O'Hora Et Al., 2022. Clumped-Isotope-Derived Climate Trends Leading Up To The End-Cretaceous Mass Extinction in Northwestern Europe

Clim.
Past, 18, 1963–1982, 2022

https://doi.org/10.5194/cp-18-1963-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
Clumped-isotope-derived climate trends leading up to the

end-Cretaceous mass extinction in northwestern Europe
Heidi E. O’Hora1 , Sierra V. Petersen1 , Johan Vellekoop2,3 , Matthew M. Jones1,4 , and Serena R. Scholz1,5
1 Department of Earth and Environmental Sciences, University of Michigan, Ann Arbor, MI 48104, USA
2 Analytical,Environmental, and Geo-Chemistry, Vrije Universiteit Brussel, 1050 Brussels, Belgium
3 Department of Earth and Environmental Sciences, KU Leuven, 3000 Leuven, Belgium
4 Department of Paleobiology, MRC-121, National Museum of Natural History,
Smithsonian Institute, Washington, DC 20013, USA

5 Dept of Earth and Planetary Sciences, Yale University, New Haven, CT 06511, USA
Correspondence: Sierra V. Petersen ([email protected])
Received: 4 August 2021 – Discussion started: 9 September 2021

Revised: 31 March 2022 – Accepted: 22 July 2022 – Published: 1 September 2022
Abstract. Paleotemperature reconstructions of the end- neous province (LIP) on the Indian subcontinent resulted
Cretaceous interval document local and global climate in global warming (see Hull et al., 2020, and references
trends, some driven by greenhouse gas emissions from therein). This warming event (termed the Late Maastrichtian
Deccan Traps volcanism and associated feedbacks. Here, Warming Event or LMWE; Woelders et al., 2018) has been
we present a new clumped-isotope-based paleotemperature observed in multiple locations and has been dated to approx-
record derived from fossil bivalves from the Maastrichtian imately coincide with the onset of major Deccan volcanism
type region in southeastern Netherlands and northeastern (66.413±0.067 Ma; Sprain et al., 2019). Anomalous mercury
Belgium. Clumped isotope data document a mean temper- concentrations in sediments (e.g., Font et al., 2016; Sial et al.,
ature of 20.4 ± 3.8 ◦ C, consistent with other Maastrichtian 2016; Percival et al., 2018; Zhao et al., 2021) and fossil shells
temperature estimates, and an average seawater δ 18 O value (Meyer et al., 2019) link the warming to a volcanic source,
of 0.2 ± 0.8 ‰ VSMOW for the region during the latest Cre- while modeling of the climate impacts of hypothesized vol-
taceous (67.1–66.0 Ma). A notable temperature increase at canic CO2 and Hg emissions results in warming consistent
∼ 66.4 Ma is interpreted to be a regional manifestation of the with observations (Tobin et al., 2017; Fendley et al., 2019;
globally defined Late Maastrichtian Warming Event, link- Nava et al., 2021).
ing Deccan Traps volcanic CO2 emissions to climate change Prior to the LMWE, the global climate was in a “cool
in the Maastricht region. Fluctuating seawater δ 18 O values greenhouse” state (Scotese, 2021), with climate models
coinciding with temperature changes suggest alternating in- (Miller et al., 2005; Ladant and Donnadieu, 2016), sea-ice-
fluences of warm, salty southern-sourced waters and cooler, indicating dinoflagellate cysts (Bowman et al., 2013), and
fresher northern-sourced waters from the Arctic Ocean. This coastal Antarctic temperatures near the freezing point (Pe-
new paleotemperature record contributes to the understand- tersen et al., 2016a), indicating the possible existence of land
ing of regional and global climate response to large-scale vol- ice on the Antarctic continent. The LMWE may have caused
canism and ocean circulation changes leading up to a catas- such an ice cap to melt (Bowman et al., 2013; Petersen et al.,
trophic mass extinction. 2016a) and, subsequently, global sea level to rise. This is po-
tentially visible in the global sea level curve of Haq (2014),
but the resolution of this sea level record is too coarse to con-
1 Introduction fidently link cycles to the short-term LMWE.
After the Late Maastrichtian Warming Event, the climate
During the late Maastrichtian, greenhouse gas emissions cooled again gradually, potentially due to silicate weathering
from the emplacement of the vast Deccan Traps large ig-
Published by Copernicus Publications on behalf of the European Geosciences Union.

1964 H. E. O’Hora et al.: Clumped-isotope-derived end-Cretaceous climate trends in NW Europe
feedbacks (Petersen et al., 2016a; Tobin et al., 2017), before 2 Locality and sample collection
being thrown into upheaval by the arrival of the Chicxulub
meteorite impact, defining the Cretaceous–Paleogene (K–Pg) Shell specimens were collected from three locations in the
boundary (Molina et al., 2006). Ejecta from the impact cov- Liège–Maastricht region of northeastern Belgium and south-
ered the planet and resulted in a short-lived “impact winter” eastern Netherlands: the Maastrichtian type section at ENCI
event (Vellekoop et al., 2014, 2016). Taken together, Deccan quarry, the CBR Romontbos quarry, and the K–Pg boundary
Traps volcanism (and associated feedbacks) and the Chicxu- (KPB) section at the Geulhemmerberg subterranean galleries
lub bolide impact create a complex pattern of paleoenviron- (GSGs; Jagt and Jagt-Yazykova, 2012; Fig. 1c). The ENCI
mental disturbances through the end Cretaceous, ultimately quarry is located ∼ 3.3 km south of the city of Maastricht
leading to the rapid extinction of ∼ 70 % of species on Earth in South Limburg, Netherlands (Fig. 1c). Maastrichtian-
(Schulte et al., 2010). aged strata at this location include the Gulpen Formation
Previous studies have reconstructed Maastrichtian marine (∼ 67.1–78.9 Ma) overlain by the Maastricht Formation (∼
temperatures at sites around the globe using several dif- 66.0–66.8 Ma; Vellekoop et al., 2022.), the latter of which
ferent proxies, including δ 18 O of foraminifera, mollusks, was sampled for this study of the uppermost Maastrichtian
tooth enamel, and bulk carbonate material (e.g., Pucéat et stage. At this location, the stratigraphically highest expo-
al., 2007; Tobin et al., 2012; Hull et al., 2020), as well as sure of the Maastricht Formation in the Meerssen Mem-
TEX86 measurements (e.g., Vellekoop et al., 2016; Woelders ber predates the KPB by less than 100 kyr (Keutgen, 2018).
et al., 2017), Mg/Ca ratios (e.g., Woelders et al., 2018), The CBR Romontbos quarry is located ∼ 4 km to the south-
and clumped isotope thermometry (Dennis et al., 2013; To- west of ENCI quarry, near Eben Emael, Liège, Belgium
bin et al., 2014; Petersen et al., 2016a, b, Meyer et al., (Fig. 1c). The Maastricht Formation is also exposed at this
2018, 2019; Zhang et al., 2018; Tagliavento et al., 2019; location, and samples from here were correlated to align with
Table S1 in the Supplement). Taken together, these records the Maastricht Formation at ENCI quarry (Felder, 1975).
clearly show a late Maastrichtian warming (summarized by We also include a specimen from the nearby Geulhemmer-
Hull et al., 2020), although the magnitude of warming dif- berg subterranean galleries, located ∼ 7 km east-northeast of
fers from site to site. At any given site, ocean temperatures ENCI quarry (Fig. 1c). The exposed section in the GSGs lies
across the LMWE may be influenced by the primary CO2 - stratigraphically above the ENCI quarry and preserves the
driven global warming, which may have manifested in am- uppermost section of the Meerssen Member, including the
plified warming at high latitudes similar to modern Arc- KPB itself (Felder and Bosch, 1998; Keutgen, 2018). Be-
tic amplification (Petersen et al., 2016a), and also by local tween the top of the ENCI section and the short KPB interval
changes in ocean circulation, sea level, upwelling, and/or recorded in the GSGs, a time gap exists of no more than sev-
runoff (changes which may themselves be driven indirectly eral tens of thousands of years (Vellekoop et al., 2022).
by global climate change, e.g., sea level rise due to melting At ENCI quarry and in the surrounding region, the Maas-
of the Antarctic ice caps during the LMWE). Therefore, it tricht Formation contains multiple conspicuous fossiliferous
is important to consider the local paleogeographic context horizons (Felder, 1975; J. J. P. Zijlstra, 1988, unpublished
when interpreting paleoclimate trends during this tumultuous data, 1997), including nine horizons in the Maastricht For-
period. mation at ENCI quarry (Fig. 2). Fossil shell specimens were
Here, we produce a new climate record for a site in north- collected from five of these horizons (Lichtenberg, ENCI,
western Europe using the clumped and stable isotopic com- Romontbos, Lava, and Laumont), as well as from the top
position of bivalve fossils to better constrain the magnitude of the Emaël Member, the basal and middle intervals of
of climate change and possible ocean circulation fluctua- the Nekum Member, and the basal interval of the Meerssen
tions during this key geologic transition interval. We focus on Member (Fig. 2). Ages for all stratigraphic horizons, lithos-
the Maastrichtian type area (Jagt and Jagt-Yazykova, 2012) tratigraphic boundaries, and specimens were determined by
in the Belgium–Netherlands border region near Maastricht, updating the Keutgen (2018) age model using known bios-
Netherlands. The reconstructed paleogeography of this lo- tratigraphic markers and bulk carbon isotope data (Vellekoop
cality – a shallow epicontinental sea surrounded by several et al., 2022; Table S2). Most horizon and unit ages have an
low-lying landmasses (Fig. 1a) – makes this site sensitive uncertainty of ±0.05 Ma, with the exception of the Lichten-
not only to global climate changes, but also to local varia- berg Horizon (uncertainty of ±0.2 Ma), Valkenburg Mem-
tions in sea level and ocean circulation. We interpret our data ber, and St. Pieter Horizon (both ±0.1 Ma; Table S2). Two
in terms of previously described and modeled estimates for samples were collected from the middle Nekum Member at
local, regional, and global climate changes during this time. the CBR Romontbos quarry from an interval correlated with
Our study adds to a growing body of literature documenting samples above the middle Nekum Member from ENCI but
the impacts of volcanism and ocean circulation on climate, below the Kanne Horizon.
especially leading up to the K–Pg extinction. The K–Pg boundary is positively identified in the Maas-
tricht region and in particular in the GSGs through several
lines of evidence (Smit and Brinkhuis, 1996; Vellekoop et
Clim. Past, 18, 1963–1982, 2022 https://doi.org/10.5194/cp-18-1963-2022

