Stable carbon isotopes of C3 plant resins and ambers record changes in atmospheric oxygen ...

23 Pages·2013·3.15 MB·English

by  Ralf Tappert

Available online atwww.sciencedirect.com
ScienceDirect
GeochimicaetCosmochimicaActa121(2013)240–262
www.elsevier.com/locate/gca
Stable carbon isotopes of C3 plant resins and ambers
record changes in atmospheric oxygen since the Triassic
Ralf Tapperta,b,⇑, Ryan C. McKellara,c, Alexander P. Wolfea, Michelle C. Tapperta,
Jaime Ortega-Blancoc,d, Karlis Muehlenbachsa
aDepartmentofEarthandAtmosphericSciences,UniversityofAlberta,Edmonton,Alberta,Canada
bInstituteofMineralogyandPetrography,UniversityofInnsbruck,Innsbruck,Austria
cDivisionofEntomology(Paleoentomology),NaturalHistoryMuseum,UniversityofKansas,Lawrence,KS,USA
dDepartamentd’Estratigrafia,PaleontologiaiGeocie`nciesMarines,FacultatdeGeologia,UniversitatdeBarcelona,Barcelona,Spain
Received28September2012;acceptedinrevisedform10July2013;Availableonline24July2013
Abstract
Estimatingthepartialpressureofatmosphericoxygen(pO )inthegeologicalpasthasbeenchallengingbecauseofthelackof
2
reliableproxies.Herewedevelopatechniquetoestimatepaleo-pO usingthestablecarbonisotopecomposition(d13C)ofplant
2
resins—includingamber,copal,andresinite—fromawiderangeoflocalitiesandages(Triassictomodern).Plantresinsarepar-
ticularlysuitableasproxiesbecausetheirhighlycross-linkedterpenoidstructuresallowthepreservationofpristined13Csignatures
overgeologicaltimescales.Thedistributionofd13Cvaluesofmodernresins(n=126)indicatesthat(a)resin-producingplantfam-
iliesgenerallyhaveasimilarfractionationbehaviorduringresinbiosynthesis,and(b)thefractionationobservedinresinsissimilar
to that of bulk plant matter. Resins exhibit a natural variability in d13C of around 8& (d13C range: (cid:2)31& to (cid:2)23&, mean:
(cid:2)27&),whichiscausedbylocalenvironmentalandecologicalfactors(e.g.,wateravailability,watercomposition,lightexposure,
temperature,nutrientavailability).Tominimizetheeffectsoflocalconditionsandtodeterminelong-termchangesinthed13Cof
resins,weusedmeand13Cvalues(d13Cresin)foreachgeologicalresindeposit.Fossilresins(n=412)aregenerallyenrichedin13C
mean
comparedtotheirmoderncounterparts,withshiftsind13Cresin ofupto6&.Theseisotopicshiftsfollowdistinctivetrendsthrough
mean
time,whichareunrelatedtopost-depositionalprocessesincludingpolymerizationanddiagenesis.Themostenrichedfossilresin
samples,withad13Cresin between(cid:2)22&and(cid:2)21&,formedduringtheTriassic,themid-Cretaceous,andtheearlyEocene.Exper-
mean
imental evidence and theoretical considerations suggest that neither change in pCO nor in the d13C of atmospheric CO can
2 2
accountfortheobservedshiftsind13Cresin.Thefractionationof13Cinresin-producingplants(D13C),instead,isprimarilyinflu-
mean
encedbyatmosphericpO ,withmorefractionationoccurringathigherpO .Theenrichedd13Cresin valuessuggestthatatmospheric
2 2 mean
pO duringmostoftheMesozoicandCenozoicwasconsiderablylower(pO =10–20%)thantoday(pO =21%).Inaddition,a
2 2 2
correlation between the d13Cresin and the marine d18O record implies that pO , pCO , and global temperatures were inversely
mean 2 2
linked,whichsuggeststhatintervalsoflowpO weregenerallyaccompaniedbyhighpCO andelevatedglobaltemperatures.Inter-
2 2
valswiththelowestinferredpO ,includingthemid-CretaceousandtheearlyEocene,wereprecededbylarge-scalevolcanismdur-
2
ingtheemplacementoflargeigneousprovinces(LIPs).Thissuggeststhattheinfluxofmantle-derivedvolcanicCO triggeredan
2
initialphaseofwarming,whichledtoanincreaseinoxidativeweathering,therebyfurtherincreasinggreenhouseforcing.Thispro-
cessresultedintherapiddeclineofatmosphericpO duringthemid-CretaceousandtheearlyEocenegreenhouseperiods.Afterthe
2
cessationinLIPvolcanismandthedecreaseinoxidativeweatheringrates,atmosphericpO levelscontinuouslyincreasedovertens
2
ofmillionsofyears,whereasCO levelsandtemperaturescontinuouslydeclined.ThesefindingssuggestthatatmosphericpO had
2 2
aconsiderableimpactontheevolutionoftheclimateonEarth,andthatthed13Coffossilresinscanbeusedasanoveltoolto
assessthechangesofatmosphericcompositionssincetheemergenceofresin-producingplantsinthePaleozoic.
(cid:2)2013Elsevier Ltd. Allrights reserved.
⇑ Correspondingauthorat:InstituteofMineralogyandPetrography,UniversityofInnsbruck,Innsbruck,Austria.Tel.:[phone].
E-mailaddress:[email protected](R.Tappert).
