reconstructions (Heran et al., 2010; Fabre et al., 2011).Distinct oxygen isotope fractionation equations may result frombiases caused by the use of restricted datasets, different species,the skeleton parts (bones, molars or incisors), and the more or lessprecise knowledge of the oxygen isotope composition of meteoricwaters ingested by the rodents.

The aim of this study is to explore the environmental meaningof the oxygen isotope composition of phosphate from rodentteeth. Therefore we provide a new set of 107 data sampled from375 extant rodents collected in Europe from 381N (Portugal) to651N (Finland). The dataset is used to propose a new oxygenisotope fractionation equation between phosphate and ambientwater for the Muroidea superfamily (Rodentia). Finally, the originof isotopic variability in tooth composition, observed at the intra-species and inter-species scale, is discussed as a potential recordof continental climate seasonality.

2. Materials and methods

2.1. Extant rodents

Oxygen isotope ratios were measured in 107 samples definedon the basis of teeth obtained from 375 specimens of extantrodents, either from a single tooth or up to 14 teeth which havebeen pooled to constitute a single isotope analysis (Tables 1 and2). For each site, a minimum number of teeth were collectedwhenever possible according to the recommendations made byseveral authors in order to infer the local water d18O values (e.g.,Clementz and Koch, 2001; Navarro et al., 2004; Fox-Dobbs et al.,2007). Teeth were mainly extracted from owl’s pellets that ensuretooth concentration before their potential incorporation intosedimentary deposits. Twenty-six rodents from 8 pellets wereanalyzed in order to test the intra- and inter-species variabilitywhereas the remaining samples represent for each locality a poolof several teeth from one or several specimens belonging tothe same species. As a whole, teeth belong to 11 extant speciesof the Muroidea superfamily, mainly arvicolids, which are thewood mouse (Apodemus sylvaticus), the southern water vole

(Arvicola sapidus), the water vole (A. terrestris), the snow vole(Chionomys nivalis), the Cabrera’s vole (Iberomys cabrerae), thecommon vole (Microtus (Microtus) arvalis), the field vole (M. (M.)agrestis), the Mediterranean pine vole (Microtus (Terricola) duode-

cimcostatus), the Lusitanian pine vole (M. (T.) lusitanicus), thecommon pine vole (M. (T.) subterraneus), and the bank vole(Myodes glareolus). It is noteworthy that all these studied extantrodent species have been commonly recovered from excavationsof Pleistocene deposits.

Rodents were extracted from preserved modern pellets (underthe roof of bell-towers, in barns, and at the base of trees) sampledfrom 79 European localities (Fig. 1), distributed over a latitudinalgradient ranging from 381N (Portugal) to 651N (Finland). Undermid-latitude climate conditions, pellets are preserved a fewmonths to a few years depending on their location and expositionto weather conditions. Samples that come from France and Spainwere collected at various elevations (from E0 m to 1500 m) anddistances (from E0 km to 500 km) from coastlines, thus belong-ing to several microclimatic modes such as the temperatemaritime, continental and mountain ones.

d18Op values of rodents that were used to establish formeroxygen isotope fractionation equations were not taken intoaccount in this study because the oxygen isotope measurementsof phosphate were based on different analytical methods andstandard values (Luz and Kolodny, 1985; D’Angela and Longinelli,1990; Longinelli et al., 2003), or various methods to infer thed18Omw values (Navarro et al., 2004).

2.2. Estimates of the oxygen isotope composition of meteoric waters

Each sample location was georeferenced with elevation, lati-tude and longitude coordinates (Tables 1 and 2). Minimum andmaximum values of the monthly oxygen isotope compositions ofmeteoric waters (d18Omw), as well as the mean annual d18Omw,were calculated (Tables 1 and 2) by using an algorithm developedby Bowen and Wilkinson (2002), and refined by Bowen andRevenaugh (2003). The dataset was derived from the Interna-tional Energy Association/World Meteorological OrganizationGlobal Network for Isotopes in Precipitation by using the OnlineIsotopes in Precipitation Calculator (Bowen et al., 2005; Bowen,2010).

2.3. Estimates of air temperatures

Mean annual air temperatures along with both minimum andmaximum monthly air temperatures were calculated for eachstudied locality by using the software New Local ClimateEstimator (LocClim v1.11) (Grieser et al., 2006). These climaticparameters were obtained through the interpolation of 10meteorological stations located in the closest vicinity to thestudied locality. Two methods were used depending on thegeographic position of the studied site relative to the locationof the meteorological stations. The Kriging method was used fora studied site surrounded by several distant (Z50 km) meteor-ological stations whereas the Nearest Neighbors method waspreferred for stations that are very close (E10 km to 20 km) tothe studied rodent site.

2.4. Regression of data

Reduced major axis (RMA) regression was used to examinethe relationship between the d18Op and the d18Omw. An intrinsicproperty of the RMA method implies that the slope of theregression y¼ f(x) is the reciprocal of the slope of x¼ f(y). Thisprocedure is appropriate for the analysis of an oxygen isotopefractionation equation because variations in both independent (x)

Table 1Oxygen isotope compositions (% V-SMOW) of tooth phosphate from extant rodents recovered from various pellets reported along with the geographic location, the number of specimens and tooth type, as well as mean annual,

minimum, and maximum d18Omw values estimated by using the OIPC software (Bowen, 2010). The symbol ‘‘*’’ indicates samples measured by using the high-temperature pyrolysis continuous flow technique according to

Lecuyer et al. (2007).

Sample Country Location Latitude Longitude Elevation Location

number

Taxon Number of

specimens

Pellet

number

Tooth type d18Op

(% V-SMOW)

Mean annual

d18Omw

Minimum

monthly d18Omw

Maximum

monthly d18Omw

(deg.) (deg.) (m) (% V-SMOW) (% V-SMOW) (% V-SMOW)

Act 53-31 France Pointis-Inard 43.1 0.8 343 13 Arvicola terrestris 1 #1 Molars 15.8 �6.6 �8.4 �4.4

Act 54-31 France Pointis-Inard 43.1 0.8 343 13 Microtus agrestis 1 #1 Molars 18.0 �6.6 �8.4 �4.4

Act 55-31 France Pointis-Inard 43.1 0.8 343 13 Microtus arvalis 1 #1 Molars 16.7 �6.6 �8.4 �4.4

Act 59-31 France Pointis-Inard 43.1 0.8 343 13 Microtus arvalis 1 #2 Molars 19.2 �6.6 �8.4 �4.4

Act 60-31 France Pointis-Inard 43.1 0.8 343 13 Arvicola terrestris 1 #2 Molars 17.1 �6.6 �8.4 �4.4

Act 1-16 France Marillac-le-Franc 45.7 0.4 112 34 Microtus arvalis 1 #3 Incisors and

molars

17.7 �6.9 �8.7 �4.6

Act 109-16 France Marillac-le-Franc 45.7 0.4 112 34 Microtus arvalis 1 #3 Incisors 16.9 �6.9 �8.7 �4.6

Act 122-16 France Marillac-le-Franc 45.7 0.4 112 34 Microtus arvalis 1 #3 Molars 17.6 �6.9 �8.7 �4.6

Act 34-67 France Beinheim 48.9 8.1 115 49 Microtus arvalis 1 #4 Molars 13.2 �8.1 �10.5 �5.8

Act 35-67 France Beinheim 48.9 8.1 115 49 Myodes glareolus 1 #4 Molars 16.3 �8.1 �10.5 �5.8

Act 36-67 France Beinheim 48.9 8.1 115 49 Microtus agrestis 1 #4 Molars 13.1 �8.1 �10.5 �5.8

Act 47-67 France Beinheim 48.9 8.1 115 49 Myodes glareolus 1 #5 Molars 16.6 �8.1 �10.5 �5.8

Act 48-67 France Beinheim 48.9 8.1 115 49 Microtus arvalis 1 #5 Molars 11.4 �8.1 �10.5 �5.8

Act 49-67 France Beinheim 48.9 8.1 115 49 Myodes glareolus 1 #5 Molars 16.0 �8.1 �10.5 �5.8

Act 50-67 France Beinheim 48.9 8.1 115 49 Microtus arvalis 1 #5 Molars 12.6 �8.1 �10.5 �5.8

Act 56-67 France Beinheim 48.9 8.1 115 49 Microtus arvalis 1 #6 Molars 13.2 �8.1 �10.5 �5.8

Act 57-67 France Beinheim 48.9 8.1 115 49 Microtus agrestis 1 #6 Molars 13.7 �8.1 �10.5 �5.8

Act 58-67 France Beinheim 48.9 8.1 115 49 Myodes glareolus 1 #6 Molars 15.6 �8.1 �10.5 �5.8

Act 101-Esp* Spain Palacios de la Sierra 42.0 �3.1 1050 65 Microtus arvalis 1 #7 Incisors 16.4 �7.6 �9.1 �5.2

Act 102-Esp* Spain Palacios de la Sierra 42.0 �3.1 1050 65 Microtus arvalis 1 #7 Incisors 14.8 �7.6 �9.1 �5.2

Act 103-Esp* Spain Palacios de la Sierra 42.0 �3.1 1050 65 Microtus (Terricola)

lusitanicus

1 #7 Incisors 13.1 �7.6 �9.1 �5.2

Act 104-Esp* Spain Palacios de la Sierra 42.0 �3.1 1050 65 Microtus arvalis 1 #7 Incisors 15.7 �7.6 �9.1 �5.2

Act 110-H* Hungary Sarvar 47.3 16.9 155 79 Microtus arvalis 1 #8 Incisors 12.5 �7.8 �10.8 �5.5

Act 111-H* Hungary Sarvar 47.3 16.9 155 79 Microtus arvalis 1 #8 Incisors 12.5 �7.8 �10.8 �5.5

Act 112-H* Hungary Sarvar 47.3 16.9 155 79 Microtus arvalis 1 #8 Incisors 12.3 �7.8 �10.8 �5.5

Act 113-H* Hungary Sarvar 47.3 16.9 155 79 Microtus arvalis 1 #8 Incisors 12.5 �7.8 �10.8 �5.5

A.

Ro

yer

eta

l./

Ea

rtha

nd

Pla

neta

ryScien

ceLetters

36

1(2

01

3)

25

8–

27

12

60

Table 2Oxygen isotope compositions (% V-SMOW) of tooth phosphate from extant rodents recovered from various European locations reported along with the geographic location, the number of specimens and tooth type, as well as

mean annual, minimum, and maximum d18Omw values estimated by using the OIPC software (Bowen, 2010). The symbol ‘‘*’’ indicates samples measured by using the high-temperature pyrolysis continuous flow technique

according to Lecuyer et al. (2007). The symbol ‘‘þ ’’ indicates samples that were excluded from the dataset before establishing Eqs. (2) and (3).