H. E. O’Hora et al.: Clumped-isotope-derived end-Cretaceous climate trends in NW Europe 1965
Figure 1. Paleogeography and sample localities used in this study. (a) Late Cretaceous paleogeographic reconstruction of Europe and
surrounding regions based on Vellekoop et al. (2022) with the study area (yellow dot) and hypothesized cold- and warm-water currents (blue
and red arrows) marked. (b) A map of modern national borders showing the position of the study area. (c) A zoomed-in map of the Maastricht
region, showing the location of the ENCI quarry, CBR Romontbos quarry, and Geulhemmerberg subterranean galleries relative to each other
and to the city center of Maastricht.
al., 2020). The typical planktonic foraminiferal “disaster” canaliculate, Entolium membranaceum, Neithea regularis,
assemblage (Smit and Zachariasse, 1996), the presence of Pinna sp., Pseudoptera sp., Ceratostreon sp., and Rastellum
the earliest Paleocene dinocyst marker taxon Senoniasphaera sp. and indet. Ostreida). Six of the samples were only iden-
inornata (Brinkhuis and Schiøler, 1996; Herngreen et al., tified as some type of bivalve or oyster. The majority (40 of
1998), and 87 Sr/86 Sr analyses of well-preserved foraminifera 66) are either Acutostrea uncinella or Agerostrea ungulata.
from the clay layers of the Geulhemmerberg underground All identified specimens were sessile epifaunal suspension
galleries (Vonhof and Smit, 1996) have all demonstrated feeders and are assumed to have avoided significant lateral
that the K–Pg boundary occurs at the base of IVf-7 of the postmortem transport. In these units, all biogenic aragonite
Meerssen Member at the hardground known as the Berg en has dissolved, leaving behind hollowed molds. All taxa sam-
Terblijt Horizon (Fig. 2). Directly above this horizon, a se- pled here originally formed calcite shells, which continue to
quence of interbedded layers of shell hash and clays repre- be preserved as calcite today (Fig. 3). Water depths for this
sents a period of unique deposition interpreted during a se- location are estimated to be less than 20 m at all times and
ries of high-energy events (storms or seismites) in the 100– often shallower (Hart et al., 2016), so water temperatures re-
300 years immediately following the Chicxulub meteorite constructed from bottom-dwelling mollusks are reasonably
impact (Roep and Smit, 1996; Smit and Brinkhuis, 1996; interpreted to approximate mean annual surface air tempera-
Vellekoop et al., 2020). The one oyster sample collected from tures.
these shell hash layers was inferred to have been alive ei- One sample of carbonate-rich matrix material from the
ther within this short interval following the KPB or to be Meerssen Member was analyzed for comparison to fossil
reworked from just underneath the Berg en Terblijt Hori- shells. The carbonate in this specimen represents a combi-
zon and was thus assigned an age of 66.0 Ma (Clyde et al., nation of calcitic microfossils and possibly secondary, diage-
2016; Sprain et al., 2018) with an uncertainty of ±0.01 Ma netic calcite. As such, comparison between its isotopic and
(Vellekoop et al., 2022), equivalent to the KPB at the scale of trace element composition and that of fossil shells can help
our time series (Table S2). assess the potential for diagenetic alteration of shell samples.
Fossil shell specimens collected from the Maastricht For-
mation at the ENCI quarry, the CBR Romontbos quarry,
and the GSGs included 66 bivalve specimens representing
11 identified taxa of mostly oysters, with a few scallops and
clams (from most to least common: Acutostrea uncinella,
Agerostrea ungulata, Pycnodonte vesicularis, Gryphaeostrea
https://doi.org/10.5194/cp-18-1963-2022 Clim. Past, 18, 1963–1982, 2022

Figure 2. Composite stratigraphic section through the Maastricht Formation at ENCI quarry, Romontbos quarry, and Geulhemmerberg
subterranean galleries. Samples collected in the Romontbos quarry and GSGs were placed on the ENCI quarry stratigraphic framework using
the age model from Vellekoop et al. (2022; Table S2). From each collection horizon or zone, the number of specimens collected, the number
that passed diagenesis screening, the number of well-preserved specimens successfully analyzed for clumped isotopes, and the total number
of specimens attempted for clumped isotopes (including some poorly preserved specimens) are listed as follows (n = passed/collected,
N = good clumped/attempted). Out of 66 shells sample, 54 shells passed screening, 33 were attempted, and 27 well-preserved shells were
successful for 147 , plus one sample of matrix material from the Meerssen Member.
3 Methods Taxa-specific clumped isotope vital effects in bivalves are

considered uncommon based on studies of modern bivalve
3.1 Sample preparation taxa (Eagle et al., 2013; Henkes et al., 2013; Came et al.,
2014), although one recent study documented apparent vital
After removal of matrix material through deionized water effects in the juvenile portion only of the oyster Magellana
rinses and scrubbing with a toothbrush, fossil shell spec- gigas (Huyghe et al., 2022). In shells sampled for this study,
imens were sampled by either hand-drilling powder from we estimate that juvenile shell material made up a volumet-
larger fragments or manually crushing smaller fragments in rically small fraction of sampled shell areas. Combined with
their entirety in a mortar and pestle. The one sample of matrix the fact that our selected species differ from the taxa demon-
material from the Meerssen Member was scraped off a fos- strating vital effects (Huyghe et al., 2022), we interpret well-
sil and powdered in a mortar and pestle to achieve roughly preserved specimens as faithful recorders of paleoenviron-
uniform grain size. The sample powders were divided for mental temperatures at the time of life.
trace element analysis and isotopic analysis. Fragments of
some shell specimens were mounted for scanning electron
microscopy (SEM) observation to assess their preservation
state (Fig. 3).