0016-7037/$-seefrontmatter(cid:2)2013ElsevierLtd.Allrightsreserved.
http://dx.doi.org/10.1016/j.gca.2013.07.011
R.Tappertetal./GeochimicaetCosmochimicaActa121(2013)240–262 241
1. INTRODUCTION Epstein, 1960; Brugnoli and Farquhar, 2000). In
addition,someplantgroupsusedistinctivemetabolic
Our knowledge of the composition of the atmosphere pathways that fundamentally affect their isotopic
through time and of its impact on climate change and composition. For example, C4 and CAM plants are
the evolution of life is based on evidence preserved in generally more enriched in 13C compared to C3
the fossil record. For the Pleistocene and Holocene, a di- plants(O’Leary,1988;Farquharetal.,1989).There-
rect and continuous record of the partial pressures and fore,ifstableisotopesareusedinchemostratigraphic
isotopic compositions of atmospheric gases is preserved studies,itiscrucialtodeterminetheorganiccompo-
in the form of air vesicles entrapped in polar ice sheets sition of the material under investigation and its
(e.g., Petit et al., 1999). With increasing age, the record botanicalsource.
of atmospheric compositions becomes less directly accessi- (2) Environmental and ecological factors can play an
ble, and inferences about the composition of the atmo- importantroleintheisotopicfractionationofplants,
sphere in the geological past rely on indirect measures and individual isotope values of fossil or modern
or proxies. The quality of such information depends di- plantmaterialmayreflectlocalinsteadoflarge-scale
rectlyontheabilityofproxiestoreliablycapturethecom- environmental conditions.
position of ancient atmospheres. (3) Plant constituents can undergo substantial composi-
Attempts have been made to reconstruct atmospheric tionalchangesduringandafterburialanddiagenesis.
pCO and pO using mass balance calculations and flux Thesechangeshavethepotentialtoaltertheisotopic
2 2
modeling(Berner,1999;BernerandKothavala,2001;Ber- compositionofthefossilplantmaterial,totheextent
ner et al., 2007). Other approaches to estimate the paleo- of preventing a meaningful paleoenvironmental
pCO haveutilizedthedensityofstomataonfossilleafcuti- interpretation.
2
cles(Retallack,2001,2002;BeerlingandRoyer,2002)and
the carbon stable isotopic composition of pedogenic car- To determine environmentally mediated changes in the
bonates(CerlingandHay,1986;Ekartetal.,1999;Bowen d13C of fossil plant organic matter through time, we ana-
andBeerling,2004).Oneoftherichestsourcesofinforma- lyzedmodernandfossilresinsfromawiderangeoflocali-
tion on ancient atmospheric compositions comes from the ties and ages. Compared to other terrestrial plant
marine sedimentary record, in particular from the stable constituents, resins have chemical properties that make
isotopecomposition(d13C,d18O,d11B)ofbiogeniccarbon- them particularly suitable as proxies of environmental
ates (e.g., Clarke and Jenkyns, 1999; Pearson and Palmer, changes over geological time. Resins can polymerize into
2000;Royeretal.,2001;Zachosetal.,2001).Thisinforma- highly cross-linked hydrocarbons that have considerable
tionhasbeenusedto inferchangesin theisotopic compo- preservationpotentialinthegeologicalrecord(Fig.1).This
sition of atmospheric CO2 through time and to derive means that a reliable terrestrial stable isotope history can
paleo-temperature estimates. More recently, the d13C of potentiallybereconstructedwellintothePaleozoic,consid-
marine phytoplankton biomarkers have been used to esti- eringthattheearliestknownfossilresinsarefromtheCar-
mate the pCO2 of Cenozoic atmospheres (Pagani et al., boniferous (White, 1914; BrayandAnderson, 2009).
2005).Additionalconstraintsonthecompositionofancient Unlike many other fossil plant organic constituents,
atmospheres were obtained from the marine record using resinscanoftenbeassignedtoaspecificsourceplantfam-
the biochemical cycles of sulfur (Berner, 2006; Halevy ily, and only a few plant families produce significant
et al.,2012). amounts of preservable resins, all of which have C3
Forterrestrialenvironments,attemptstoestimatepaleo- metabolism. Furthermore, resins are chemically conserva-
pCO2 have frequently exploited the d13C of fossil organic tive, which means that the composition of the resins did
matter, primarily plant organic matter (Strauss and Pe- not change significantly as plants evolved. These factors
ters-Kottig,2003;Jahrenetal.,2008).Arangeofplantcon- limit not only the chemical variability of fossil resins,
stituents have been targeted for d13C analysis, including but also any potential family-specific biases, which could
wood, foliage, and cellulose extracts, as well as the coals influence the isotopic composition of fossil resins in
to which they contribute (Stuiver and Braziunas, 1987; chemostratigraphic studies.