Sample Country Location Latitude Longitude Elevation Locationnumber

Taxon Number ofspecimens

Tooth type d18Op

(% V-SMOW)

Mean annual

d18Omw

Minimum

monthly d18Omw

Maximum

monthly d18Omw

(deg.) (deg.) (m) (% V-SMOW) (% V-SMOW) (% V-SMOW)

Act 6-Belg Belgium Sommethonne 49.6 5.4 230 1 Microtus agrestis 5 Molars 14.9 �8.2 �10.4 �5.9

Act 5-Belg Belgium Bellefontaine 49.7 5.5 360 2 Microtus arvalis 5 Molars 16.8 �8.5 �10.7 �6.1

Act 26-Belg Belgium Grez-Doiceau 50.7 4.7 50 3 Microtus agrestis 5 Molars 17.1 �8.0 �10.0 �5.8

Act 4-Fin Finland Oulu 64.7 26.4 50 4 Arvicola terrestris 1 Incisors and

Molars

9.5 �12.5 �16.3 �9.6

Act 45-Fin Finland Oulu 64.7 26.4 50 4 Microtus agrestis 4 Molars 8.7 �12.5 �16.3 �9.6

Act 29-2b France Piediquacio 42.0 9.4 325 5 Apodemus sylvaticus 5 Incisors 18.4 �6.4 �8.2 �3.9

Act 31-2b France Pancheraccia 42.2 9.4 570 6 Apodemus sylvaticus 5 Incisors 18.8 �7.0 �8.9 �4.3

Act 14-66 France Montbolo 42.5 2.7 600 7 Microtus agrestis 4 Molars 18.8 �7.0 �9.0 �4.7

Act 76-66 France Rabouillet 42.7 2.4 755 8 Microtus agrestis 3 Incisors 16.3 �7.3 �9.4 �4.9

Act 23-65 France Troubat 43.0 0.6 555 9 Microtus arvalis 1 Molars 12.6 �7.0 �8.5 �4.6

Act 87-68* France Troubat 43.0 0.6 555 9 Chionomys nivalis 3 Molars 14.4 �7.0 �8.5 �4.6

Act 15-65 France Argel �es-Gazost 43.0 �0.1 470 10 Microtus agrestis 4 Molars 15.7 �6.8 �8.5 �4.6

Act 21-09 France Camon 43.0 2.0 345 11 Microtus agrestis 5 Molars 15.7 �6.6 �8.5 �4.4

Act 74-11 France Saint-Polycarpe 43.0 2.3 290 12 Microtus agrestis 3 Incisors 14.8 �6.5 �8.4 �4.3

Act 51-31 France Pointis-Inard 43.1 0.8 345 13 Arvicola terrestris 3 Molars 14.7 �6.6 �8.4 �4.4

Act 75-11 France Fontcouverte 43.2 2.7 75 14 Microtus agrestis 3 Incisors 17.5 �6.1 �8.1 �4.0

Act 17-13 France La Treille 43.3 5.5 165 15 Microtus (Terricola)

duodecimcostatus

4 Molars 19.7 �6.4 �8.5 �4.1

Act 25-13 France Salin-de-Giraud 43.4 4.7 10 16 Microtus (Terricola)

duodecimcostatus

1 Incisors 17.8 �6.1 �8.1 �3.9

Act 37-13 France Rognes 43.7 5.3 330 17 Microtus (Terricola)

duodecimcostatus

1 Molars 14.8 �6.8 �9.0 �4.4

Act 39-13 France Mallemort 43.7 5.2 145 18 Apodemus sylvaticus 1 Incisors 17.2 �6.5 �8.6 �4.2

Act 33-84 France Caumont 43.9 4.9 55 19 Microtus (Terricola)

duodecimcostatus

5 Molars 16.2 �6.4 �8.5 �4.1

Act 30-82 France Pedrech 43.9 1.3 90 20 Microtus arvalis 6 Incisors 17.8 �6.3 �8.2 �4.2

Act 97-04* France Alpes de Hautes

Provence

44.0 6.0 1000 21 Chionomys nivalis 11 Incisors 15.5 �7.3 �9.3 �4.8

Act 24-40 France Arue 44.1 �0.4 95 22 Myodes glareolus 1 Incisors 18.2 �6.4 �8.0 �4.2

Act 69-38 France Clelles 44.8 5.6 755 23 Microtus (Terricola)

duodecimcostatus

8 Molars 15.4 �8.0 �10.5 �5.4

Act 38-43 France Coubon 45.0 3.9 645 24 Microtus agrestis 1 Molars 16.1 �7.8 �10.2 �5.3

Act 99-43* France Beauzac 45.3 4.1 540 25 Arvicola terrestris 1 Incisors 12.9 �7.6 �10.0 �5.2

Act 84-19* France Hublanges 45.3 1.8 340 26 Arvicola sapidus 1 Incisors 15.0 �7.3 �9.3 �4.9

Act 9-19 France Lamazi �ere-Basse 45.4 2.2 570 27 Microtus arvalis 4 Molars 14.4 �7.7 �9.9 �5.3

Act 63-19 France Saint-Yrieix-le-

Dejalat

45.4 1.9 515 28 Arvicola sapidus 1 Incisors 16.6 �7.6 �9.7 �5.2

Act 81-19* France Treignac 45.5 1.8 530 29 Arvicola sapidus 1 Incisors 16.4 �7.7 �9.8 �5.2

Act 115-19* France La Besse 45.5 2.1 780 30 Microtus (Terricola)

subterraneus

5 Incisors 16.1 �8.2 �10.4 �5.6

Act 65-38 France Chabons 45.4 5.4 500 31 Microtus agrestis 10 Incisors 14.0 �7.8 �10.1 �5.2

Act 73-38 France Montferrat 45.5 5.6 540 32 Microtus agrestis 6 Incisors 12.9 �7.9 �10.2 �5.3

Act 13-73 France Domessin 45.5 5.7 330 33 Microtus agrestis 3 Molars 9.9 �7.5 �9.8 �5.1

Act 3-16 France Artenac 45.8 0.4 145 35 Microtus arvalis 10 Molars 20.9 �6.9 �8.4 �4.5

Act 71-01 France Les Echets 45.9 4.9 275 36 Arvicola terrestris 4 Incisors 13.8 �7.5 �9.8 �5.1

Act 72-01 France Les Echets 45.9 4.9 275 36 Microtus arvalis 5 Incisors 15.4 �7.5 �9.8 �5.1

Act 28-17 France Lussant 46.0 �0.8 10 37 Microtus arvalis 6 Molars 21.7 �6.7 �8.1 �4.3

Act 52-01 France Poncin 46.1 5.4 280 38 Microtus arvalis 5 Molars 16.4 �7.6 �9.9 �5.1

Act 10-71 France Prisse la Crouze 46.3 4.7 230 39 Microtus arvalis 6 Molars 15.7 �7.5 �9.6 �5.1

Act 40-23 France Lavaufranche 46.3 2.3 420 40 Microtus arvalis 5 Molars 16.1 �7.7 �9.8 �5.3

A.

Ro

yer

eta

l./

Ea

rtha

nd

Pla

neta

ryScien

ceLetters

36

1(2

01

3)

25

8–

27

12

61

Table 2 (continued )

Sample Country Location Latitude Longitude Elevation Location

number

Taxon Number of

specimens

Tooth type d18Op

(% V-SMOW)

Mean annual

d18Omw

Minimum

monthly d18Omw

Maximum

monthly d18Omw

(deg.) (deg.) (m) (% V-SMOW) (% V-SMOW) (% V-SMOW)

Act 64-85 France Sainte-Radegonde-

des-Noyers

46.4 �1.1 10 41 Microtus arvalis 12 Incisors 17.2 �6.7 �8.4 �4.5

Act 41-71 France Gourdon 46.6 4.4 465 42 Microtus arvalis 7 Molars 14.7 �8.0 �10.3 �5.5

Act 78-85* France Les Lucs-sur-

Boulogne

46.8 �1.5 70 43 Microtus arvalis 4 Incisors 15.7 �6.9 �8.6 �4.7

Act 44-49 France Longue-Jumelles 47.4 �1.1 30 44 Microtus arvalis 3 Molars 17.5 �7.0 �8.7 �4.7

Act 42-37 France Villedomer 47.5 0.9 100 45 Microtus arvalis 7 Molars 14.9 �7.3 �9.2 �5.0

Act 20-89 France Saint-More 47.6 3.8 125 46 Microtus agrestis 4 Molars 15.6 �7.6 �9.7 �5.2

Act 116-56* France Goezillio 48.2 �3.1 220 47 Microtus (Terricola)

subterraneus

8 Incisors 17.0 �7.4 �8.9 �5.0

Act 12-35 France Bazouges-la-

Perouze

48.4 -1.1 90 48 Microtus agrestis 3 Molars 15.4 �7.4 �9.0 �5.0

Act 8-50 France Hauteville-La-

Guichard

49.1 �1.3 70 50 Microtus arvalis 4 Molars 15.5 �7.5 �9.1 �5.1

Act 43-27 France Marais-Vernier 49.4 0.5 10 51 Microtus agrestis 5 Molars 16.2 �7.5 �9.3 �5.3

Act 32-76 France Fultot 49.8 0.8 130 52 Microtus arvalis 6 Molars 17.0 �7.9 �9.7 �5.5

Act 11-80 France Famechon 49.8 2.0 80 53 Microtus arvalis 5 Molars 13.1 �7.8 �9.7 �5.5

Act 114-80* France Famechon 49.8 2.0 80 53 Microtus (Terricola)

subterraneus

3 Incisors 13.0 �7.8 �9.7 �5.5

Act 7-08 France Hautes-Rivi �eres 49.9 4.8 210 54 Microtus agrestis 5 Molars 16.1 �8.2 �10.3 �5.9

Act 120-Por* Portugal Amareleja 38.2 �7.2 150 55 Arvicola sapidus 1 Incisors 22.2 �5.0 �5.5 �2.8

Act 117-Port* Portugal Redondo 38.6 �7.5 280 56 Microtus (Terricola)

duodecimcostatus

3 Incisors 19.7 �5.3 �5.8 �3.0

Act 118-Port* Portugal Redondo 38.6 �7.5 280 56 Arvicola sapidus 1 Incisors 16.0 �5.3 �5.8 �3.0

Act 119-Port* Portugal Tomar 39.6 �8.4 130 57 Microtus (Terricola)

duodecimcostatus

2 Incisors 19.2 �5.1 �5.8 �3.0

Act 16-Slo Slovenia Ljubljana 46.0 14.7 385 58 Microtus arvalis 6 Molars 10.6 �7.9 �9.9 �5.1