Figure 3. SEM images of nine shell specimens representing varying levels of calcite preservation. Scale bars represent 1 µm. Images were
adjusted for brightness in Adobe Photoshop 2022. Trace element concentrations (in ppm) and SEM preservation indices quantifying the
level and prevalence of dissolution and secondary growth (SG) on a scale of 0–3 are shown below each image for comparison. Samples
in (a) through (e) were interpreted. Samples in (f) through (i) were not interpreted further due to too much evidence of secondary growth
and/or trace elements exceeding thresholds (see Fig. 4).
3.2 Preservation assessment (SEM and trace elements) tions (Fig. 3) using a Thermo Scientific Quadrupole-ICP-MS
(iCAP Q) equipped with an Elemental Scientific PrepFAST 2
The isotopic composition of fossil shells is susceptible to automation system. KED mode (He gas) was utilized for all
post-depositional alteration via recrystallization, which can isotopes. All samples were analyzed in 2 % HNO3 (w/v)
skew interpreted formation temperatures (Brand and Mor- (Optima grade), and sensitivity drift was corrected for with
rison, 1987). Elevated levels of trace elements (Mn and Fe standard-sample bracketing methods (Table S4).
especially) in calcitic fossil material may indicate diagene- A subset of 21 specimens representing most horizons
sis and/or recrystallization of the calcite (Land, 1967; Möller were photographed under SEM using a JEOL JSM-7800FLV
and Kubanek, 1976; Brand and Morrison, 1987; de Winter et field-emission SEM in the Robert B. Mitchell Electron Mi-
al., 2018). Additionally, the prevalence of diagenetic calcite crobeam Analysis Lab at the University of Michigan Ann
overgrowth on a sample can be identified easily under scan- Arbor (Fig. 3). SEM images were used to assign samples a
ning electron microscopy (SEM; Fig. 3), as diagenetic calcite score of 0–3 for both prevalence of dissolution (Diss) and
does not show the same microstructures as molluskan calcite. secondary growth (SG; Table S3). Samples that scored a
All but one of the fossil shells collected for this study were 2 or 3 for SG were deemed “bad” and not considered for
screened for recrystallization using a combination of SEM isotopic interpretation as the addition of diagenetic calcite
imaging and trace element detection methods (Figs. 3 and 4, with a (presumably) different isotopic composition than the
Table S3). base shell would skew environmental interpretations. Sam-
All but one of the 66 shells (as well as one matrix sample) ples with some evidence of dissolution (score of 2 or 3)
were analyzed in the Michigan Elemental Analysis Labora- but no secondary growth were deemed “okay” and were still
tory at the University of Michigan for Mn and Fe concentra-

each sample, individual replicates were deemed to be outliers

if they fell more than 4 standard deviations outside the mean
of the remaining three to four replicates for a given sample. A
total of 108 replicates representing 34 unique samples were
analyzed over a span of 7 years using two different machines
and procedures (see below). Only samples with at least three
acceptable replicates were considered (n = 27 shells + one
matrix sample prior to preservation screening; Table 1).
Between 2016 and 2020, a first set of fossil shell pow-
ders were measured for clumped isotopic composition (147 )
in the University of Michigan Stable Isotope Lab using a
Thermo-Finnigan MAT 253 dual-inlet isotope ratio mass
spectrometer and a manual sample preparation device de-
scribed in detail by Defliese et al. (2015), with updates to
Figure 4. Iron and manganese concentrations of all samples used Porapak temperature outlined by Petersen et al. (2016c).
for diagenesis screening. Samples with Fe concentrations above Briefly, 3.5–5 mg of powder is reacted at 75 ◦ C in a common
2500 ppm or Mn concentrations above 100 ppm (or both) were not acid bath, and evolved CO2 is cryogenically separated from
considered for further interpretation. These thresholds were cho- water and other contaminants through two −80 ◦ C ethanol–
sen based on prior studies (see text), then adjusted to lower, more liquid nitrogen traps and one PorapakQ and silver wool trap.
conservative levels to screen out most samples showing secondary Clean CO2 is measured for five acquisitions of 12 cycles
growth while keeping all samples showing good preservation or at a major beam (m/z 44) voltage of 16 V. Raw voltages
solely dissolution under SEM (see Fig. 4).
of masses 44 to 49 were converted to bulk isotopic values
(δ 45 to δ 49 ) and clumped isotopic values (147 , 148 , and 149 )
using IUPAC (Brand) parameters with an R code presented
considered further because the material that remained likely in Petersen et al. (2019). Raw 147 values were placed in a
still represented inhabited environmental conditions. Sam- gas-based absolute reference frame following methods intro-
ples with scores of 0 or 1 for both SG and dissolution were duced by Dennis et al. (2011) using theoretical equilibrium
deemed “good” and were fully considered. values for heated (1000 ◦ C) and water-equilibrated (25 ◦ C)
standard gases (0.0266 and 0.9198, respectively) and the
3.3 Stable and clumped isotope analysis 75 ◦ C acid fractionation factor of +0.072 ‰ for 147 from Pe-
tersen et al. (2019). Boundaries between each 1- to 3-week-
Powders drilled or crushed from all shell specimens and one long correction window were selected to account for drift in
matrix sample were analyzed for oxygen (δ 18 O) and carbon equilibrium gas line slopes and to maintain consistent val-
(δ 13 C) stable isotope compositions (Figs. S1–S3 in the Sup- ues of in-house carbonate standards through time. δ 18 O of
plement, Table 1). Powders from 39 shells were analyzed us- carbonate (δ 18 Ocarb ) was calculated from δ 18 O of evolved
ing a Thermo-Finnigan Kiel IV automated carbonate device CO2 using an empirical acid fractionation factor defined us-
coupled to a MAT 253 gas source dual-inlet mass spectrom- ing in-house carbonate standards independently calibrated to
eter in the University of Michigan Stable Isotope Laboratory. NBS-18 and NBS-19 (α = 1.007888 for calcite at 75 ◦ C; Ta-
Data were corrected for acid fractionation and normalized to ble S4). Long-term standard deviations of δ 13 C, δ 18 O, and
the VPDB (Vienna Pee Dee Belemnite) scale using a single- 147 on the MAT253 were consistent across time at approx-
step transfer function between two anchor points, the NBS- imately 0.1 ‰, 0.3 ‰, and 0.020 ‰, respectively, based on
18 and NBS-19 standards. Precision was maintained at less in-house carbonate standards run many times over the mea-
than 0.1 ‰ for both δ 18 O and δ 13 C (Tables 1 and S4). surement interval (Table S4).
The remaining 27 samples plus seven of the above 39 were In 2021 and 2022, additional powders were analyzed in
analyzed for δ 18 O and δ 13 C in conjunction with clumped the University of Michigan SCIPP lab using a Nu Perspec-
isotope analysis in the University of Michigan Stable and tive isotope dual-inlet ratio mass spectrometer connected to
Clumped Isotopes for Paleoclimatology and Paleoceanogra- a NuCarb automated sample preparation device (Jones et al.,
phy (SCIPP) Laboratory (Figs. S1–S3 in the Supplement, Ta- 2022). This device reacts sample powders in individual vials
ble 1). During clumped isotope analysis, one measurement under vacuum at 70 ◦ C by slowly injecting 150 µL of 103 %
simultaneously acquires 147 , which is a function of carbon- phosphoric acid. Vials are isolated for the first 5 min of the
ate formation temperature (or past seawater temperature in reaction to reduce excess bubbling, then opened to a vari-
the case of fossil shells), δ 13 C, and δ 18 O of CO2 gas evolved able temperature trap held at −160◦ for an additional 15 min,
from CaCO3 (Eiler, 2011). Samples are typically replicated continuously collecting evolved CO2 . Next, this variable-
a minimum of three times, and replicates are averaged to cal- temperature trap is elevated to −60 ◦ C, releasing CO2 to
culate a sample mean. Here, after combining all replicates for pass through a static trap filled with Porapak Q material held