Gro¨cke,1998,1999,2002;Lu¨ckeetal.,1999;Peters-Kottig Although large samples of gem-quality fossil resins, or
etal.,2006;PooleandVanBergen,2006;Pooleetal.,2006; amber, are found in fewer than (cid:3)50 locations worldwide,
Bechtel et al., 2008). However, three key issues need to be small pieces (<5mm) of non-precious resin, which are
addressedwhenplantorganicmatterisusedforstablecar- sometimes referred to as resinites, are found frequently,
bon isotopic analyses in geochemical and chemostrati- and are often associated with coal deposits (Stach et al.,
graphicstudies: 1982).Previous studiesofthecarbonisotopic composition
ofresinswerelimitedbecausetheywereconductedononly
(1) Plants are composed of a complex range of organic afewlocalities,andinmostcases,usingonlyafewsamples
compounds, including polysaccharides (e.g., cellu- (NissenbaumandYakir,1995;Murrayetal.,1998;Nissen-
lose),lignin,lipids,cuticularwaxes,andphotosynth- baum et al., 2005; McKellar et al., 2008; DalCorso et al.,
ates. Due to differential fractionation of carbon 2011). Our study is the first attempt to integrate the d13C
isotopes during biosynthesis, the isotopic composi- data of resins from a wide range of localities and ages,
tionofindividualplantfractionstypicallydifferfrom and to determine their potential as proxies for changes in
eachotherandfromthatofthebulkplant(Parkand atmosphere composition through geologicaltime.
242 R.Tappertetal./GeochimicaetCosmochimicaActa121(2013)240–262
A B
C
D E
Fig.1. Examplesofmodernandfossilresins.(A)FreshresinoozingoutofthetrunkofAraucariacunninghamii,AdelaideBotanicalGarden,
Adelaide,Australia.Fieldofview(FOV):(cid:3)20cm.(B)Insitulenticularresinitefragment(arrow)withinacoalseam,HorseshoeCanyon
Formation (Campanian),Edmonton,Canada(forcepsforscale).(C)SelectionofambersamplesfromtheHorseshoeCanyonFormation
(Campanian),Drumheller,Canada.Thelargestspecimenis(cid:3)5cm.(D)InclusionsoffoliagefromParataxodiumsp.(anextinctcupressaceous
conifer)inamber,ForemostFormation(Campanian),GrassyLake,Canada.FOV:(cid:3)1cm.(E)Triassicresindropletsincarbonatematrix,
HeiligkreuzkofelFormation(Carnian),Dolomites,Italy.FOV:(cid:3)10cm.
2. SAMPLES ANDMETHODS old),whicharereferredtoascopal.Thesamplesetincludes
specimensfrommanywell-knowndeposits(e.g.,Balticam-
In total, 538 d13C measurements were conducted on ber,Bitterfeldamber,Dominicanamber,MyanmarorBur-
modern (n=126) and fossil resins (n=412) from various meseamber,andLebaneseamber), butalso materialfrom
locationsworldwide,ranginginagefromrecenttoTriassic newly discovered occurrences(Table 1).
(Fig.2,Table1).Analyticalworkwasperformedonhard- For many amber deposits, the age of amber formation
enedresinsthatwerevisiblyfreeofinclusionsandvesicles. can be tightly constrained (±2Ma or better) by indepen-
Modern resins were collected primarily from the trunks of dentstratigraphiccontrolsorradiometricdating(Table1).
native trees growing under a range of climatic conditions However, for some of the deposits, the ageconstraints are
(i.e., tropical to subarctic), although specimens from culti- more provisional, especially for localities that show evi-
vars were also analyzed in some cases. We restricted the denceofre-workingorre-deposition(e.g.,BalticandBitter-
samplingofmodernresinstorepresentativesofthefamilies feld ambers). For these deposits, the age estimates have
Pinaceae(Pseudotsuga,Picea,Pinus),Cupressaceae(Meta- relatively large uncertainties ((cid:3)±4Ma).
sequoia, Thuja), Araucariaceae (Agathis), and Fabaceae Toexaminethechemicalcompositionoftheresins,rep-
(Hymenaea),asthesefamiliesrepresentthemostimportant resentative samples of modern and fossil resins from each
resin producers in modern environments and the fossil re- localitywereanalyzedusingmicro-Fouriertransforminfra-
cord(Langenheim, 2003). red (FTIR) spectroscopy. The resulting absorption spectra
The fossil resin samples include many small, non-pre- were used to (1) classify the resins into distinct composi-
cious varieties, in addition to genuine ambers. To simplify tional groups (see below), and (2) determine the botanical
the terminology, all fossil resins are referred to herein as affinity of the fossil resins. Particular emphasis was placed
amber, with the exception of sub-recent resins (>200years on identifying the botanical source of fossil resins from
R.Tappertetal./GeochimicaetCosmochimicaActa121(2013)240–262 243
ResinType Age
pinaceous Neogene
cupressaceous-araucarian Paleogene
fabaceous Late Cretaceous
dipterocarpaceous Early Cretaceous
Triassic
Fig.2. Locationsofamberdepositsandtypesofamberanalyzedinthisstudy.
previously undescribed localities. Absorption spectra were dominant terpenoid structures, but they can also be influ-
collectedbetween4000and650cm(cid:2)1(wavenumbers)using enced by the presence of additional organic components
aThermoNicoletNexus470FTIRspectrometerequipped and the degreeof polymerization.