Act 92-Esp* Spain San Martin de

Valdeiglesias

40.4 �4.4 670 59 Iberomys cabrerae 1 Incisors 17.6 �6.5 �7.6 �4.2

Act 121-Esp* Spain Caminreal 40.8 �1.3 920 60 Arvicola sapidus 1 Incisors 14.9 �7.1 �8.7 �4.9

Act 70-Esp Spain Yanguas de Eresma 41.1 �4.2 895 61 Microtus arvalis

astiusanes

5 Incisors 15.7 �7.0 �8.2 �4.6

Act 83-Esp* Spain Zaragoza 41.7 �0.9 200 62 Arvicola sapidus 1 Incisors 15.2 �5.8 �7.4 �4.0

Act 82-Esp* Spain Sobradiel 41.7 �1.0 200 63 Arvicola sapidus 1 Incisors 18.2 �5.9 �7.4 �4.1

Act 79-Esp* Spain Cernadilla 42.0 �6.4 900 64 Microtus (Terricola)

lusitanicus

8 Incisors 16.7 �7.4 �8.4 �4.8

Act 80-Esp* Spain Uncastillo 42.4 �1.1 620 66 Microtus (Terricola)

duodecimcostatus

8 Incisors 16.7 �6.9 �8.5 �4.7

Act 95-Esp*þ Spain Pisto Larr �ede Olivan 42.5 �0.7 1230 67 Chionomys nivalis 4 Incisors 17.5 �8.1 �10.0 �5.6

Act 93-Esp*þ Spain Basaran 42.5 �0.2 1380 68 Chionomys nivalis 7 Incisors 16.9 �8.4 �10.4 �5.8

Act 94-Esp*þ Spain Espot 42.6 1.1 1500 69 Chionomys nivalis 1 Incisors 19.0 �8.9 �10.9 �6.0

Act 77-Esp* Spain Berastegui 43.1 �2.0 400 70 Microtus agrestis 10 Incisors 15.6 �6.6 �8.1 �4.5

Act 22-Sue Sweden Flurkmark 64.0 20.2 60 71 Microtus agrestis 3 Molars 14.0 �12.5 �15.6 �9.4

Act 18-Sue Sweden Vindeln 64.2 19.7 150 72 Microtus agrestis 3 Molars 14.2 �12.7 �16.1 �9.6

Act 46-Sue Sweden Vindeln 64.2 19.7 150 72 Arvicola terrestris 1 Molars 12.7 �12.7 �16.1 �9.6

Act 68-Sui Switzerland Villars-Burquin 46.9 6.6 1290 73 Arvicola terrestris 1 Incisors 9.1 �9.8 �12.5 �6.8

Act 66-Sui Switzerland Les Bayards 46.9 6.5 1100 74 Arvicola terrestris 1 Incisors 11.4 �9.5 �12.1 �6.6

Act 67-Sui Switzerland Emposieu du

Voisinage

47.0 6.7 1100 75 Arvicola terrestris 1 Molars 10.8 �9.5 �12.2 �6.6

Act 2-Eng England Dorset 50.8 �2.3 120 76 Microtus agrestis 8 Incisors 17.8 �7.9 �9.5 �5.6

Act 19-Eng England Norfolk 52.6 0.9 70 77 Microtus agrestis 14 Incisors 11.4 �8.4 �9.9 �5.9

Act 27-Eng England Shropshire 52.6 �3.2 120 78 Microtus agrestis 10 Molars 16.6 �8.4 �10.0 �6.0

A.

Ro

yer

eta

l./

Ea

rtha

nd

Pla

neta

ryScien

ceLetters

36

1(2

01

3)

25

8–

27

12

62

49

353458

79

5556

57

64

5960

76

7778

12

3

73-75

6562-63

7066-68 69 56

78910 111213 14 151617-1920 21222324-2526-30 31-333637 38394041 4243 44 45 46

47 4850 51

5253 54

61

60

50

40

7271

4

Fig. 1. Geographic location of rodent samples. Location reference numbers are the

same as in Tables 1 and 2. The colour gradient illustrates the d18Omw gradient in

Western Europe. Source of the map is: www.waterisotopes.org.

A. Royer et al. / Earth and Planetary Science Letters 361 (2013) 258–271 263

and dependent (y) variables are taken into account. In theframework of this study, a rodent tooth is considered to poten-tially record any month in a year (Grimes et al., 2008; Podlesaket al., 2008), thus generating a floating independent variable inthe range of possible d18Omw seasonal variations. Statistics werecomputed by using the PAST software (Hammer et al., 2001).

2.5. Analytical techniques

The small size of rodent teeth precluded the separation ofenamel from dentine, therefore the oxygen isotope compositionsof phosphate were obtained on bulk teeth. Before being crushedin an agate mortar, the teeth were cleaned with double deionizedwater (DDW) in an ultrasonic bath. Organic matter was thenoxidized by 10% H2O2 for 24 h. Tooth samples were treatedfollowing the wet chemistry protocol described by Crowsonet al. (1991) and slightly modified by Lecuyer et al. (1993). Thisprotocol consists in the isolation of phosphate PO4

3� from apatiteas Ag3PO4 crystals using acid dissolution and anion-exchangeresin. After dissolution of 15–30 mg of powdered tooth in 2 M HFat 25 1C for 24 h, the CaF2 that precipitated was separated fromthe solution that included the phosphate by centrifugation. TheCaF2 precipitate was rinsed 3 times using DDW and the rinsedwater was added to the solution that was neutralized with a 2 MKOH solution. Cleaned Amberjet

TM

resin (2 ml) was added tothe neutralized solution in polypropylene tubes. The tubes wereplaced on a shaker table for 12 h to promote the ion exchangeprocess. Excess solution was discarded and the resin was washed5 times with DDW to remove the traces of ionic contaminants. Toelute the phosphate ions quantitatively from the resin, 25–30 mlof 0.5 M NH4NO3 was added to bring the pH of the solution to7.5–8.5, and the tubes were gently shaken for about 5 h. Silverphosphate was precipitated from the eluted solution following

the method of Firsching (1961). The solution was placed in a250 ml Erlenmeyer flask and about 1 ml of concentrated NH4OHwas added to raise the pH to 9–10. Fifteen millilitre of ammo-niacal AgNO3 solution was added to the flask. Upon heating thissolution to 70 1C in a thermostatic bath, millimeter-size yellowishcrystals of Ag3PO4 were quantitatively precipitated. The crystalsof silver phosphate were collected on a Millipore filter, washed 3times with DDW, and air dried at 50 1C.

The first part of the oxygen isotope data was obtained accord-ing to a ‘‘conventional off-line’’ method (Tables 1 and 2). The18O/16O ratios were measured after reaction of silver phosphatewith graphite to form CO2 (O’Neil et al., 1994; Lecuyer et al.,1998). Aliquots of 8 mg of silver phosphate along with 0.5 mg ofpure graphite were weighed into tin reaction capsules, trans-ferred to quartz tubes and degassed for 30 min at 80 1C invacuum. The samples were then heated at 1100 1C for 1 min topromote the redox reaction. CO2 samples were analyzed at theUniversity of Lyon with a GV IsoPrime

TM

or a GV PRISM II stableisotope ratio mass spectrometer using dual inlet systems withautomated cold fingers. Isotopic compositions are quoted in thestandard d notation relative to V-SMOW. The reproducibility ofmeasurements carried out on tooth samples was close to 0.2%(1s). Silver phosphate precipitated from standard SRM 120c(formerly NBS 120c; natural Miocene phosphorite from Florida)was repeatedly analyzed (d18O¼21.69%, S.D.¼0.10; n¼28) alongwith the silver phosphate samples derived from the rodent toothcollection.

The second part of the oxygen isotope data (Tables 1 and 2)was obtained according to a high-temperature pyrolysis contin-uous flow technique developed by Lecuyer et al. (2007). Fivealiquots of 400–500 mg of silver phosphate samples were mixedwith 500 mg of nickelized carbon in silver foil capsules. Pyrolysiswas performed at 1270 1C using a EuroVector EA3028-HT

TM

elemental analyser. A ceramic reaction tube was packed withglassy carbon in the presence of nickelized graphite to generateCO that was then transferred to a GV IsoPrime

TM

mass spectro-meter in continuous flow mode with helium (He) as the carriergas. The d18O value of SRM 120c was fixed at 21.7% for correctionof instrumental mass fractionation during CO isotopic analysis.The average standard deviation equals 0.2570.13%. Aliquots ofsilver phosphate from SRM 120c were analyzed several times aday in order to account for possible instrumental drift.

The average value of 21.7% obtained for the standard SRM120c was early proposed by Lecuyer et al. (1993) who fluorinatedsilver phosphate crystals with bromine pentafluoride and cali-brated the oxygen isotope ratios against the NBS28 quartzinternational standard. Recently, Halas et al. (2011) confirmedthe absence of any sizable isotopic fractionation effect during theconversion of silver phosphate into CO and proposed a mean d18Ovalue of 21.870.2% for SRM 120c on the basis of an inter-laboratory calibration. It is also noteworthy that Chenery et al.(2010) proposed a comparable value of 21.770.7% after a6-month period of repeated measurements of SRM 120c calibratedagainst NBS127 barium sulphate, that is in agreement with theaverage value of 21.570.5% calculated from published data indifferent laboratories worldwide (Chenery et al., 2010).

3. Results

3.1. Oxygen isotope compositions of rodent teeth

3.1.1. Rodents sampled from pellets

Oxygen isotope compositions of rodent tooth phosphate mea-sured in specimens sampled from the same pellets are summar-ized in Table 1 and Fig. 2. A pellet contains skeletal remains of 2–4

10 11 12 13 14 15 16 17 18 19 20

Microtus arvalis

Arvicola terrestrisMicrotus agrestis

Microtus (Terricola)lusitanicusMyodesgl areolus

δ18Op of rodent teeth (‰ V-SMOW)

Pointis-Inard (France) Pellet 1Pellet 2

Marillac-Le-Franc (France) Pellet 1

Palacios de la Sierra (Spain) Pellet 1

Pellet 2Beinheim (France) Pellet 1

Pellet 3

Sarvar (Hungary) Pellet 1

Fig. 2. Intra- and inter- specific variations in d18Op values of extant rodent teeth sampled from various pellets. Each point corresponds to one sampled specimen.

A. Royer et al. / Earth and Planetary Science Letters 361 (2013) 258–271264

preys, which are mainly rodents captured by one owl in the sameday over a small territory of about 10–25 km2 (Andrews, 1990).The analysis of rodents preserved in pellets provides data inrelation to inter- and intra-species oxygen isotope variability at agiven location.