Table 1. All isotope results. Sample names were created following a convention of three to four capital letters relating to the horizon, then three to four letters relating to the taxa, and
then one capital letter to separate multiple shells of the same taxa from that horizon. The age model is from Vellekoop et al. (2022; see Table S2). The preservation index (P index)
column combines the results of trace element analysis (good trace – GT, elevated iron – Fe, elevated manganese – Mn, or both – FeMn; or NA if not measured) and SEM imagery
(GOOD, OK, BAD, or NA if not imaged) in the format of TRACE/SEM. Any sample that failed ≥ 1 criteria was not considered further. Errors for 147 , temperature, and δ 18 Osw are
external 1 SE (see text).
Sample name Age (Ma) Height (m) P index δ 18 O δ 18 O SD δ 13 C δ 13 C SD 147 147 SE Temp (◦ C) Temp SE (◦ C) δ 18 Osw δ 18 Osw SE
LIC-Psop 67.1 41.45 GT/NA −1.3 0.04 2.4 0.02 – – – – – –
LIC-Pycn 67.1 41.45 Fe/BAD −1.3 0.1 2.2 0.1 0.698 0.009 22.0 2.9 0.4 0.6
LIC-AcutA 67.1 41.45 GT/GOOD −1.2 0.1 2.7 0.1 0.699 0.011 21.7 3.6 0.5 0.8
LIC-Gry 67.1 41.45 GT/NA −1.2 0.2 2.5 0.1 0.695 0.013 23.1 4.4 0.7 0.9
LIC-AcutB 67.1 41.45 GT/NA −1.1 0.1 2.9 0.03 – – – – – –
LIC-Ager 67.1 41.45 GT/NA −1.0 0.02 2.5 0.01 – – – – – –
LIC-Biv 67.1 41.45 GT/NA −1.7 0.04 1.8 0.01 – – – – – –
LIC-Ento 67.1 41.45 GT/NA −2.3 0.01 1.4 0.04 – – – – – –
VAL-Ager 67.0 42.5 GT/NA −1.7 0.1 1.7 0.05 0.707 0.015 19.2 5.1 −0.5 1.1
ENC-Pinna 66.85 44.4 GT/NA −0.7 0.1 2.6 0.03 – – – – – –
ENC-BivA 66.85 44.4 FeMn/NA −1.5 0.1 1.4 0.04 0.692 0.003 24.0 0.9 0.7 0.2
ENC-BivB 66.85 44.4 FeMn/BAD −1.8 0.1 1.0 0.1 0.664 0.005 34.1 2.0 2.4 0.3
GRO-Biv 66.7 46 Mn/NA −1.6 0.1 1.7 0.04 0.701 0.011 20.9 3.6 0 0.8
GRO-Ost 66.7 46 GT/NA −1.4 0.04 2.5 0.05 0.698 0.001 22.0 0.3 0.4 0.1
ROT-Acut 66.55 52.7 GT/NA −1.4 0.2 2.6 0.1 0.710 0.005 17.8 1.8 −0.5 0.5
ROT-Ento 66.55 52.7 GT/NA −1.0 0.1 1.7 0.03 0.695 0.011 22.9 3.9 1.0 0.8
LAV-AcutG 66.4 56.75 GT/GOOD −1.4 0.1 3.2 0.03 – – – – – –
LAV-NeitA 66.4 56.75 Fe/NA −1.3 0.1 1.1 0.03 – – – – – –
LAV-NeitB 66.4 56.75 GT/GOOD −1.3 0.02 1.6 0.01 – – – – – –
LAV-AcutA 66.4 56.75 GT/NA −0.6 0.2 2.8 0.1 0.704 0.019 20.8 6.4 0.8 1.4
LAV-AcutB 66.4 56.75 GT/GOOD −0.8 0.1 2.8 0.2 0.699 0.022 22.3 6.9 0.9 1.5
LAV-AcutC 66.4 56.75 GT/NA −1.2 0.1 3.2 0.02 – – – – – –
LAV-AcutD 66.4 56.75 GT/NA −1.2 0.04 2.9 0.02 – – – – – –
LAV-AcutE 66.4 56.75 GT/NA 0.0 0.05 3.3 0.03 0.738 0.013 9.5 3.9 −0.9 0.8
LAV-AcutF 66.4 56.75 GT/NA −1.0 0.04 2.4 0.03 – – – – – –
LAV-Ost 66.4 56.75 GT/NA 0.7 0.1 3.0 0.04 – – – – – –
LAV-Ager 66.4 56.75 GT/NA −0.9 0.2 2.5 0.1 0.724 0.016 14.2 4.9 −0.9 1.1
H. E. O’Hora et al.: Clumped-isotope-derived end-Cretaceous climate trends in NW Europe
LAV-Psop 66.4 56.75 GT/NA −1.7 0.1 2.9 0.1 – – – – – –

EMAU-AgerA 66.375 60 GT/NA −1.2 0.04 3.3 0.04 – – – – – –
EMAU-AgerB 66.375 60 GT/BAD −1.4 0.1 3.3 0.1 – – – – – –
EMAU-AgerC 66.375 60 GT/GOOD −1.2 0.3 3.1 0.2 0.704 0.007 19.8 2.1 0.2 0.5
EMAU-AcutA 66.375 60 NA/NA −1.3 0.2 3.1 0.1 0.704 0.014 20.1 4.4 0.1 1.0
EMAU-AcutC 66.375 60 GT/NA −1.0 0.04 2.8 0.02 – – – – – –
EMAU-AcutD 66.375 60 GT/NA −1.4 0.1 2.9 0.01 – – – – – –
EMAU-AcutE 66.375 60 GT/NA −1.3 0.1 3.0 0.04 – – – – – –
EMAU-AcutF 66.375 60 GT/GOOD −0.5 0.01 3.4 0.01 – – – – – –
EMAU-AgerD 66.375 60 GT/NA −1.3 0.01 3.3 0.01 – – – – – –
Clim. Past, 18, 1963–1982, 2022