with a Nicolet Continuum IR microscope, with a spectral Four compositional types of resins are distinguished in
resolution of 4cm(cid:2)1. A total of 200 interferograms were thisstudy.Theseresintypeswereproducedbyspecificplant
collectedandco-addedforeachspectrum.Spectrawerecol- families and are, accordingly, classified into the following
lectedfrominclusion-free,freshlybrokenresinchipsplaced fourcategories:
on an infrared-transparent NaCl disc. To avoid the effects
ofoversaturationofthespectra,samplethicknesswaskept (1) Pinaceousresinsareexclusivelyproducedbyconifers
below 5lm. Depending on the quality of the sample, the belonging to the familyPinaceae. The infrared spec-
spot size was set to values between 50(cid:4)50 and tra of pinaceous resins are characterized by distinc-
100(cid:4)100lm(squareaperture).Furtherdetailsconcerning tive absorption peaks at 1460 and 1385cm(cid:2)1
the FTIR analytical procedures are presented in Tappert (Fig. 3). The amplitude of the peak at 1460cm(cid:2)1 is
et al.(2011). generallylowerthanthepeakat1385cm(cid:2)1.Another
Thestablecarbonisotopecompositionsoftheresinswere trait is the absence or the weak development of a
determinedusingaFinniganMAT252dualinletmassspec- peakat2848cm(cid:2)1.Thispeakistypicallywell devel-
trometer at the University of Alberta. Samples of untreated oped in cupressaceous-araucarian resins (see below).
resin ((cid:3)1–5mg) were combusted together with (cid:3)1g CuO Further details about the spectral characteristics of
as an oxygen source in a sealed and evacuated quartz tube pinaceous resins can be found in Tappert et al.
at 800(cid:3)C for (cid:3)12h. The d13C analyses were normalized to (2011). Based on results of gas chromatography–
NBS-18 and NBS-19, and the results are given with respect mass spectrometry(GC–MS),pinaceousresinsconsist
to V-PDB (Coplen et al., 1983). Instrumental precision and primarily of diterpenoids that are based on pimarane
accuracy was on the order of ±0.02&. Repeated analyses andabietanestructures(Langenheim,2003).
ofindividualsampleswerewithin±0.1&. (2) Cupressaceous-araucarian resins are mainly pro-
duced by conifers of the families Cupressaceae and
3. RESULTS Araucariaceae, but also by Sciadopitys verticillata,
thesoleextantmemberofthefamilySciadopytaceae
3.1.Infrared spectroscopy (Wolfeetal.,2009).Theinfraredspectraofcupressa-
ceous-araucarian resins are characterized by distinc-
The comparison of infrared spectra from modern and tive absorption peaks at 1448, 887, and 791cm(cid:2)1
fossil resins (Fig. 3) shows that resins, irrespective of their (Fig. 3). In fossil resins, the 887cm(cid:2)1 peak is often
ageandbotanicalorigin,producespectrathatareverysim- reducedorabsentasaresultofthefossilizationpro-
ilar.Thissimilarityisduetothedominanceofditerpenoids cess. The amplitude of the absorption peak at
inthecompositionoftheresins.However,somevariability 1385cm(cid:2)1istypicallylowerthanthatoftheadjacent
in the spectra exists and differences in the position and peakat1448cm(cid:2)1.Furtherdetailsaboutthespectral
heightofspecificabsorptionfeaturescanbeusedtodistin- characteristics of cupressaceous-araucarian resins and
guishdifferentresintypes.Thesesubtlespectroscopicdiffer- their distinction from pinaceous resins can be found
ences primarily reflect variations in the structure of the in Tappert et al. (2011). Cupressaceous-araucarian
244 R.Tappertetal./GeochimicaetCosmochimicaActa121(2013)240–262
Table1
Overviewofthefossilresinsanalyzedinthisstudy,includingtheirageandbotanicalsource.
No.Sample Location Stratigraphic Age/Epoch AbsoluteCompositional ClassaBotanical Referencestoage
setting offormation Age Type(basedon source and/orbotanical
[Ma] FTIR source
spectroscopy)
1 Colombian Colombia Undetermined Pliocene– 0.0002– Fabalean Ic Hymenaeasp. Ragazzietal.(2003),
copal Holocene 2.5 Lambertetal.(1995)
2 Malaysian Borneo Merit-Pilacoal Middle 12 DipterocarpaceousII DipterocarpaceaeLangenheimandBeck
amber field Miocene (1965),
SchleeandChan(1992)
3 Mexican Chiapas LaQuintaFm.– EarlyMiocene 13–19 Fabalean Ic Hymenaeasp. LangenheimandBeck
amber BalumtunFm. (1965),
Cunninghametal.,1983,
Solo´rzanoKraemer(2007)
4 Dominican Dominican LaTocaFm.– LateOligocene–15–20 Fabalean Ic Hymenaeasp. LangenheimandBeck
amber Rep. YaniguaFm. EarlyMiocene (1965),
Iturralde-Vinentand
MacPhee(1996)
5 Baltic Baltic BlaueErdeFm. LateEocene– 33–40 Cupressaceous- Ia Sciadopityssp. Langenheim(2003),
amber coast (redeposited) EarlyOligocene araucariansuc Wolfeetal.(2009)
6 Bitterfeld Germany Bitterfeldcoal LateEocene– 33–40 Cupressaceous- Ia Sciadopityssp. Fuhrmann(2005),
amber field(redeposited)EarlyOligocene araucariansuc Wolfeetal.(2009)
7 Giraffe Canada Peatsequencein Eocene 39–38 Cupressaceous- Ib Metasequoiasp. Tappertetal.(2011)
kimberlite (NT) kimberlitecrater araucarian
amber
8 Tiger USA(WA) TigerMountain MiddleEocene 47–48 Cupressaceous- Ib Metasequoiasp. Mustoe(1985)
Mountain Fm. araucarian
amber
9 Ellesmere Canada EurekaSound MiddleEocene 50 PinaceousSUC V Pseudolarixsp. Francis(1988),thisstudy
Islandamber(NU) Fm.