Specimens of Microtus arvalis from Sarvar, Hungary, havecomparable d18Op values ranging from 12.3% to 12.5% (n¼4),which are within analytical uncertainties, whereas those fromMarillac-Le-Franc, France, have a narrow variation in d18Op valuesbetween 16.9% and 17.7% (n¼3). Large inter-species isotopicvariations are observed inside a given pellet (Fig. 2). Indeed, datafrom Beinheim, France, show that d18Op values of Myodes glar-

eolus (n¼4: min–max¼15.6%–16.6%) are 3% higher in averagethan d18Op values of other coexisting Microtus sp. The twoanalyzed species M. arvalis and M. agrestis are close to each other(M. arvalis: n¼4: min–max¼11.3%–13.2%; M. agrestis: n¼2:min–max¼13.1%–13.7%). In the case of Palacios de la Sierra,Spain, a significant difference in d18Op values is observed betweenM. (T.) lusitanicus (13.1%) and M. arvalis (n¼3: min–max¼14.8%–16.4%). An inter-species isotopic difference up to 3% isalso observed between Arvicola terrestris and other Microtus sp.specimens for the two pellets collected at Pointis-Inard, France.

3.1.2. Rodents sampled over western Europe

Oxygen isotope measurements of rodent teeth sampled invarious locations across Europe are reported in Table 2. Thelowest d18Op value of 8.7% is documented at the highest latitudes(Finland: 641N) and the highest one of 22.2% at the lowestlatitudes (Portugal: 381N). No significant correlation (R2

¼0.09;po0.005) is observed between rodent tooth d18Op values and anindex of increasing continentality that was defined as the distancefrom the closest westward coastline (except samples under theinfluence of the Mediterranean climate). The large datasetsavailable for France and Spain reveal a large isotopic variabilityin rodent d18Op values at the hundred-kilometer scale. The largestisotopic variability ranging from 9.9% to 21.7% is observed inFrance whereas Spain samples have d18Op values comprisedbetween 13.1% and 19.0%. Mountain areas (France, Spain andSwitzerland) are also characterized by a large variability in d18Op

values ranging from 9.1% to 19.0%. As a whole, oxygen isotopecompositions of rodent teeth are roughly positively correlated(R2¼0.32; n¼90; po0.0001) to the mean annual d18Omw values

(1) estimated by ‘Waterisotope.org’ (Fig. 3a). Note that d18Op

values from the same species coming from the same locality weregrouped together. The linear regression of data provides thefollowing equation:

d18Op ¼ 1:73 70:15ð Þd18Omwþ28:81 71:32ð Þ ð1Þ

Compared to Eq. (1), the correlation between the oxygenisotope compositions of rodent teeth and the mean annual airtemperatures is of similar quality (R2

¼0.27; n¼90; po0.0001)(Fig. 3b). It is noteworthy, however, that the d18Op values areroughly negatively correlated (R2

¼0.31; n¼90; po0.0001) to theamplitude of seasonal temperatures (Fig. 3c).

4. Discussion

4.1. Possible causes of the variability of d18Op values at a given

location

The p value of 0.0001 associated with the linear regressionEq. (1) points out the existence of a robust relation between theoxygen isotope compositions of rodent teeth and local meteoricwaters. Nevertheless, the large data scattering, from about 2% to6% at a given location, observed in Fig. 2 indicates that second-order parameters control the oxygen isotope compositions ofrodent teeth. Several empirical and modelling studies of oxygenisotope fractionation by vertebrates revealed in detail complexrelationships with environmental (source of drinking water, meanair temperature, seasonality and relative humidity), and ecologicaland physiological parameters such as body mass, skeletal growth,metabolic activity, diet, and water turnover (Bryant and Froelich,1995; Kohn, 1996; Kohn et al., 1996; Iacumin and Longinelli, 2002).

4.1.1. Body mass and metabolic activity

Water turnover is known to be scaled to body mass (Nagy,1987; Nagy and Peterson, 1988; Kohn et al., 1996), it is thereforeexpected to be rapid for small mammals such as rodents. Podlesaket al. (2008) deduced that ‘‘the turnover of rodent body water wasrapid and had nearly reached isotopic equilibrium two weeksafter a change in drinking water’’. Consequently, the d18O valueof rodent body water records a restricted period of the oxygenisotope compositions of waters in their environment that isstrongly season-dependent at mid-latitudes.

Metabolic processes have a strong impact on the oxygenisotope composition of body water. Indeed, the basal metabolicrate (BMR) of small mammals is strongly influenced by the bodymass, ambient temperature, and air humidity. In general, bothBMR and body temperature increase with latitude. Extant rodentsare not strictly homoeotherms, they have a fluctuating bodytemperature (35–39 1C) according to ambient temperatures(Lovegrove, 2003). Grimes et al. (2008) noted that fluctuatingbody temperatures could generate substantial errors for estimat-ing past air temperatures. Other species-dependent metabolicprocesses as evaporative water loss or metabolic water consump-tion (Getz, 1963) can also modify the oxygen isotope composition

8

10

12

14

16

18

20

22

24

-18 -16 -14 -12 -10 -8 -6 -4 -2

Mean Annual δ18Omw (‰ V-SMOW)

δ18O

p of

rode

nt te

eth

(‰ V

-SM

OW

)

δ18O

p of

rode

nt te

eth

(‰ V

-SM

OW

)

δ18O

p of

rode

nt te

eth

(‰ V

-SM

OW

)

-10 -5 0 5 10 15 20 25 30

Mean Annual Air Temperature (°C)

Apodemus sylvaticusArvicola sapidusA. terrestrisChionomys nivalisI. cabrerae

Microtus arvalisM. agrestisMicrotus Terricola duodecimcostatusM. T. subterraneusM. T. lusitanicusMyodes glareolus

δ18Op = 1.73(±0.15)δ18Omw + 28.81(±1.32)R2=0.32

δ18Op = 0.92(±0.08)Τ + 05.73(±0.89)R2=0.27

δ18Op = -0.93(±0.08)ΔΤ + 31.09(±1.91)R2=0.32

8

10

12

14

16

18

20

22

24

-15

14 16 18 20 22 24 26 28 30

Seasonal amplitude (°C)

12108

10

12

14

16

18

20

22

24

Fig. 3. d18Op values (% V-SMOW) of rodent teeth reported against: (a) mean annual oxygen isotope compositions of local meteoric waters (d18Omw). The mean annual

d18Omw values of each location were estimated by using the OIPC software (Bowen, 2010). Bars illustrate minimum and maximum intra-annual variations in d18Omw

values estimated for each location. The plain line illustrates Eq. (1). (b) Temperature (1C). The mean annual temperature of each location was estimated by using the

software New Local Climate Estimator. Bars illustrate minimum and maximum intra-annual variations in air temperature values estimated for each location. (c) Seasonal

temperature amplitude (1C). Temperature values of each location were estimated by using the software New Local Climate Estimator.

A. Royer et al. / Earth and Planetary Science Letters 361 (2013) 258–271 265

of body water that results from a mass balance between outputand input oxygen fluxes.

4.1.2. Ecological parameters

Both intra- and inter-species oxygen isotope differences areobserved in pellets (Fig. 2), suggesting the influence of ecologicalpreferences upon the isotopic record in rodent teeth. Laboratoryexperiments led Podlesak et al. (2008) to quantify the respectivecontributions of drinking water (56%), solid food (14%), andrespired molecular O2 (30%) to the total amount of oxygenpresent in rodent body water. Differences in diet induce variousratios between oxygen derived from food and drinking water,consequently there is a potential differential 18O-enrichment ofbody water among various rodent species. Rodents may haveaccess to various sources of drinking water with distinct oxygenisotope ratios such as streams, still water, dew, plant exudates,snow or moisture adsorbed on food. Moreover, most smallmammals such as rodents are less water-dependent than largemammals and their low daily water requirement means that thespecific evaporative water loss may have a strong impact on theoxygen isotope budget of their body water (Getz, 1963, 1968). Assuggested by Lindars et al. (2001), the magnitude of body water18O-enrichment should also depend on the composition of thesolid food diet composition with possible varying proportions offruits, seeds or leaves whose isotopic compositions may differfrom those of local rainfall.

4.1.3. Timing of tooth growth

The formation of rat and mouse incisors is completed in 9–12days following birth. The teeth grow at rates between 0.6 and1 mm per day throughout the rodent life (Smith et al., 2005).Klevezal (2002) studied the incremental layers in dentine andconcluded that incisors were totally renewed in about one monthfor the adult Microtus arvalis against 45–50 days for Apodemus sp.,thus indicating a rapid turn-over in tooth renewal. Tooth growthrate also varies as a function of season and animal’s age (Klevezal,2002). By combining the parameters of a rapid tooth growth andmineralization with a rapid turn-over, it is expected that rodentincisors record a short period before animal’s death. Podlesaket al. (2008) have also shown that subsequently to a switch inthe oxygen isotope composition of drinking water, the incisorsof bushy-tailed woodrat (Neotoma cinerea) have reached a newisotopic steady-state in 2 weeks. In conclusion, Podlesak et al.(2008) demonstrated that these small teeth record indirectly theoxygen isotope composition of environmental water over a shortperiod that does not exceed a season.

Rodent molars have different growing modes depending onspecies (Hillson, 2005), for example, Apodemus sylvaticus has bra-chyodont teeth whereas Myodes glareolus has hypsodont teeth.Contrary to the incisors, the low-crowned teeth and the molars ofMyodes sp. record the period of birth. All the other Arvicolinaeanalyzed in this study have molars with a continuous growth mode(hypselodont), thus compensating their abrasion. The rate of toothabrasion also depends on the quantity and quality of enamel that

A. Royer et al. / Earth and Planetary Science Letters 361 (2013) 258–271266

varies among species. For example, Koenigswald and Golenishev(1979) calculated tooth growth rates of 0.4–0.5 mm per week forthe steppe lemming (Lagurus lagurus), 0.5–0.7 mm per week forMicrotus sp., and 0.8–0.9 mm per week for the collared lemming(Dicrostonyx torquatus). Koenigswald and Golenishev (1979) alsocalculated the growth rate of molars from Myodes glareolus at about0.05 mm per week by measuring the elongation of the molar roots. Itcan be questioned whether tooth size may have an impact or not onthe oxygen isotope record of environmental water. For the example ofspecimens sampled from the pellets of Pointis-Inard that have distinctd18Op values (Fig. 2), Arvicola terrestris have bigger molars thanMicrotus sp. and could have therefore recorded a d18Omw valueaveraged over a longer period than for Microtus sp. The role of thebiological age of specimens and the breeding period must also beevaluated. In fact, the raptor makes no age-dependent selection oftheir prey because of the small size of arvicolids and murids (Pearson,1985). Birth periods of captured preys may differ by several weeksand the breeding period may vary as a function of the studied species.Consequently, diachronic teeth are concentrated in pellets, whichcould explain the isotopic difference observed between Microtus sp.and Myodes glareolus sampled from the Beinheim pellets (Fig. 2).