1969
H. E. O’Hora et al.: Clumped-isotope-derived end-Cretaceous climate trends in NW Europe
Table 1. Continued.
Sample name Age (Ma) Height (m) P index δ 18 O δ 18 O SD δ 13 C δ 13 C SD 147 147 SE Temp (◦ C) Temp SE (◦ C) δ 18 Osw δ 18 Osw SE
EMAU-AcutG 66.375 60 GT/GOOD −1.5 0.03 3.0 0.02 – – – – – –
EMAU-AcutH 66.375 60 GT/NA −1.3 0.03 4.0 0.01 – – – – – –
EMAU-Cer 66.375 60 GT/NA −0.7 0.01 1.9 0.01 – – – – – –
LAU-Unoy 66.35 61.15 GT/OK −1.3 0.02 3.2 0.02 – – – – – –
LAU-AcutA 66.35 61.15 FeMn/NA −0.8 0.1 3.2 0.04 – – – – – –
LAU-AcutB 66.35 61.15 GT/NA −1.3 0.03 3.2 0.02 – – – – – –
LAU-AcutC 66.35 61.15 GT/NA −1.2 0.02 3.2 0.01 – – – – – –
LAU-AcutD 66.35 61.15 GT/NA −1.0 0.1 3.3 0.01 – – – – – –
LAU-AcutE 66.35 61.15 GT/GOOD −0.7 0.1 3.4 0.04 – – – – – –
LAU-AcutF 66.35 61.15 GT/NA −1.1 0.3 2.8 0.2 0.717 0.014 15.8 4.3 −0.7 1.1
LAU-AcutG 66.35 61.15 GT/GOOD −1.3 0.1 3.3 0.03 0.698 0.003 21.8 1.2 0.4 0.3
NEKB-AcutA 66.325 62 GT/NA −0.7 0.2 3.4 0.1 0.683 0.008 27.2 3.0 2.1 0.7
NEKB-AcutBC 66.325 62 GT/GOOD −1.2 0.1 3.3 0.1 – – – – – –
NEKB-AcutD 66.325 62 GT/GOOD −1.7 0.01 3.1 0.01 0.690 0.007 24.7 2.2 0.7 0.5
NEKB-AcutE 66.325 62 GT/BAD −0.9 0.03 2.7 0.04 – – – – – –
NEKB-AcutF 66.325 62 Fe/NA −0.5 0.02 2.8 0.01 – – – – – –
NEKB-Ager 66.325 62 GT/NA −1.2 0.1 2.6 0.03 0.689 0.003 25.1 0.9 1.2 0.2
NEK-Neit 66.3 64 GT/BAD −1.4 0.1 1.8 0.1 – – – – – –
NEK-Acut 66.3 64 GT/BAD – – – – – – – – – –
NEK-NeitB 66.3 64 GT/NA −1.7 0.2 1.6 0.03 0.697 0.013 22.3 4.3 0.1 0.8
CBR-Acut 66.275 65 GT/NA −2.1 0.1 2.8 0.05 0.702 0.013 20.8 4.4 −0.5 1.0
CBR-Biv 66.275 65 GT/NA −1.2 0.1 3.1 0.1 0.706 0.003 19.4 1.0 0 0.3
Clim. Past, 18, 1963–1982, 2022

MEER-RasA 66.07 76 GT/GOOD −2.0 0.02 2.3 0.05 0.698 0.001 21.9 0.4 −0.2 0.1
MEER-RasB 66.07 76 GT/GOOD −1.3 0.2 2.1 0.1 0.719 0.019 15.9 5.6 −0.9 1.2
MEER-mat 66.07 76 GT/NA −0.1 0.1 1.3 0.04 0.705 0.011 19.8 3.5 1.2 0.8
GCAV-OystA 66.0 85 GT/GOOD −2.3 0.1 2.7 0.02 0.699 0.003 21.6 1.0 −0.6 0.2
1970
at −30◦ . CO2 is frozen on the far size in another variable- dards CM, OO, and CORS) or comparison to ETH standards
temperature trap set at −160◦ for 800 s. CO2 is warmed up over a 6-month measurement session in 2020 (for all five;
to the gas phase, and the yield is recorded with a transducer. Table S4). Close agreement in 147 , δ 18 O, and δ 13 C between
The Nu Perspective and NuCarb can handle both larger (3– long-term values and session-specific values on all three in-
6 mg) and smaller (350–450 µg) samples. Following purifi- struments confirms that data from different measurement ses-
cation, depending on the transducer reading, the automated sions and instruments can be directly compared (Table S4).
sequence directs CO2 from a larger sample to equilibrate Comparison of replicates of the same sample powder run on
into the bellows or freezes CO2 from a smaller sample into a each machine supports this.
cold finger between the bellows and the change-over block. Finally, acid-corrected 147 values (CDES25) from both
All sample replicates in this study were 3–5 mg in size and measurement periods were converted to temperature us-
were therefore analyzed in “bellows mode”. Previous stud- ing the composite calibration of synthetic carbonates from
ies using a Nu Perspective and NuCarb for clumped isotope Petersen et al. (2019) [Temp(◦ C) = (0.0383 × 106 )/(147 −
analysis described “cold-finger mode” (Mackey et al., 2020). 0.258) − 273.15]. This equation was chosen in favor of the
In bellows mode, the bellows are compressed until a desired ETH-standard-based calibration of Anderson et al. (2021)
beam strength of 80 nA is achieved on the major (m/z 44) ion due to the use of a gas-based reference frame for the early
beam. Gas is analyzed for four blocks of 20 reference sample measurements and the fact that calibration data acquired on
cycles, with 20 s of integration on each half-cycle. Bellows the MAT 253 by Defliese et al. (2015) and Winkelstern et
are continually adjusted between each cycle to maintain the al. (2016) are included in the Petersen et al. (2019) composite
initial beam strength at all times. calibration equation. Use of the Anderson et al. (2021) equa-
As above, raw beam intensities of masses 44–49 were con- tion results in temperatures 1.5–2.5 ◦ C cooler and δ 18 Osw
verted to bulk isotopic values (δ 45 to δ 49 ) and clumped iso- values proportionally lower but does not change any of the
topic values (147 , 148 , and 149 ) using IUPAC (Brand) pa- conclusions in this study. Carbonate δ 18 O values and 147 -
rameters (Brand et al., 2010) with an R code presented in derived temperatures were used to calculate seawater δ 18 O
Petersen et al. (2019), adjusted slightly to accommodate dif- (δ 18 Osw ) for each sample via the calcite–water relationship
fering data formats from the Nu instrument. Isotopic val- of Kim and O’Neil (1997). All δ 18 Osw values are reported
ues were converted into the Intercarb Carbon Dioxide Equi- on the VSMOW scale (Table 1).
librium Scale (I-CDES25) absolute reference frame using Long-term reproducibility (1SD) of 147 in carbonate stan-
147 values for four ETH standards defined by the Inter- dards (0.020 ‰ for MAT253 and 0.018 ‰ for Nu) was used
carb project (Meckler et al., 2014; Bernasconi et al., 2021) to calculate a long-term standard error in 147 for each sam-
and a 147 acid fractionation factor of +0.066 ‰ for 70 ◦ C ple. For temperature and δ 18 Osw values, a long-term standard
from Petersen et al. (2019). First, a single slope was fitted error was calculated using the long-term standard error in
through ETH 1 and ETH 2 (adjusted by 0.0033 ‰) in δ 47 147 and the pseudo-linear relationship between internal 1 SE
vs. 147 space, then all four ETH standards were used for the on 147 and either temperature or δ 18 Osw in all samples (for
empirical transfer function step. Some heated gases (1000 this study: long-term 1 SE in temperature = 306.7× (long-
and 200 ◦ C) and equilibrated gases (25 ◦ C) were analyzed term 1 SE in 147 ) + 0.1778, and long-term 1 SE in δ 18 Osw =
for cross-comparison but were not used in reference frame 66.33× (long-term 1 SE in 147 )+0.0461). The external 1 SE
calculations. Carbonate δ 18 O and δ 13 C were each calculated for all three parameters was selected as the larger of either the
via a single-step transfer function defined by the four ETH internal or the long-term standard error. All replicate-level
standards using values published in Bernasconi et al. (2018). clumped isotope and stable isotope data have been submit-
Although use of ETH standards does not require an acid frac- ted to the permanent data archive EarthChem as part of the
tionation factor when applied to an entire dataset, combining ClumpDB database.
data from gas-based and carbonate-based reference frames
requires normalization to the same acid digestion tempera-
4 Results
ture. Long-term standard deviations of δ 13 C, δ 18 O, and 147
on this instrument were consistent across time at better than 4.1 Preservation and sample screening
0.05 ‰, 0.1 ‰, and 0.018 ‰, respectively (Table S4).
Clumped isotope (147 ) values for between two and five Isotopic signatures may be affected by both bulk chemical
in-house carbonate standards were tracked to ensure con- alteration (dissolution and recrystallization replacing origi-
sistency and comparability of isotope values through time nal shell material and/or overgrowth of secondary material
and across machines. Stable isotope (δ 13 C and δ 18 O) val- changing the bulk isotopic composition) and solid-state bond
ues for in-house carbonate standards were determined rela- reordering (changes in the number of heavy isotope clumps
tive to NBS-18 and NBS-19 on the same Kiel IV + MAT253 in the crystal lattice, overprinting the original recorded tem-
used for bulk sample δ 13 C and δ 18 O (Table S4). 147 values perature without any mass transfer). Overall, SEM preser-
for these standards were defined based on comparison to gas vation ratings and trace element concentrations were low in
standards over multiple years (2015 and 2018–2019 for stan- most samples (Figs. 3 and 4, Table S3), ruling out signifi-