10 TadkeshwarIndia Cambayshale Eocene 50–52 DipterocarpaceousII DipterocarpaceaeRustetal.(2010)
amber
11 Panda Canada KimberlitecraterEarlyEocene 53 Cupressaceous- Ib Metasequoiasp. Wolfeetal.(2012)
kimberlite (NT) infill araucarian
amber
12 Alaskan USA(AK) ChickaloonFm. Paleocene–Early53–55 Cupressaceous- Ib Metasequoiasp. Triplehornetal.(1984),
amber Eocene araucarian thisstudy
13 Evansburg Canada UpperScollard Paleocene 59–62 Cupressaceous- Ib Metasequoiasp. Thisstudy
amber (AB) Fm. araucarian
14 Genesee Canada ScollardFm. EarlyPaleocene65 Cupressaceous- Ib Metasequoiasp. Thisstudy
amber (AB) araucarian
15 Albertan Canada Horseshoe Campanian– 70–73 Cupressaceous- Ib Parataxodiumsp.McKellarandWolfe
amber (AB) CanyonFm. Maastrichtian araucarian (2010)
16 GrassyLakeCanada ForemostFm. Campanian 76–78 Cupressaceous- Ib Parataxodiumsp.McKellarandWolfe
amber (AB) araucarian (2010)
17 CedarLake Canada ForemostFm. Campanian 76–78 Cupressaceous- Ib Parataxodiumsp.McKellarandWolfe
amber (MT) (redeposited) araucarian (2010)
18 NewJersey USA(NJ) RaritanFm. Turonian 90–94 Cupressaceous- Ib Cupressaceae Grimaldietal.(1989),
amber araucarian Anderson(2006)
19 Burmese Myanmar HukawngBasin LateAlbian– 96–102 Cupressaceous- Ib Cupressaceae Grimaldietal.(2002),
amber Early araucarian CruickshankandKo
Cenomanian (2003)
20 Spanish Spain, EschuchaFm. Albian 100–110Cupressaceous- Ib CheirolepidiaceaeAlonsoetal.(2000)
amber various araucarian
21 Lebanese Lebanon Gre`sdeBaseFm.LateBarremian–123–127Cupressaceous- Ib Cupressaceae NissenbaumandHorowitz
amber EarlyAptian araucarian (1992),Azaretal.(2010)
22 Wealden UK(Isle Hastingsbeds Berriasian– 138–142Cupressaceous- Ib Cupressaceae Nicholasetal.(1993)
amber ofWight) (WealdenGroup)Valanginian araucarian
23 Italian Italy Heiligkreuzkofel Carnian 220–225Cupressaceous- Ib CheirolepidiaceaeRoghietal.(2006),
amber (Dolomites) Fm. araucarian Ragazzietal.(2003)
SUC:containssuccinicacidesters(succinate).
a ClassificationbasedonAndersonetal.(1992)andAndersonandBotto(1993).
R.Tappertetal./GeochimicaetCosmochimicaActa121(2013)240–262 245
resinsconsistprimarilyofditerpenoidsthatarebased Although pinaceous resins are produced in extremely
on labdane skeletal structures, such as communic highabundanceinnorthern-hemisphereforests,theirpres-
acidpolymers(Thomas,1970;Andersonetal.,1992). ervation potential in the fossil record is limited. The main
(3) Fabalean resins are produced by the tropical angio- constituent of pinaceous resins—diterpenes with pimarane
sperm Hymenaea (Fabaceae, Fabales). The infrared and abietane skeletal structures—do not possess the func-
spectra of these resins superficially resemble spectra tionalgroupsrequiredtoformstablepolymers.Therefore,
frompinaceousandcupressaceous-araucarianrepre- they are prone to decomposition. Nevertheless, under
sentatives(Fig.3).Similartopinaceousresins,faba- favorable conditions, even pinaceous resins can persist for
leanresinsproduceanabsorptionpeakat1460cm(cid:2)1, millions of years (Wolfe etal.,2009).
buttheamplitudeofthe1385cm(cid:2)1peakisloweror In accordance with previous studies (Langenheim and
equivalent to the 1460cm(cid:2)1 peak. A distinctive fea- Beck,1965;Broughton,1974;Tappertetal.,2011),allres-
ture of fabalean resin spectra is the lack of strong ins of a given type produce very similar FTIR absorption
absorption features between 970 and 1050cm(cid:2)1. spectra, irrespective of whether the samples are modern
Fabalean resins also produce an absorption peak at or fossil, which indicates that chemical transformations
887cm(cid:2)1. Like cupressaceous-araucarian resins, during burial and maturation are minor. This observation
fabaleanresinsareprimarilybasedonlabdanestruc- is important because it supports the premise that resins
tured diterpenoids, with zanzibaric and ozic acid can resistisotopic exchangeovergeological timescales.