Passey and Cerling (2002) have shown that the process ofenamel mineralization may be responsible for the dampening ofthe seasonal amplitude during the oxygen isotope record inapatite phosphate. Rates of enamel maturation and lengthwisegrowth are the 2 main factors controlling this dampening effect.The short time of enamel mineralization of the small mammalsand the small size of their teeth imply a minimal dampeningeffect, thus providing a more reliable record of the seasonalamplitude of air temperatures (Kohn 2004).

4.1.4. Role of the predator on the sampling bias

Owls, which are the main predators of the studied rodents, may beresponsible for two main sampling biases. The first one concerns theperiod of prey capture because the predators adapt their diet to theresource availability that fluctuates throughout the year. In the case ofnon-nomadic predators, they move from place to place until they findenough preys to survive. The second one is related to hunt territoriespotentially inhabited by different rodent communities, and which arepeculiar to distinct predators. For example, rodents associated withmountain ecosystems and captured at different altitudes wouldgenerate pellets containing teeth with distinct d18Op values as theresult of various drinking sources and diets.

4.2. Average d18Op values and the calculation of isotopic

fractionation equations

Ecological parameters are well constrained for some rodentspecies, for example, Lindars et al. (2001) studied fat dormouse (Glis

glis) specimens that are known for their limited period of reproduc-tion, few litters and non-continuous growth teeth. In this case, thed18O values of drinking waters can be directly related to the recordedseason. The voles, for example, may have different seasons ofreproduction, which means that their d18Op cannot be related to aknown season of reproduction such as spring, which is the mostcommon period of reproduction of mammals. We question whetherthe mean d18Op values can be used or not to infer the oxygen isotopecomposition of meteoric waters. There is also the need to investigateunder which conditions this mean d18Op value would be able toreflect the yearly mean d18Omw value.

Rodents, such as the voles, constitute the main diet componentof some predators like owls. Generally, a strong relationshipexists between the predator and the prey density (e.g. Carboneand Gittleman, 2002). Populations of small mammals fluctuateduring the year with a rising of the population during the spring

and a peak of density until the summer. Even if the predatorscapture preys all along the year, the rising number of rodentsincreases the probability of predators to capture a high number ofspecimens during the summer, which moreover corresponds tothe period when owls nest and feed their brood. As a consequenceof the close prey–predator relationship, the mean d18Op value isexpected to reflect the preferential period of prey capture thatshould correspond to the peak of density in rodent populations.

Seasonal amplitude has crucial effects on the amplitude ofseasonal variation in d18Omw as well on the demography, repro-duction and breeding periods of rodents (Bronson, 1985). Dura-tion and magnitude of peaks of population density vary withlatitude and climate modes. Indeed, high latitudes tend to delaythe period of reproduction and to produce a high peak of densityduring the short warm season. Middle latitudes offer the possi-bility to rodents of extending their period of reproduction thattakes place early in the year and goes on several months.Consequently, the mean d18Op values would reflect more or lesslonger periods of the year depending on the climate mode thatconditions the demographic rodent peak and the associatedpreferential hunting period of their predators.

From a dataset composed of several d18Op values, it is possibleto calculate 2 different arithmetical averages that are the meanand the midrange values obtained following this calculation:(minimum valueþmaximum value)/2. The midrange value canbe used assuming the working hypothesis that owls capturerodents throughout the year. Consequently, for samples fromsites with a given temperature interval, both minimum andmaximum d18Op values can be related to both minimum andmaximum monthly d18Omw values, respectively. Even though therange of seasonal d18Omw varies locally, it generally increaseswith decreasing temperatures. Consequently, mean d18Op valuesand their variability can be explored from pools of data comingfrom locations characterized by close mean annual air tempera-tures. d18Op values measured from extant rodents (Tables 1 and 2)were grouped by averaging data over mean annual air tempera-ture intervals of 1 1C, thus providing mean and midrange d18Op

values (Table 3).For a given temperature, it is observed that the mean d18Op

values are generally slightly higher than the midrange d18Op

values (Table 3). This result suggests that the mean d18Op valuescan bias the expected corresponding mean d18Omw value espe-cially in areas of high seasonal amplitude where peaks of rodentpopulations are shifted to the warmest period of the year.

Two oxygen isotope fractionation equations for the studiedrodents were calculated by using either mean or midrange d18Op

values as well as mean d18Omw values. A RMA regression analysisof d18Omw and d18Op, paired values generated the following linearEqs. (2) and (3) (Fig. 4):

Mean d18Op ¼ 24:35 72:02ð Þþ1:14 70:18ð Þd18Omw

with R2¼ 0:82 n¼ 9; po0:0001ð Þ ð2Þ

Midrange d18Op ¼ 24:76 72:70ð Þþ1:21 70:20ð Þd18Omw

with R2¼ 0:79 n¼ 9; po0:0001ð Þ ð3Þ

In both cases, the data obtained from rodents living in thePyrenean Mountains were considered as outliers and excludedfrom the 2 regressions lines. In the Pyrenean Mountains, thepoorly known altitude effect on d18Omw values could alter therelationship between the d18Op of rodents and the d18O of waterthey ingest in their living environment.

The previous calibration lines between d18Op and d18Omw

obtained for rodents were characterized by highly different slopes.Luz and Kolodny (1985) obtained a slope of 0.49 by measuring d18Op

of rats that grew in laboratory-controlled conditions. D’Angela and

Table 3

Mean and midrange d18Op values (% V-SMOW) of extant rodent teeth calculated by averaging values of samples recovered from locations belonging to the same mean air

temperature bracket (1 1C). Corresponding mean d18Omw values were obtained by using the OIPC software (Bowen, 2010). The minimum and maximum d18Op and d18Omw

values correspond to extreme values within each group. The symbol ‘‘*’’ indicates specimens coming from the Pyrenees Mountains.

Mean annual

temperature

Number of

samples

Minimum

d18Op

Maximum

d18Op

Midrange

d18Op

Mean

d18Op

Minimum

d18Omw

Maximum

d18Omw

Mean

d18Omw

(% V-SMOW) (% V-SMOW) (% V-SMOW) (% V-SMOW) (% V-SMOW) (% V-SMOW) (% V-SMOW)

2–3 1C 5 8.7 14.2 11.4 11.8 �16.3 �9.4 �12.6

7–8 1C* 3 16.9 19.0 17.9 17.8 �10.9 �5.6 �8.5

8–9 1C 5 9.1 16.1 12.6 12.6 �12.5 �4.8 �8.9

9–10 1C 11 10.6 17.8 14.2 14.5 �10.8 �4.9 �7.9

10–11 1C 23 9.9 18.8 14.3 15.1 �10.7 �4.6 �7.6

11–12 1C 14 13.1 17.5 15.3 15.5 �9.9 �4.0 �7.3

12–13 1C 16 15.0 21.7 18.3 17.5 �9.3 �4.2 �6.7

13–14 1C 4 14.8 18.4 16.6 16.7 �8.5 �3.9 �6.5

14–15 1C 4 15.2 19.7 17.5 17.7 �8.5 �3.9 �6.1

15–16 1C 4 16.0 22.2 19.1 19.3 �5.8 �2.8 �5.2

Mean Annual δ18Omw (‰V-SMOW)

Mean Annual δ18Omw (‰ V-SMOW)

δ18O

pof r

oden

t tee

th ( V

‰-S

MO

W)

δ18O

pof r

oden

t tee

th (‰

V-S

MO

W)

8

10

12

14

16

18

20

22

-17 -15 -13 -11 -9 -7 -5 -3

B A

Pyrenees Mountains

(A) Midrange: δ18Op = 24.76(± 2.70) + 1.21(±0.20)δ18Omw; R²=0.79

(B) Mean: δ18Op = 24.35(± 2.02) + 1.14(±0.18)δ18Omw; R²=0.82

-15 -13 -11 -9 -7 -5 -35

10

15

20

abc

def

a: Longinelli et al. (2003)b and c:This studyd: d’Angela and Longinelli (1990)e: Luz and Kolodny (1985)f: Navarro et al. (2004)

Fig. 4. Mean (black circles) and midrange (white circles) oxygen isotope compositions (% V-SMOW) of extant rodent teeth plotted against the mean annual oxygen

isotope compositions of local meteoric waters estimated by using the OIPC software (Bowen, 2010). Mean and midrange d18Op values are reported in Table 3. Plain lines

illustrate the oxygen isotope fractionation equations established using midrange d18Op values (A) and mean d18Op values (B). Mean (black square) and midrange (white

square) d18Op values of rodents living in the Pyrenees Mountains were considered as outliers and excluded from the two regression lines. Vertical bars illustrate the

minimum and maximum d18Op values of rodent teeth measured in each group. Horizontal bars illustrate the minimum and maximum intra-annual variations in d18Omw

values estimated by using the OIPC software. The upper left inset shows linear regressions obtained in this study compared to published oxygen isotope fractionation

equations.

A. Royer et al. / Earth and Planetary Science Letters 361 (2013) 258–271 267

Longinelli (1990) determined a slope of 0.79 based on the analysis ofwild mice, further revised to 1.14 through the addition of new dataobtained from voles (Longinelli et al., 2003). Based on a new corpus ofwild voles, Navarro et al. (2004) proposed an isotopic fractionationequation with a slope of 0.57. Several studies devoted to the d18Op ofmammals discussed the expected dependence of d18Op on d18Omw

(Longinelli, 1984; Luz et al., 1984; Bryant and Froelich, 1995; Kohn,1996; Kohn and Cerling, 2002). Slopes of linear regressions aregenerally lower than 1.0, most often close to 0.9 according to Kohnand Cerling (2002). For goats, the slope is 0.88 (Huertas et al., 1995),0.65 for deers (Luz et al., 1990), 0.73 for horses (Sanchez Chillon et al.,1994) whereas a general linear regression of mammal data compiledfrom the literature provided a slope of 0.90 (Amiot et al., 2004).Few studies provided slopes larger than 1.0 for larger mammalssuch as foxes (slope¼1.34; Iacumin and Longinelli, 2002), sheep

(slope¼1.48), cattle (slope¼1.01) or deer (slope¼1.13; D’Angela andLonginelli, 1990). For small mammals, Longinelli et al. (2003) haveobtained a slope of 1.07. Slopes larger than 1 have been put intoquestion by Bryant and Froelich (1995) who underlined the statisticaluncertainty associated with the determination of isotopic fractiona-tion equations. The original interpretation of the slope value wasprovided by Luz et al. (1984) who considered that it reflects the ratiobetween the amount of ingested water and the rate of metabolicactivity. In conclusion, slopes obtained for Eqs. (2) and (3), bothhigher than 1.0, may be considered as surprising results. However, itmust be noticed that time-dependent box modelling (Langlois et al.,2003) has shown that slopes of the fractionation equations aresensitive to body mass. In the case study of these small mammals,their small body mass, body water fluxes and high metabolic ratesmay be key factors responsible for the observed slopes.