cant chemical alteration, and a mild thermal history suggests geochemical alteration. This also explains how three sam-
reordering is not a concern. ples failed SEM screening (Fig. 3) but showed low trace el-
In the 21 shells imaged for SEM, 6 displayed evidence ement concentrations (Fig. 4). We infer that in these sam-
of secondary growth (SG = 2 or 3) and were not considered ples, although some secondary growth was seen under SEM
further. One sample showed evidence of dissolution but no (Fig. 3), it must have been volumetrically insignificant in the
secondary growth and was deemed “okay” and included in total powdered sample, not present in the portion of the shell
interpretations. A total of 14 samples showed little to no ev- that was powdered (which differed from the portion imaged
idence of secondary growth (SG = 0 or 1) and were deemed under SEM), and/or did not have elevated trace element con-
“good” and included. In the 65 shell specimens tested for centrations.
trace elements, Mn concentration ranges from 17–210 ppm, In total, 12 of 65 shell samples failed one or both screening
while Fe concentration ranges from 96–7151 ppm. The ma- techniques, with 9 failing trace element thresholds, 6 failing
trix sample was measured to have 60 ppm Mn and 664 ppm SEM criteria, and 3 of the 6 or 9 failing both. The one sample
of Fe. Threshold values of 100 and 2050 ppm were selected that was not measured was assumed to pass screening based
for Mn and Fe, respectively, to exclude most samples that on the majority of samples passing screening. An additional
failed SEM screening, while including all samples that ex- three samples failed stable isotope measurements for analyt-
hibited only original material under SEM (Figs. 3 and 4). ical reasons, and two more failed to achieve the three good
These thresholds are more conservative than those proposed replicates necessary for 147 but were still used for δ 18 O and
by other studies (Brand and Morrison, 1987; Voigt et al., δ 13 C. In the six samples that were measured on both the Kiel
2003; Ullmann et al., 2013, de Winter et al., 2018) but, in and one of the clumped isotope machines, clumped-isotope-
total, still only result in the flagging of 9 specimens out of derived δ 18 O and δ 13 C values were used (but only differed
the total 65 (Fig. 4, Table S3). by 0.1 ‰–0.2 ‰ from Kiel-derived values). After removing
Bond reordering without exchange of mass can occur if poorly preserved and insufficiently analyzed samples, the
samples are heated above 100–150 ◦ C (Henkes et al., 2014; dataset includes 51 well-preserved shell samples analyzed
Winkelstern and Lohmann, 2016), such as during burial. for δ 18 O and δ 13 C and 23 well-preserved shell samples with
Heating breaks and reforms bonds between atoms in the car- acceptable 147 analyses (Table 1). All further interpretation
bonate lattice and changes the clumped isotopic composition excludes discussion of removed samples.
– and therefore interpreted temperature – without affecting
the bulk δ 18 O or δ 13 C composition appreciably. Regional ge- 4.2 Stable isotopes
ologic evidence suggests a relatively shallow burial history
for these strata. In the southern Netherlands, the Roer Valley The 51 shell specimens exhibit a range of carbonate δ 18 O
Graben contains several basin-wide unconformities dated to (δ 18 Ocarb ) values between −2.3 ‰ and −0.5 ‰ VPDB, with
the Late Cretaceous and middle Paleocene that signify two 2 samples showing higher values of 0.0 ‰ and 0.7 ‰ VPDB
basin inversions, leaving the Maastrichtian and Danian strata (Fig. S1 in the Supplement). The average δ 18 Ocarb value
relatively undisturbed (Luijendijk et al., 2011, and references throughout the section is −1.2±0.5‰ (1 SD) VPDB. In con-
therein). Outstandingly well-preserved mosasaur skeletons trast, the carbonate fraction of bulk matrix in the one sample
and hollow aragonitic molds in the Maastricht Formation fur- from the Meerssen Member measured here and in a high-
ther indicate a lack of burial or significant diagenetic fluid resolution record from the Lichtenberg Horizon to the top
flow during this time (Dortangs et al., 2002; Jagt et al., 2016). of the Nekum Member (Vellekoop et al., 2022) falls any-
Overall, these burial conditions are not expected to result in where from ∼ 0 ‰ to ∼ 1 ‰ higher than molluskan car-
bond reordering in any samples. bonate (Fig. S1 in the Supplement), with a mean value of
Isotopic evidence supports the idea that selected trace el- −0.6 ± 0.4 ‰ (1 SD) VPDB and a range of −1.7 ‰ to 0.0 ‰
ement and SEM thresholds are conservative and that over- VPDB through the section.
all diagenetic alteration in this region is mild to absent. De- Carbonate δ 13 C (δ 13 Ccarb ) values range from 1.4 ‰ to
spite being excluded based on trace element and/or SEM re- 3.4 ‰ VPDB, with one high sample at 4.0 ‰ VPDB
sults, the isotopic values of excluded samples were often not (Fig. S2). The average δ 13 Ccarb value is 2.8 ± 0.6 ‰ (1 SD)
substantially different than other shells that passed diagene- VPDB. Bulk matrix δ 13 C is clearly lower than most mol-
sis screening (e.g., only one of four measured for clumped luskan carbonate by anywhere from 1 ‰ to 2 ‰ VPDB
isotopes had a higher-than-normal 147 -derived temperature; (Fig. S2). The average bulk matrix δ 13 C value is 1.7 ± 0.1 ‰
Fig. 5). Additionally, the matrix sample taken from the (1 SD) VPDB with a range of 1.3 ‰ to 1.9 ‰ VPDB through
Meerssen Member showed trace element concentrations be- the section.
low thresholds (Fig. 4) and a temperature of 19.8 ± 3.7 ◦ C In most cases, the number of specimens representing
(Fig. 5), very similar to temperatures from fossil shells, in- each taxon is too small and/or the stratigraphic range is
dicating that the bulk matrix – which is more likely to be too limited to make reliable interspecies comparisons (e.g.,
affected by recrystallization or reordering than a dense cal- Rastellum sp.). Even considering this, the pectinids Entolium
cite shell – has also not undergone noticeable isotopic or membranaceum and Neithea regularis appear to record car-