polymers being typical components (Cunningham
etal., 1983). 3.2. Carbonstable isotopes
(4) Dipterocarpaceous resins are produced by members
of the angiosperm family Dipterocarpaceae, which 3.2.1.Modern resins
isafamilyofdiverseandwidespreadtropicalrainfor- Thed13Cvaluesofmodernresinsanalyzedinthisstudy
est trees. The infrared spectra of dipterocarpaceous rangefrom(cid:2)31.2&to(cid:2)23.6&.Thesevaluesareconsistent
resins are quite distinct from other resin spectra with previously published d13C analyses of modern resins
(Fig. 3). The most characteristic spectroscopic fea- from comparable plant taxa (Nissenbaum and Yakir,
tures are the distinctive absorption triplet between 1995; Murray et al., 1998; Nissenbaum et al., 2005). The
1360 and 1400cm(cid:2)1, and an additional absorption d13C values of the modern resins follow a near Gaussian
peak at 1050cm(cid:2)1. Dipterocarpaceous resins are distribution, with a mean value of (cid:2)26.9& (Fig. 4). It is
primarily based on sesquiterpenoids, such as cadin- notablethatresinsfromindividualplantfamilieshavesim-
ene, but triterpenoids can also be present (Ander- ilar d13C distributions and similar mean d13C values
son et al., 1992). The resin produced by (Fig. 5). This indicates that the isotopic fractionation that
dipterocarpaceous trees is colloquially referred to as occurs during resin biosynthesis is nearly identical for the
Dammar. plant families considered in this study, despite differences
in the terpenoid profileof individualresin types.
Some of the resins, including Baltic amber and resins Fig. 4 also shows the carbon isotopic compositions of
fromEllesmereIsland,werefoundtoproduceanadditional bulkleafmaterialfrommodernC3-plantsusingdatacom-
anddistinctivespectralfeatureat1160cm(cid:2)1,whichisasso- piledbyKo¨rneretal.(1991),Diefendorfetal.(2010),and
ciatedwithabroadshoulderorabroadpeakbetween1200 Kohn (2010). The leaf material was sampled from a wide
and1300cm(cid:2)1(Fig.3).Thesespectralfeaturesgenerallyre- range of plant taxa growing under diverse climatic condi-
latetothepresenceofesterifiedsuccinicacid(succinate)as tions,fromalpineandpolartotropical.Therefore,theiso-
an additional component in the resin (Beck et al., 1965; topic composition should approximate the average
Beck,1986).Althoughsuccinatehasbeenidentifiedinsome compositionofmodernC3plantorganicmatter.However,
pinaceousresins—namelyinresinsfromPseudolarixamabi- samplesfromextremelydryenvironmentsandsamplesthat
lisandinsomefossilresinsofPseudolarix-affinityfromthe werepotentiallyinfluencedbythecanopyeffect(seebelow)
CanadianArctic (Anderson andLePage, 1995; Poulin and were excluded. It is notable that the isotope values of the
Helwig,2012)—itisthehallmarkconstituentofBalticam- leaf material ((cid:2)32.8& to (cid:2)22.2&, mean: (cid:2)27.1&) and
ber, which itself is a cupressaceous-araucarian resin. The the distribution of isotope values are very similar to those
presence of succinate has important implications in the reportedformodernresins.Inconclusion,wefindnocon-
assessmentofthebotanicalsourceoffossilresins,andcon- sistent differences between the distributions of d13C values
sequently,succinate-bearingresinsaremarkedseparatelyin in modern leaves andmodern resins.
Table1.
Most of the fossil resins analyzed for this study are of 3.2.2.Fossil resins
cupressaceous-araucarian type. Along with fabalean res- Carbon isotope values of fossil resins were found to
ins (e.g., Dominican and Mexican ambers), cupressa- range from (cid:2)28.4& to (cid:2)18.2&, which means that some
ceous-araucarian resins make up the most productive fossil resins are more enriched in 13C compared to their
commercial amber deposits (Table 1). The predominance modern counterparts. Data for individual samples are
of these resins in the fossil record is due to their terpe- shown in Fig. 6. Similar to modern resins, the fossil resins
noid composition because labdane-based resins polymer- at each deposit produce a range of d13C values. These
ize rapidly and are highly resistant to chemical ranges are similar to those observed for modern resins
breakdown (Fig. 1B). (i.e.,(cid:3)8&)providedthatacomparablenumberofsamples
246 R.Tappertetal./GeochimicaetCosmochimicaActa121(2013)240–262
Single bonds (x-H) Double bonds Single bonds/Skeletal vibrations
O-H C-H C=C, C=O C-O, C-C
2848
1160
14601448
1050 791
1385 887
succinate
region
Pinaceous resins
Pseudolarix amabilis
-modern-
Ellesmere Island amber (Canada)
-Middle Eocene-
Cupressaceous-
araucarian resins
Wollemia nobilis
-modern-
)
y
rit
cla Tiger -MMoidudnltea iEno acbeenre (-USA)
r
o
f
est Genesee amber (Canada)
off -Early Paleocene-
(
e
c Fabalean
n
a resins
b Hymenaea courbaril
or -modern-
s
b
A
Mexican amber
-Early Miocene-
Dominican amber
-Late Oligocene-Early Miocene-
Dipterocarpaceous
resins
Shorea rubriflora
-modern-
Malaysian amber
-Middle Miocene-
3600 3400 3200 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800
Wavenumbers (cm-1)
Fig. 3. Micro-FTIR spectra of different resin types distinguished in this study (selected modern and fossil examples). Dashed lines mark
spectralfeaturesdiscussedinthetext.Yellowbackgroundshighlightthemostdistinctivespectralregions.Characteristicspectralregionsfor
molecularvibrationsrelevanttoresinsaremarkedinblue.(Forinterpretationofthereferencestocolorinthisfigurelegend,thereaderis
referredtothewebversionofthisarticle.)