A. Royer et al. / Earth and Planetary Science Letters 361 (2013) 258–271268

4.3. Normalized d18Omw values and seasonal variations

Oxygen isotope compositions of phosphate from incisors andmolars from various extant species of rodents were used in order toobtain a general isotopic fractionation equation that could beapplied to fossil species. Nevertheless, tooth type and taxonomyare parameters that may have important consequences uponthe climatic interpretation of the oxygen isotope record in toothphosphate. The respective roles of these 2 parameters were inves-tigated by applying the ranging method to the complete dataset. Theranging method is a mathematical procedure for adjusting para-meters to a common scale in order to make comparisons possible.

Oxygen isotope compositions of rodent tooth phosphate wereconverted into d18Omw values (yi) according to the general oxygenisotope fractionation Eq. (3) established in Section 4.2. Themethod of ‘ranging’ (Legendre and Legendre 1998) was appliedto normalize yi values into y0 i values by reducing to the hypothe-tical interval [0, 1] both minimum and maximum monthly meand18Omw values estimated by the OIPC:

y’i ¼ yi�ymin

� �= ymax�ymin

� �ð4Þ

with for each location ymin¼minimum monthly mean d18Omw,and ymax¼maximum monthly mean d18Omw.

In this study, magnitudes between minimal and maximald18Omw values of the various localities were reduced to the interval[0, 1]. If a d18Op value recorded the mean annual d18Omw of thelocality, the d18Omw values (y0i) would be close to 0.5. If a d18Op

value recorded an extreme seasonal value, the d18Omw values (y0 i)would be close to 0 or 1. The d18Omw values (y0 i) might be locatedout of the interval [0, 1], meaning that rodent teeth have recordedan isotopic signal out of the range of d18Omw estimated by OIPC.

Distribution according toin δ18Omw values (‰

-0.9 -0.6 -0.3 0 0.3 0.6 0.9 1.2 1.5 1.80

2

4

6

8

10

12

14

16

18

Num

ber o

f spe

cim

ens

Teeth:Incisors

Teeth:Molars

Distribution according to tooth type

Num

ber o

f spe

cim

ens

Normalized δ18Omw (Y’i values)

Normalized δ18O

0

2

4

6

8

10

12

14

16

18

-0.9 -0.6 -0.3 0 0.3

Fig. 5. Distributions of normalized d18Omw values (y0 i) obtained by converting d18Op

fractionation Eq. (3), then normalized into y0 i values by using Eq. (4). Frequency histogr

(a) tooth type, (b) taxonomy, and (c) seasonal amplitude in d18Omw values from the sa

Thus this method allows comparing the influence of the examinedparameters on the dispersion of d18Op values converted to d18Omw

by the use of Eq. (3). Three parameters were compared in thisstudy: the influence of taxonomy, the teeth analyzed and theseasonal amplitude of the sampled site.

Normalized d18Omw values (y0i) were calculated by distinguish-ing incisors from molars, their distributions reported in a fre-quency histogram (Fig. 5a) are not statistically different (Mann–Whitney: p¼0.215 U¼818; n incisor¼45; n molars¼43) and didnot differ from normal distributions (Shapiro–Wilk W test: W

incisors¼0.98, p¼0.71; W molars¼0.98, p¼0.80). It can be there-fore concluded that 18O/16O ratios can be measured either inincisors or molars. Taxonomy parameter was considered in Fig. 5band the observed distributions are not statistically significantbetween Microtus, Microtus (Terricola), Arvicola and Chionomys

(Kruskal–Wallis: p¼0.42). However, it can be observed thatArvicola records lower d18Omw (mean y0 i¼0.33) than Microtus

(mean y0 i¼0.46) and Microtus (Terricola) (mean y0 i¼0.44), havingitself a lower mean value than Chionomys (mean y0i¼0.71), thusreflecting increasing shifts towards high d18Omw overestimatingmean air temperatures. The last examined parameter is theseasonal amplitude of a given location (expressed as % d18Omw)with 4 defined categories that are lower than or equal to 4%,comprised between 4% and 5%, comprised between 5% and 6%,and higher than 6% (Fig. 5c). A high mean d18Omw (y0i) of 0.83 isobserved in the case of the highest seasonal amplitude larger than6% compared to mean values close to 0.43 for lower seasonalamplitudes (Kruskal–Wallis: p¼0.14). It is concluded from theseobservations that the oxygen isotope compositions of toothphosphate from Microtus and Arvicola sampled at mid-latitudesprovide reliable estimates of the mean annual oxygen isotope

the seasonal amplitude) at each location

-0.9 -0.6 -0.3 0 0.3 0.6 0.9 1.2 1.5 1.80

2

4

6

8

10

12

14

16

18

Num

ber o

f spe

cim

ens

Species: Microtus sp.Species: Arvicola sp.

Species: M. Terricola sp.Species: Chionomys sp.

Distribution according to the genus

Seasonal amplitude: ≤ 4‰5Seasonal amplitude: ]4‰ ; ‰]Seasonal amplitude: ]5‰ ;6‰] Seasonal amplitude: > 6‰

Normalized δ18Omw (Y’i values)

mw (Y’i values)

0.6 0.9 1.2 1.5 1.8

of extant rodents into d18Omw values (yi) by using the general oxygen isotope

ams show how the oxygen isotope compositions of rodent teeth are influenced by

mpled location.

A. Royer et al. / Earth and Planetary Science Letters 361 (2013) 258–271 269

composition of meteoric waters. The small size of Microtus sp.implies that they are more able to record the seasonal variabilitythan bigger rodents such as Arvicola sp. In contrast to theseresults, data obtained either from Chionomys or rodent speciesexposed to high seasonal amplitudes (e.g. high-latitudes, moun-tains, continental interiors) constitute valuable proxies of themean temperature of the warmest season.

Such a seasonal bias most likely concerned the oxygen isotopefractionation equation determined for Arvicolinae by Navarro et al.(2004) who were the first to include data obtained from high-latitudes (Scandinavia). Navarro et al. (2004) noticed that theserodent specimens from high-latitudes were trapped between Juneand August. It is likely that the oxygen isotope compositions of theirtooth phosphate closely reflect summer d18Omw values instead ofmean annual d18Omw values. This bias could explain the significantdifferences of slope values between the isotopic fractionation equa-tion proposed by Navarro et al. (2004) and those determined byD’Angela and Longinelli (1990) and Longinelli et al. (2003) thatresemble the general isotopic fractionation Eq. (3) determined inthe framework of this study (Fig. 4).

4.4. Application to fossil rodent teeth

In sedimentary deposits, especially in cave fillings, accumula-tions of rodent teeth result from the activity of their predators. Ourstudy shows that the distribution of isotopic data must be normalwith a mode occurring during the mid-season. This patternillustrates that predators may capture rodents all along the yearwith a peak of maximal activity during the warm seasons when thepopulation of rodents may increase by a factor 10 to 100.

On the basis of the results obtained from this study of extantrodents, the following recommendations should be taken intoaccount for oxygen isotope studies of fossil rodent remains:

-

The distribution of normalized data needs to be analyzedbecause only a normal distribution would authorize the inter-pretation of d18Op mean values as mean annual d18Omw values.In this favourable scenario, the seasonal amplitude can bededuced from both minimum and maximum d18Op values. Inthe case of inter-specimen differences in d18Op values largerthan 8%, which reflect high seasonal climatic amplitude, a biastowards the record of the warmest season must be envisionedas observed for extant rodents living under high-latitudes.

-

Teeth belonging to the same species should be preferred,keeping in mind that the small rodent Microtus sp. recordsmore reliably the seasonal variability as well as minimum andmaximum seasonal d18Omw values than Arvicola sp. because ofits lower body mass and higher body water fluxes.

-

Isolation of enamel from small teeth of rodents is not alwayspossible. The amount of enamel extracted from one tooth maynot be enough to ensure a high-quality oxygen isotopemeasurement of silver phosphate. Given the fact that dentineis prone to diagenetic alteration, it must be kept in mind thatd18Op values obtained from bulk fossil teeth may induce apotential bias for palaeoclimatic reconstructions.

5. Conclusion

This study investigates the oxygen isotope composition ofphosphate measured in extant rodents coming from pellets gen-erated by owls. The following main considerations are underlined:

-

Average values of oxygen isotope compositions of Muroideateeth are linearly correlated to the mean annual oxygen isotopecompositions of meteoric waters. Despite a large isotopic

variability observed at a given location, a robust oxygen isotopefractionation equation can be determined and used to recon-struct past continental climates.

-

This isotopic variability is partly explained by a heterochronyin tooth formation related to the short time of mineralizationfor all rodent species as well as the duration of mineralizationthat is species-dependent. Consequently, tooth phosphate ofrodents is able to record a short seasonal interval in theoxygen compositions of meteoric waters.

-

Oxygen isotope compositions of tooth phosphate from Micro-

tus and Arvicola sampled at mid-latitudes provide reliableestimates of the mean annual oxygen isotope composition ofmeteoric waters. The small size of Microtus sp. implies thatthey are more able to record the seasonal variability thanbigger rodents such as Arvicola sp. In contrast to these results,data obtained either from Chionomys or rodent species exposedto high seasonal amplitudes (e.g. high-latitudes, mountains,continental interiors) constitute valuable proxies of the meantemperature of the warmest season.

-

The distribution of normalized data needs to be performedbecause only a normal distribution would authorize the inter-pretation of d18Op mean values as mean annual d18Omw values. Inthis favourable scenario, the seasonal amplitude can be deducedfrom both minimum and maximum d18Op values. A bias towardthe record of the warmest season must be envisioned in the caseof inter-specimen d18Op differences larger than 8%.

-

In the case of rodent fossils, several teeth should be analyzedwithin a sedimentary level to maximize the probability forobtaining a normal distribution of d18Op mean values and totest the variability of d18Op values. Oxygen isotope ratiosshould be measured from teeth belonging to the same speciesto minimize biases related to different ecological and physio-logical parameters.