Figure 5. Maastrichtian 147 -derived paleotemperatures for all samples. Samples shown in red (n = 4) did not pass screening thresholds
and were not included in further interpretation. Error bars represent the larger of either the internal or long-term 1 SE (see text). Onset of
the main Deccan Traps eruptions (grey shading) at 66.413 ± 0.067 Ma (Sprain et al., 2019). Member names and horizon abbreviations are
labeled across the top of the figure (see Fig. 2). ∗ Schiepersberg Member.
bon isotope compositions ∼ 1 ‰–1.5 ‰ more depleted rel- ception is the Lava Horizon (66.4 Ma), which shows a large
ative to other species, which mainly represent Ostreida, spread in temperatures from 9.5 ± 3.8 ◦ C (external 1 SE) to
while oxygen isotopic compositions appear similar (Fig. S3). 22.3±6.9 ◦ C (external 1 SE; Fig. 5). This may represent sea-
This inter-order difference may represent differing use of sonal aliasing in samples (e.g., powdering a portion of shell
metabolic (respired) carbon vs. environmental carbon be- representing wintertime growth vs. summertime growth), al-
tween these two orders of bivalvia. Between the two most though these samples were not treated any differently than
common taxa (Acutostrea uncinella and Agerostrea ungulate other horizons where this is not observed. This may also rep-
– both oysters), there does not appear to be any systematic resent long-term time averaging into a single horizon, where
offset in oxygen or carbon isotopic compositions (Fig. S3). individual shells may have lived thousands of years apart.
It is not a species-related offset, as the highest and lowest
temperatures both occur in Acutostrea uncinella (Fig. 5). A
4.3 Temperature estimates
similar argument can be extended to other horizons where a
A total of 4 of the 27 shell specimens successfully measured larger spread is seen, such as the lower Meerssen Member
for 147 were excluded from further interpretation due to ev- (66.1 Ma) and the Romontbos Horizon (66.55 Ma), although
idence of diagenesis (described above). The 23 samples that samples in these layers are still within error at the external
pass diagenetic screening criteria exhibit paleotemperatures 1 SE level.
ranging from 9.5–27.2 ◦ C, with a mean value of 20.4±3.8 ◦ C A fluctuating temporal trend is observed in which pale-
(1 SD; Fig. 5). In all instances in which multiple taxa were otemperatures rise from a low at 66.4 Ma to a peak in the
sampled from the same horizon, measured temperatures are basal Nekum Member at 66.325 Ma, followed by a decrease
within error of each other, suggesting no interspecies offset by 66.275 Ma that persists through the KPB (Fig. 5). The
due to vital effects. In the absence of vital effects in any taxa, onset of warming coincides with the onset of Deccan vol-
all species are expected to record the same temperature due to canism at 66.413 ± 0.067 Ma (Sprain et al., 2019; Fig. 5),
their common benthic habitat. The high agreement between within the error of our age model (Table S1 in the Supple-
taxa from the same horizon seen here is similar to that seen ment). We therefore identify this warming event as the named
in other deposits of the same age from the Gulf Coastal Plain Late Maastrichtian Warming Event (LMWE; Woelders et al.,
(Meyer et al., 2018), although further replication to reduce 2018; Hull et al., 2020).
error bar size would be necessary to rule out small vital ef-
fects.
In most horizons with multiple samples, estimated temper-
atures overlap within error for all samples (Fig. 5). The ex-

4.4 Seawater δ 18 O estimates pCO2 levels also display a similarly warmer climate glob-
ally as a result of CO2 -induced warming (Tabor et al., 2016;
Calculated δ 18 Osw values ranged from −0.9 ‰ to +2.1 ‰ Ladant et al., 2020). Modeled zonal mean temperatures at a
VSMOW (Fig. S4). The average δ 18 Osw value is +0.2 ± paleolatitude of 40◦ N are 20.2 ◦ C for a 2× CO2 (560 ppm)
0.8 ‰ (1 SD) VSMOW, which is above the ice-free mean Maastrichtian scenario or 21.4 ◦ C for a 4× CO2 (1120 ppm)
ocean water estimate of −1.0 ‰ VSMOW. This heavier Maastrichtian scenario (Tabor et al., 2016).
average δ 18 Osw value is consistent with the presence of Multiple paleoclimate records document a Late Maas-
some continental ice volume (presumed to be located on trichtian Warming Event (LMWE) (e.g., Stott and Kennett,
the Antarctic continent) during the cool greenhouse Maas- 1990; Li and Keller, 1998; Petersen et al., 2016a; Woelders
trichtian interval (Miller et al., 2005; Petersen et al., 2016a; et al., 2017, 2018; Barnet et al., 2019; Gao et al., 2021; Nava
Ladant and Donnadieu, 2016). Just like with temperature, et al., 2021) beginning gradually at 66.4 Ma with more rapid
where multiple taxa were sampled within a given horizon warming at 66.3 Ma (Hull et al., 2020; Fig. 7). The timing of
δ 18 Osw values are within error of each other, although there this LMWE appears to closely follow the onset of LIP vol-
is fairly large spread in the Lava Horizon (Fig. S4). canism on the Indian subcontinent based on 40 Ar/39 Ar ages
Seawater δ 18 O values show a similar temporal trend as for the oldest basalts of the Deccan Traps (66.413±0.067 Ma
that of temperature (Fig. 6c and d). We observe peak δ 18 Osw from the Jawhar Formation; Sprain et al., 2019) and shifts
values at the base of the Nekum Member (1.4 ± 0.7 ‰ VS- in marine osmium isotope ratios towards mantle values be-
MOW, horizon mean and 1 SD of three samples), and the ginning ∼ 400 kyr prior to the KPB (Ravizza and Peucker-
lowest δ 18 Osw values occur at the Lava Horizon in the Emaël Ehrenbrink, 2003; Hull et al., 2020), although others argue
Member (−1.0 ± 1.8 ‰ VSMOW, horizon mean and 1 SD for a slightly later onset around 66.311 ± 0.051 Ma (Schoene
of three samples; Figs. 6d and S4). This relationship is not et al., 2019; Nava et al., 2021). Volcanically derived CO2 is
a function of calculations of temperature or δ 18 Osw and, in- widely considered to be the cause of this warming.
stead, falls naturally out of the δ 18 Ocarb –δ 18 Osw –temperature The 147 -derived temperature record generated here doc-
relationship. Across the entire dataset, both temperature and uments a warming in the Nekum Member beginning from
δ 18 Osw show weak correlation with δ 18 Ocarb , with neither as a temperature low in the Lava Horizon at 66.4 ± 0.05 Ma
the primary sole driver (Fig. S5). (16.6 ± 5.9 ◦ C, 1 SD, N = 4) and increasing to the basal
Nekum at 66.325±0.05 Ma (25.7±1.4 ◦ C, 1 SD, N = 3) over
5 Discussion
75 kyr (Fig. 5). The magnitude of this warming is calculated
as 9.1 ◦ C going from these listed horizons or, more conserva-
5.1 Further evidence for a warmer Maastrichtian world tively, 5.8 ◦ C comparing the warmest horizon (basal Nekum)
and a Late Maastrichtian Warming Event to a pre-event baseline averaging the preceding seven hori-
zons (19.9 ± 2.0 ◦ C, N = 14). Although this warming event
New data presented here indicate that mean annual ocean is only barely statistically distinguishable as differing from
temperatures in the Maastricht region (paleolatitude ∼ the mean of the full record when considering the size and
40◦ N; van Hinsbergen et al., 2015) varied from 9.5 ± 3.9 to meaning of the 1 SE error bars and the fairly small number of
27.2 ± 3.7 ◦ C between 67.1 and 66.0 Ma, with a mean tem- samples representing peak warmth (n = 3; Fig. 5), the close
perature of 20.4 ± 3.8 ◦ C (1 SD) over this interval. This is agreement between all three points from this horizon and the
much warmer than the modern-day climate of Maastricht, following additional lines of evidence support the interpreta-
which has a mean annual air temperature of 9.8 ◦ C (Maas- tion of this warming as an expression of the LMWE.
tricht Climate, 2021). Warmer-than-modern temperatures in Based on a newly updated age model (Vellekoop et al.,
the Maastrichtian have been seen before in many studies 2022; Fig. S1 in the Supplement), the warmest 147 -derived
(Table S1 in the Supplement) and are expected due to el- temperatures slightly follow or align with the onset of earli-
evated atmospheric CO2 levels during the Maastrichtian of est Deccan eruptions (depending on whether a date of 66.413
∼ 1000 ppm on average over the study interval (Zhang et or 66.3 Ma is used) and the general age of the globally de-
al., 2018; Henehan et al., 2019; Hoenisch, 2021). Recon- fined LMWE (Hull et al., 2020; Fig. 7). The dinocyst Palyn-
structed Maastrichtian marine temperatures from sites lo- odinium grallator – a marker for the LMWE at northern mid-
cated between 30 and 50◦ N show an average temperature latitudes (Vellekoop et al., 2018; 2019) – first appears in the
of 19.8 ± 5.2 ◦ C (1 SD, n = 23 studies; Table S1 in the Sup- Maastricht Formation in the Nekum Member (Schiøler et al.,
plement), in close agreement with our site mean. Outside the 1997; Vellekoop et al., 2019). Thermophilic hermatypic scle-
marine realm, Maastrichtian-aged fossil plants also suggest ratinian macrofossils also first emerge in high abundances in
that terrestrial temperatures at 40◦ N were warmer, measur- the Nekum Member (Liebau, 1978; Leloux, 1999; Fig. 6e).
ing around 15 ◦ C (Golovneva, 2000), and that near 49◦ N, Together, these records suggest that the LMWE is preserved
air temperatures varied from 10–18 ◦ C during this time in- in the Nekum Member at ENCI quarry and that the warmer
terval (Wilf et al., 2003). General circulation models using 147 -derived temperatures found there are likely real.
Late Cretaceous paleogeography and elevated atmospheric