were analyzed. For some of the amber deposits, only a To quantify the isotopic shift through time, and to mini-
small number of samples were analyzed due to sample mizetheeffectofsamplesize,weusedthemeand13Cvalues
availability. Consequently, the range of isotope values for ofthefossilresins(d13Cresin)asarepresentativemeasurefor
mean
these deposits tends to be narrower. Despite differences in each deposit(Table2).
thenumberofsamplesfordifferentdeposits,aconsiderable Thed13Cresin ofsubrecentColombiancopal((cid:2)27.5&)is
mean
shift in d13C between deposits is observed (Figs. 6 and 7). nearly identical to modern resins, but with increasing age,
R.Tappertetal./GeochimicaetCosmochimicaActa121(2013)240–262 247
Fractionation efficiency
high low
Growth conditions (interpreted)
optimal typical poor
A
Modern resins mean: -26.9‰
= 1.39
n = 160
y
c
n
e
u
q
e
Fr
B
Bulk C3 leaf mean = -27.1‰
= 1.67
n = 891
y
c
n
e
u
q
e
Fr
13C (‰vs PDB)
Fig.4. Distributionofd13Cvaluesin(A)modernresins(datafromthisstudywithadditionaldataofHymenaearesinsfromNissenbaum
etal.,2005),and(B)bulkleafmatterfromawiderangeofC3planttaxa(datafromKo¨rneretal.,1991;Diefendorfetal.,2010;Kohn,2010).
resinsbecomecontinuouslymoreenrichedin13Candcon- TheearlyEoceneischaracterizedbyadramaticshiftto-
sequently less negative in their d13Cresin values. This trend wardsmoredepletedresinisotopecompositions,asdemon-
mean
towards more enriched isotopic compositions in fossil res- strated by amber from the Tadkeshwar coal field of India
inscontinuesthroughtheNeogeneandintothemiddleEo- (d13Cresin =(cid:2)25.8&), and amber preserved in the Panda
mean
cene. Some of the most productive amber deposits, kimberlitepipeinCanada(d13Cresin =(cid:2)24.8&).Theearli-
mean
includingtheDominican(d13Cresin =(cid:2)24.8&)andtheBal- estEoceneamberfromtheChickaloonFormationinAlas-
mean
tic(d13Cresin =(cid:2)23.5&)amberdeposits,formedduringthis ka marks a return to slightly more 13C-enriched
mean
timeinterval.EoceneambersfromTigerMountain(Wash- compositions, witha d13Cresinof(cid:2)23.6&.
mean
ington,USA)andEllesmereIsland(Nunavut,Canada)are Paleocene amber from Evansburg (d13Cresin ¼(cid:2)22:3&)
mean
more than 5.3&enriched in 13C compared to modern res- and Genesee (d13Cresin ¼(cid:2)23:1&) in Alberta, Canada,
mean
ins, with a d13Cresin of (cid:2)21.7& and (cid:2)21.6&, respectively. are more enriched than their early Eocene counterparts.
mean
Thelatter arethe isotopically most enrichedCenozoic res- However, from the Paleocene into the Late Cretaceous
ins in the presentsampleset. (Campanian),resincompositionsbecomeincreasinglymore
248 R.Tappertetal./GeochimicaetCosmochimicaActa121(2013)240–262
35
Pinaceae
30 (mean=-26.9‰, =1.12, n=86)
Cupressaceae-Araucariaceae
(mean=-26.6‰, =1.77, n=50)
25
y Fabaceae (angiosperm)
c (mean=-27.3‰, =1.28, n=24)
n
e 20
u
q
e
Fr 15
10
5
0
-31 -30 -29 -28 -27 -26 -25 -24 -23
13C (‰vs PDB)
Fig.5. Distributionofd13Cvaluesfromdifferenttypesofmodernresins,separatedaccordingtotheirsourceplantfamilies.Datainclude
previouslypublishedanalysesofHymenaea(Fabaceae)resins(n=17)fromNissenbaumetal.(2005).
depletedin13C.ThemostdepletedCretaceousresinswitha of parasites, etc. (Park and Epstein, 1960; Guy et al.,
d13Cresin of (cid:2)23.9& were found in the Horseshoe Canyon 1980;O’Leary,1981;Ko¨rneretal.,1991;ArensandJahren,
mean
Formation (late Campanian) of southern and central Al- 2000; Edwards et al., 2000; Dawson et al., 2002; Gro¨cke,
berta.FromthelateCampanianandintotheEarlyCreta- 2002;McKellar etal.,2011)(Fig.8).Eachofthesefactors
ceous, this trend reverses and the d13Cresin shifts to more has a direct influence on the efficiency of fractionation in
mean
13C-enrichedcompositions,asshownbythecoevalGrassy plantsduringphotosynthesis,andeachcancauseanisoto-
LakeandCedarLakeamber(d13Cresin ¼(cid:2)23:6&),andthe pic shift on the order of several permil in the d13C of a
mean
New JerseyRaritanamber (d13Cresin ¼(cid:2)22:1&)(Fig. 7). plant.