Acknowledgements

Authors are grateful to Mr. Alarmargot, M. Allanic (LPO Ile-de-France), Antoine Andre (GEPMA Alsace), Yves Armand, StephaneAulagnier (I.N.R.A), Eric Barthelemy (Association la Chev�eche),Ladislas Biegala (Groupe Mammalogique Normand), Josselin Boireau(Groupe Mammalogique Breton), Jean-Philip Brugal (UMR CNRS6636), Gaelle Caublot (Groupe Mammalogique et Herpetologiquedu Limousin), David Cochard (Archeosph�ere), Aude Gaborit Loret(IDE-Environnement), Roland Jamault (GEOCA), Sylvie Kruzik (LPOmission rapaces), Audrey Maurin (LPO Ile-de-France), Roland Libois(Universite de Li�ege), Veronique Laroulandie (UMR CNRS 5199),Rui Lourenc-o (University of Evora), Julien Oppliger (Museum d’his-toire naturelle), Carlos Pimenta (Instituto de Gest~ao do PatrimonioArquitectonico e Arqueologico), Catherine Price (University of Oxford),Francis Raoul (Universite de Franche-Comte), Pascal Rolland (GroupeMammalogique Breton), Henri Pierre Rouault (Coglais Communaute),Jerome Tourneur (Naturalistes Angevins), and Heikki Henttonen(University of Helsinki) for providing the pellets samples. AurelienRoyer was granted by the department of Charente (CG16; France).

References

Amiot, R., Lecuyer, C., Buffetaut, E., Fluteau, F., Legendre, S., Martineau, F., 2004.Latitudinal temperature gradient during the Cretaceous Upper Campanian–Middle Maastrichtian: d18O record of continental vertebrates. Earth Planet. Sci.Lett. 226, 255–272.

Anderson, D.E., Goudie, A., Parker, A., 2007. Global Environments Through theQuaternary: Exploring Environmental Change. Oxford University Press, USA.

Andrews, P., 1990. Owls, Caves, and Fossils: Predation, Preservation, and Accu-mulation of Small Mammal Bones in Caves, With an Analysis of the Pleistocene

A. Royer et al. / Earth and Planetary Science Letters 361 (2013) 258–271270

Cave Faunas from Westbury-sub-Mendip. University of Chicago Press,Somerset, UK. (239 pp.).

Bernard, A., Daux, V., Lecuyer, C., Brugal, J.P., Genty, D., Wainer, K., Gardien, V.,Fourel, F., Jaubert, J., 2009. Pleistocene seasonal temperature variations recordedin the d18O of Bison priscus teeth. Earth Planet. Sci. Lett. 283, 133–143.

Bowen, G.J., 2010. The Online Isotopes in Precipitation Calculator, Version 2.2./http://www.waterisotopes.orgS.

Bowen, G.J., Revenaugh, J., 2003. Interpolating the isotopic composition of modernmeteoric precipitation. Water Resour. Res. 39, 1299, http://dx.doi.org/10.1029/2003WR002086.

Bowen, G.J., Wassenaar, L.I., Hobson, K.A., 2005. Global application of stablehydrogen and oxygen isotopes to wildlife forensics. Oecologia 143, 337–348.

Bowen, G.J., Wilkinson, B., 2002. Spatial distribution of d18O in meteoric precipita-tion. Geology 30, 315–318.

Bronson, F.H., 1985. Mammalian reproduction: an ecological perspective. Biol.Reprod. 32, 1–26.

Bryant, J.D., Froelich, P.N., 1995. A model of oxygen isotope fractionation in bodywater of large mammals. Geochim. Cosmochim. Acta 59, 4523–4537.

Carbone, C., Gittleman, J.L., 2002. A common rule for the scaling of carnivoredensity. Science 295, 2273–2276.

Chaline, J., Brunet-Lecomte, P., Campy, M., 1995. The last glacial/interglacial recordof rodent remains from the Gigny karst sequence in the French Jura used forpalaeoclimatic and palaeoecological reconstructions. Palaeogeogr. Palaeocli-matol. Palaeoecol. 117, 229–252.

Chenery, C., Muldner, G., Evans, J., Eckardt, H., Lewis, M., 2010. Strontiumand stable isotope evidence for diet and mobility in Roman Gloucester, UK.J. Archaeol. Sci. 37, 150–163.

Chritz, K.L., Dyke, G.J., Zazzo, A., Lister, A.M., Monaghan, N.T., Sigwart, J.D., 2009.Palaeobiology of an extinct Ice Age mammal: stable isotope and cementumanalysis of giant deer teeth. Palaeogeogr. Palaeoclimatol. Palaeoecol. 282,133–144.

Clementz, M., Koch, P.L., 2001. Monitoring habitat and foraging ecology in aquaticmammal with stable isotopes in the tooth enamel. Oecologia 129, 461–472.

Crowson, R.A., Showers, W.J., Wright, E.K., Hoering, T.C., 1991. A method forpreparation of phosphate samples for oxygen isotope analysis. Anal. Chem. 63,2397–2400.

Cuenca-Bescos, G., Melero-Rubio, M., Rofes, J., Martınez, I., Arsuaga, J.L., Blain, H.A.,Lopez-Garcıa, J.M., Carbonell, E., Bermudez de Castro, J.M., 2011. The Early–Middle Pleistocene environmental and climatic change and the humanexpansion in Western Europe: a case study with small vertebrates (GranDolina, Atapuerca, Spain). J. Hum. Evol. 60, 481–491.

D’Angela, D., Longinelli, A., 1990. Oxygen isotopes in living mammal’s bonephosphate: further results. Chem. Geol. 86, 75–82.

Dansgaard, W., 1964. Stable isotopes in precipitation. Tellus 16, 436–468.Dansgaard, W., Johnsen, S.J., Clausen, H.B., Dahl-Jensen, D., Gundestrup, N.,

Hammer, C.U., Hvidberg, C.S., Steffensen, J.P., Sveinbjornsdottir, A.E., Jouzel,J., Bond, G., 1993. Evidence for general instability of past climate from a250-kyr ice-core record. Nature 364, 218–220.

Delpech, F., Donard, E., Gilbert, A., Guadelli, J.-L., Le Gall, O., Martini-Jacquin, A.,Paquereau, M.-M., Prat, F., Tournepiche, J.-F., 1983. Contribution �a la lecturedes paleoclimats quaternaires d’apr�es les donnees de la paleontologie enmilieu continental. Quelques exemples de flore et de faune d’ongules pris dansle Pleistoc�ene superieur, Colloque de l’A. G. S. O. CNRS Editions. Bordeaux,pp. 165–177.

Delpech, F., Grayson, D.K., Rigaud, J.-P., 2000. Biostratigraphie et paleoenvironne-ments du debut du Wurm recent d’apr�es les grands mammif�eres de l’abri duFlageolet I (Dordogne, France). Paleo 12, 97–126.

Fabre, M., Lecuyer, C., Brugal, J.-P., Amiot, R., Fourel, F., Martineau, F., 2011. LatePleistocene climatic change in the French Jura (Gigny) recorded in the d18O ofphosphate from ungulate tooth enamel. Quat. Res. 75, 605–613.

Feranec, R.S., Hadly, E.A., Paytan, A., 2010. Isotopes reveal limited effects of middlePleistocene climate change on the ecology of mid-sized mammals. Quat. Int.217, 43–52.

Firsching, F.H., 1961. Precipitation of silver phosphate from homogeneous solu-tion. Anal. Chem. 33, 873–874.

Fox-Dobbs, K., Bump, J.K., Peterson, R.O., Fox, D.L., Koch, P.L., 2007. Carnivore-specific stable isotope variables and variation in the foraging ecology ofmodern and ancient wolf populations: case studies from Isle Royale, Minne-sota, and La Brea. Can. J. Zool. 85, 458–471.

Fricke, H.C., Clyde, W.C., O’Neil, J.R., 1998. Intra-tooth variations in d18O (PO4) ofmammalian tooth enamel as a record of seasonal variations in continentalclimate variables. Geochim. Cosmochim. Acta 62, 1839–1850.

Gehler, A., Tutken, T., Pack, A., 2011. Triple oxygen isotope analysis of bioapatite astracer for diagenetic alteration of bones and teeth. Palaeogeogr. Palaeoclima-tol. Palaeoecol. 310, 84–91.

Getz, L.L., 1963. A comparison of the water balance of the prairie and meadowvoles. Ecology 44, 202–207.

Getz, L.L., 1968. Influence of water balance and microclimate on the local distribution ofthe redback vole and white-footed mouse. Ecology 49, 276–286.

Grieser, J., Gommes, R., Bernardi, M., 2006. New LocClim—the local climate estimatorof FAO. Geophys. Res. Abstr. 8:SRef-ID:1607–7962/gra/EGU06-A-0830.

Griggo, C., 1996. Etablissement de courbes climatiques quantifiees �a partir descommunautes animales pleistoc�enes suivi d’une application aux gisements del’abri Suard (Charente) et de la grotte du Bois-Ragot (Vienne). Paleo 8, 81–97.

Grimes, S.T., Collinson, M.E., Hooker, J.J., Mattey, D.P., 2008. Is small beautiful? Areview of the advantages and limitations of using small mammal teeth and the

direct laser fluorination analysis technique in the isotope reconstruction ofpast continental climate change. Palaeogeogr. Palaeoclimatol. Palaeoecol. 266,39–50.

Grimes, S.T., Mattey, D.P., Collinson, M.E., Hooker, J.J., 2004. Using mammal toothphosphate with freshwater carbonate and phosphate palaeoproxies to obtainmean paleotemperatures. Quat. Sci. Rev. 23, 967–976.

Grimes, S.T., Mattey, D.P., Hooker, J.J., Collinson, M.E., 2003. Paleogene paleoclimatereconstruction using oxygen isotopes from land and freshwater organisms: theuse of multiple paleoproxies. Geochim. Cosmochim. Acta 67, 4033–4047.

Halas, S., Skrzypek, G., Meier-Augenstein, W., Pelc, A., Kemp, H.F., 2011. Inter-laboratory calibration of new silver orthophosphate comparison materials forthe stable oxygen isotope analysis of phosphates. Rapid Commun. MassSpectrom. 25, 579–584.

Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST: paleontological statisticssoftware package for education and data analysis. Palaeontol. Electron. 4 (1), 9.

Heran, M.-A., Lecuyer, C., Legendre, S., 2010. Cenozoic long-term terrestrialclimatic evolution in Germany tracked by d18O of rodent tooth phosphate.Palaeogeogr. Palaeoclimatol. Palaeoecol. 285, 331–342.

Hernandez Fernandez, M., 2001. Bioclimatic discriminant capacity of terrestrialmammal faunas. Global Ecol. Biogeogr. 10, 189–204.

Hernandez Fernandez, M., Alvarez Sierra, M.A., Pelaez-Campomanes, P., 2007.Bioclimatic analysis of rodent palaeofaunas reveals severe climatic changes inSouthwestern Europe during the Plio-Pleistocene. Palaeogeogr. Palaeoclimatol.Palaeoecol. 251, 500–526.

Hillson, S., 2005. Teeth, 2nd ed. Cambridge University Press, Cambridge.Huertas, A.D., Iacumin, P., Stenni, B., Chillon, B.S., Longinelli, A., 1995. Oxygen

isotope variations of phosphate in mammalian bone and tooth enamel.Geochim. Cosmochim. Acta 59, 4299–4305.

Iacumin, P., Longinelli, A., 2002. Relationship between d18O values for skeletalapatite from reindeer and foxes and yearly mean d18O values of environmentalwater. Earth Planet. Sci. Lett. 201, 213–219.

Jeannet, M., 2010. L’ecologie quantifiee. Essai de description de l’environnementcontinental �a l’aide des microvertebres. Prehistoires Mediterraneennes.1-2010 : Varia, 11-34, http://pm.revues.org/index492.html.

Klevezal, G.A., 2002. Reconstruction of individual life histories of rodents fromtheir teeth and bone. Acta Theriol. 47, 127–138.

Koch, P.L., Fisher, D.C., Dettman, D.L., 1989. Oxygen isotopic variation in the tusksof extinct proboscideans; a measure of season of death and seasonality.Geology 17, 515–519.

Koenigswald, W. von, Golenishev, F.N., 1979. A method for determining growthrates in continuously growing molars. J. Mammal. 60, 397–400.

Kohn, M.J., 1996. Predicting animal d18O: accounting for diet and physiologicaladaptation. Geochim. Cosmochim. Acta 60, 4811–4829.

Kohn, M.J., 2004. Comment: tooth enamel mineralization in ungulates: implica-tions for recovering a primary isotopic time-Series, by B. H. Passey and T.E.Cerling (2002). Geochim. Cosmochim. Acta 68, 403–405.

Kohn, M.J., Schoeninger, M.J., Valley, J.W., 1996. Herbivore tooth oxygen isotopecompositions: effects of diet and physiology. Geochim. Cosmochim. Acta 60,3889–3896.

Kohn, M.J., Cerling, T.E., 2002. Stable isotope compositions of biological apatite.Rev. Mineral. Geochem. 48, 455–488.

Kolodny, Y., Luz, B., Navon, O., 1983. Oxygen isotope variations in phosphate ofbiogenic apatites, I. Fish bone apatite-rechecking the rules of the game. EarthPlanet. Sci. Lett. 64, 398–404.

Langlois, C., Simon, L., Lecuyer, C., 2003. Box-modeling of bone and toothphosphate oxygen isotope compositions as a function of environmental andphysiological parameters. Isot. Environ. Health Stud. 39, 259–272.

Lecuyer, C., Fourel, F., Martineau, F., Amiot, R., Bernard, A., Daux, V., Escarguel, G.,Morrison, J., 2007. High-precision determination of 18O/16O ratios of silverphosphate by EA-pyrolysis-IRMS continuous flow technique. J. Mass Spectrom.42, 36–41.

Lecuyer, C., Grandjean, P., Barrat, J.A., Emig, C.C., Nolvak, J., Paris, F., Robardet, M., 1998.d18O and REE contents of phosphatic brachiopods: a comparison between modernand lower Paleozoic populations. Geochim. Cosmochim. Acta 62, 2429–2436.

Lecuyer, C., Grandjean, P., O’Neil, J.R., Cappetta, H., Martineau, F., 1993. Thermalexcursions in the ocean at the Cretaceous-Tertiary boundary (northernMorocco): d18O record of phosphatic fish debris. Palaeogeogr. Palaeoclimatol.Palaeoecol. 105, 235–243.

Legendre, S., 1986. Analysis of mammalian communities from the late Eocene andOligocene of Southern France. Palaeovertebrata 16, 191–212.

Legendre, S., 1989. Les communautes de mammif�eres du Paleog�ene (Eoc�enesuperieur et Oligoc�ene) d’Europe occidentale: structures, milieux et evolution.Munchner Geowiss. Abh. A 16, 1–110.

Legendre, P., Legendre, L., 1998. Numerical Ecology, 2nd ed. Elsevier Science,Amsterdam.

Legendre, S., Montuire, S., Maridet, O., Escarguel, G., 2005. Rodents and climate: a newmodel for estimating past temperatures. Earth Planet. Sci. Lett. 235, 408–420.

Lindars, E.S., Grimes, S.T., Mattey, D.P., Collinson, M.E., Hooker, J.J., Jones, T.P., 2001.Phosphate d18O determination of modern rodent teeth by direct laserfluorination: an appraisal of methodology and potential application to palaeo-climate reconstruction. Geochim. Cosmochim. Acta 65, 2535–2548.

Longinelli, A., 1973. Preliminary oxygen-isotope measurements of phosphate frommammal teeth and bones. Colloq. Int. CNRS 219, 267–271.

Longinelli, A., 1984. Oxygen isotopes in mammal bone phosphate: a new tool forpaleohydrological and paleoclimatological research. Geochim. Cosmochim.Acta 48, 385–390.

A. Royer et al. / Earth and Planetary Science Letters 361 (2013) 258–271 271

Longinelli, A., Iacumin, P., Davanzo, S., Nikolaev, V., 2003. Modern reindeer and mice:revised phosphate-water isotope equations. Earth Planet. Sci. Lett. 214, 491–498.

Longinelli, A., Nuti, S., 1973. Oxygen isotope measurements of phosphate from fishteeth and bones. Earth Planet. Sci. Lett. 20, 337–340.

Lovegrove, B.G., 2003. The influence of climate on the basal metabolic rate of smallmammals: a slow–fast metabolic continuum. J. Comp. Phys. B: Biochem. Syst.Environ. Physiol. 173, 87–112.

Luz, B., Kolodny, Y., Horowitz, M., 1984. Fractionation of oxygen isotopes betweenmammalian bone-phosphate and environmental drinking water. Geochim.Cosmochim. Acta 48, 1689–1693.

Luz, B., Kolodny, Y., 1985. Oxygen isotope variations in phophate of biogenicapatites, IV. Mamm teeth and bones. Earth Planet. Sci. Lett. 75, 29–36.

Luz, B., Cormie, A.B., Schwarcz, H.P., 1990. Oxygen isotope variations in phosphateof deer bones. Geochim. Cosmochim. Acta 54, 1723–1728.

Montuire, S., Marcolini, F., 2002. Palaeoenvironmental significance of the mam-malian faunas of Italy since the Pliocene. J. Quat. Sci. 17, 87–96.

Montuire, S., Maridet, O., Legendre, S., 2006. Late Miocene–Early Pliocene tem-perature estimates in Europe using rodents. Palaeogeogr. Palaeoclimatol.Palaeoecol. 238, 247–262.

Montuire, S., Michaux, J., Legendre, S., Aguilar, J.-P., 1997. Rodents and climate. 1. Amodel for estimating past temperatures using arvicolids (Mammalia: Roden-tia). Palaeogeogr. Palaeoclimatol. Palaeoecol. 128, 187–206.

Nagy, K.A., 1987. Field metabolic rate and food requirement scaling in mammalsand birds. Ecol. Monogr. 57, 111–128.

Nagy, K.A., Peterson, C.C., 1988. Scaling of water flux rate in animals. Univ. Calif.Publ. Zool. 120, 172.

Navarro, N., Lecuyer, C., Montuire, S., Langlois, C., Martineau, F., 2004. Oxygenisotope compositions of phosphate from arvicoline teeth and Quaternaryclimatic changes, Gigny, French Jura. Quat. Res. 62, 172–182.

O’Neil, J.R., Roe, L.J., Reinhard, E., Blake, R.E., 1994. A rapid and precise method ofoxygen isotope analysis of biogenic phosphate. Isr. J. Earth Sci. 43, 203–212.

Passey, B.H., Cerling, T.E., 2002. Tooth enamel mineralization in ungulates:implications for recovering a primary isotopic time-series. Geochim. Cosmo-chim. Acta 66, 3225–3234.

Passey, B.H., Cerling, T.E., 2006. In situ stable isotope analysis d13C, d18O of verysmall teeth using laser ablation GC/IRMS. Chem. Geol. 235, 238–249.

Pearson, O.P., 1985. Predation. In: Tamarin, R.H. (Ed.), Biology of new world

Microtus. 798 American Society of Mammalogists, Shippensburg, Pennsylvania,pp. 535–563.

Podlesak, D.W., Torregrossa, A.-M., Ehleringer, J.R., Dearing, M.D., Passey, B.H.,Cerling, T.E., 2008. Turnover of oxygen and hydrogen isotopes in the body

water, CO2, hair, and enamel of a small mammal. Geochim. Cosmochim. Acta72, 19–35.

Rozanski, K., Araguas-Araguas, L., Gonfiantini, R., 1993. Isotopic patterns in

modern global precipitation. In: Swart, P.K., Lohmann, K.C., McKenzie, J., Savin,S. (Eds.), Climate Change in Continental Isotopic Records. American Geophy-

sical Union, Washington, DC, pp. 1–37.Sanchez Chillon, B., Alberdi, M., Leone, G., Bonadonna, F.P., Stenni, B., Longinelli, A.,

1994. Oxygen isotopic composition of fossil equid tooth and bone phosphate:an archive of difficult interpretation. Palaeogeogr. Palaeoclimatol. Palaeoecol.107, 317–328.

Smith, C.E., Chong, D.L., Bartlett, J.D., Margolis, H.C., 2005. Mineral acquisition ratesin developing enamel on maxillary and mandibular incisors of rats and mice:

implications to extracellular acid loading as apatite crystals mature. J. BoneMiner. Res. 20, 240–249.

Tutken, T., Furrer, H., Vennemann, T.W., 2007. Stable isotope compositions ofmammoth teeth from Niederweningen, Switzerland: implications for the LatePleistocene climate, environment, and diet. Quat. Int. 164–165, 139–150.

Tutken, T., Vennemann, T.W., Janz, H., Heizmann, E.P.J., 2006. Palaeoenvironmentand palaeoclimate of the Middle Miocene lake in the Steinheim basin, SW

Germany: a reconstruction from C, O, and Sr isotopes of fossil remains.Palaeogeogr. Palaeoclimatol. Palaeoecol. 241, 457–491.

van Dam, J.A., 2006. Geographic and temporal patterns in the late Neogene(12-3 Ma) aridification of Europe: the use of small mammals as paleoprecipitationproxies. Palaeogeogr. Palaeoclimatol. Paleoecol. 238, 190–218.