mean
ResinsthatformedbetweentheCenomanianandBarre- Everyplantcommunitywillproducearangeofd13Cval-
mian are characterized by even more enriched d13Cresin uesbecausetheenvironmentalconditions(i.e.,growthcon-
mean
compositions. During this time interval, Burmese amber ditions) for each plant are unique. However, each plant
(d13Cresin ¼(cid:2)21:3&), Spanish amber (d13Cresin ¼ taxonhasitsownrangeofgrowthconditionsunderwhich
mean mean
(cid:2)21:4&), and Lebanese amber (d13Cresin ¼(cid:2)21:1&) were 13C fractionation is maximized. The observation that the
mean
formed. The oldest Cretaceous amber samples were recov- d13C of modern resins follows a Gaussian distribution
eredfromtheWealdenGroupofEngland(UK).Although (Fig. 4), suggests that the majority of the resin-producing
the d13Cresin ((cid:2)22.8&) of the Wealden amber is more de- plants grew under conditions that were intermediate be-
mean
pleted than that of the Burmese, Spanish, and Lebanese tween optimal (i.e., most efficient fractionation) and poor
amber, its d13Cresin is not well constrained because only (i.e., least efficient fractionation). Therefore, the relative
mean
twosamples were available for analysis. d13Cvaluesofplantmatterwithinalocalplantcommunity
Pre-Cretaceousamberswerelimitedtosamplesfromthe can be viewed primarily as a measure of the growth
Triassic(Carnian)Heiligkreuzkofel Formationofthe Dol- conditions.
omites,Italy.Withad13Cresin of(cid:2)20.9&,theseambersare Formostplants,includingthecopious resinproducers,
mean
the most 13C-enriched in the current sample set. In accor- it can be assumed that the metabolized CO was sourced
2
dancewithpreviouslypublishedd13Cvaluesofamberfrom fromisotopicallyundisturbedair,whichhadad13Ccompo-
the same locality (d13Cresin ¼(cid:2)20:12&; DalCorso et al., sition approximating a global atmospheric average. How-
mean
2011), our results confirm that these Triassic resins ever, local isotopic effects can alter the d13C of the CO in
2
are>6&more13C-enriched relativeto modern resins. the ambient air. Limited air circulation, for example, can
lead to a reprocessing of respired air, which is depleted in
4.DISCUSSION 13C (i.e., canopy effect). Metabolizing this depleted air
causesthed13Cofplantstoshifttowardsmorenegativeval-
4.1.Influences onthed13C ofmodern plants ues. In extreme cases, it can result in a depletion in 13C of
plant matter of >6& (Medina and Minchin, 1980). The
4.1.1.Localenvironmental factors canopy effect primarily affect plants in the understory of
Experimentalstudies,includinggrowthchamberexperi- densetropicalrainforests.Amongtheplantfamiliesconsid-
ments, have shownthat plant d13C is influenced bya wide ered in this study, however, only the Dipterocarpaceae
range of local environmental factors, such as water avail- grow in such dense rainforest environments, and only the
ability, water composition (e.g., salinity, nutrients), light Miocene amber from Borneo and the Eocene amber from
exposure, local temperature, altitude, humidity, presence Tadkeshwar, India, are of dipterocarpaceanorigin.
R.Tappertetal./GeochimicaetCosmochimicaActa121(2013)240–262 249
13C (‰vs VPDB)
-32 -30 -28 -26 -24 -22 -20 -18 -16
Pinaceae
Cupressaceae-Araucariaceae modern
Fabaceae
Columbian copal
Malaysian amber
Mexican amber
Dominican amber
Baltic amber
Bitterfeld amber
Giraffe kimberlite amber
Tiger Mountain amber
Ellesmere Island amber
Tadkeshwar amber
Panda kimberlite amber
Alaskan amber
Evansburg amber
Genesee amber
Albertan amber
Cedar Lake amber
Grassy Lake amber
New Jersey amber
Burmese amber
Spanish amber
Lebanese amber
Wealden amber
Italian amber
Fig.6. d13Cofmodernandfossilresinsanalyzedforthisstudy.Modernresindataareseparatedaccordingtotheirbotanicalsource,i.e.,resin
type.Datafromfossilresins(grey)areshownforindividualdeposits.Thedataarearrangedaccordingtotherelativeageofthefossilresins
(seeTable1).Blackrectanglesrepresentonestandarddeviation.Blacksquaresmarkmeanvaluesandverticalbarsmedianvalues.
4.1.2.Regionalandglobal environmental factors (MAP>2000mm/year), in which fractionation is in-
Theresultsofrecentstudiesontheglobaldistributionof creased.Forplantsthatgrewinenvironmentswithaninter-
d13C values in bulk leaf matter indicate that the fraction- mediate MAP, average D13C values only show a weak
ation of 13C in plants is not only influenced by local envi- correlation with MAP. Since most of the resin-producing
ronmental factors, but also by regional precipitation plant families investigated here grow typically in environ-
patterns (Diefendorf et al., 2010; Kohn, 2010; Fig. 8). It ments with intermediate MAP, the effect of MAP on their
was found that the average d13C of bulk leaf matter and D13C can be considered minor. This notion is supported
thecalculatedfractionationvalues(D13C)showsomecorre- by the observation that the d13Cresin values of resins from
mean
lationwiththemeanannualprecipitation(MAP)atagiven different modern plant families investigated in this study
location. The 13C fractionation thereby increases with are very similar. However, the Dipterocarpaceae represent
increasing MAP (Diefendorf et al., 2010; Kohn, 2010). an exception, because they commonly grow in areas with
However, the correlation is strongly influenced by plants high MAP. Therefore, they are predisposed towards in-
that were sampled from very dry environments creased 13C fractionation. This ecological preference may
(MAP<500mm/year), in which 13C fractionation is re- explainthelowd13Cvaluesofdipterocarpaceousresinsob-
duced, and those sampled from tropical rainforests served byMurrayet al.(1998).