Figure 6. Isotopic compositions and paleotemperatures of fossil shells compared to inferred relative sea level in the Maastricht Formation.
Error bars represent 1 SD for δ 18 O (a) and δ 13 C (b) and represent external 1 SE for temperature (c) and δ 18 Osw (d) (see Sect. 3.3). All
samples failing diagenetic screening criteria have been removed. Colored lines pass through horizon means. Age model from Vellekoop et
al. (2022; Table S3). The relative sea level curve (e) is based on Schiøler et al. (1997), adjusted to the age model used in this study. Hermatypic
coral appearance data (e) are from Leloux (1999). Onset of the main Deccan Traps eruptions (grey dotted line) at 66.413 ± 0.067 Ma (Sprain
et al., 2019).
Figure 7. Temperature evolution through the Late Maastrichtian Warming Event. Comparison of new temperature data to the compilation
of late Maastrichtian temperatures by Hull et al. (2020). Grey and light blue vertical bars indicate onset of Deccan volcanism and the KPB,
as in Fig. 5. Vertical error bars on horizon averages represent 1 SD of all temperatures from a given horizon. Horizons with no error had
only one sample. Horizontal error bars represent uncertainty in the age model of Vellekoop et al. (2022). The brown dashed line indicates
a temperature profile calculated from δ 18 Ocarb assuming a fixed δ 18 Osw value of −1 ‰ VSMOW and, similarly to the Hull et al. (2020)
composite record, shows peak temperatures delayed relative to the 147 -based temperature peak.

Interestingly, a δ 18 Ocarb -based temperature reconstruction to reflect marginal-marine to restricted-marine conditions un-
assuming constant δ 18 Osw (red dashed line, Fig. 7) incor- der high hydrodynamic conditions (Brinkhuis and Schiøler,
rectly shows a slightly later peak in temperature. The true 1996; Schiøler et al., 1997), observed fluctuations are inter-
peak, as indicated by the 147 -derived record, is hidden in preted as local sea level changes. Alternatively, environmen-
δ 18 Ocarb due to simultaneous increases in temperature and tal factors other than sea level might have driven the observed
enrichment in δ 18 Osw (Figs. 5, 6, and S4). The timing of high abundances of Paralecaniella in the type Maastrichtian,
warming in our 147 -derived record appears to predate that such as biological responses to local changes in hydrody-
of the composite record of Hull et al. (2020) as well (Fig. 7), namic conditions and salinity, which would also be reflected
although uncertainty in age in both records may not sustain in the δ 18 Osw record presented here. Nevertheless, seagrass
this relationship. If age models hold as shown, the later tim- and foraminifera indicate that water depths remained less
ing seen in the composite record may reflect the same issue than 20 m at all times and often shallower (Hart et al., 2016).
of a hidden LMWE peak, as many of the records included Although this local sea level record derived from paly-
are based on foraminiferal δ 18 Ocarb and include an assump- nological abundances may have some issues at a fine scale
tion of constant δ 18 Osw . Another 147 -derived temperature related to minimal consideration of sedimentology and geo-
record across this period from Seymour Island, Antarctica chemistry (Vellekoop et al., 2022), we see a notable correla-
(Petersen et al., 2016a), also shows simultaneous increases in tion between it and many of our isotopic parameters (Fig. 6).
temperature and δ 18 Osw as well as a delayed apparent warm- Peak δ 18 Osw values, high temperatures, and, to a lesser de-
ing in a δ 18 Ocarb -based temperature reconstruction, although gree, higher δ 13 Ccarb values correlate with periods of maxi-
the absolute timing of the LMWE in that record is apparently mum sea level, while lower δ 18 Osw values, cooler tempera-
later (∼ 66.24–66.2 Ma). Faulty assumptions of constant sed- tures, and lower δ 13 Ccarb values correspond to periods of rel-
imentation rate between magnetostratigraphic datums could atively lower sea level (Fig. 6). In particular, sea level is inter-
potentially explain the difference in absolute timing. preted to rise from the Lava Horizon up to the Laumont Hori-
Following on the assumed identification of the LMWE zon, peaking around the base of the Nekum Member and then
in the basal Nekum Member, our temperature record in- declining again by the Kanne Horizon (Fig. 6e). This aligns
dicates that seawater temperatures in this region returned with the temperature and δ 18 Osw peak in the basal Nekum
to baseline levels by the middle Nekum Member within highlighted above (Fig. 6c and d), coinciding with the on-
∼ 100 kyr (Fig. 5). This is consistent with the timing of sili- set of Deccan volcanism. A second interval of peak sea level
cate weathering feedbacks, as previously suggested (Caldiera occurs during the Gronsveld Member (Fig. 6e), which also
and Rampino, 1990; Dessert et al., 2001; Petersen et al., displays temperatures and δ 18 Osw values slightly higher than
2016a; Tobin et al., 2017; Hull et al., 2020), further support- preceding and following horizons (Fig. 6c and d). δ 13 Ccarb
ing the volcanically derived CO2 as the source of the initial values also rise into the Gronsveld Member and again rise
warming event. Others have suggested that the emplacement from the Lava to Laumont horizons (Fig. 6b).
of large igneous provinces can actually be a long-term CO2 We propose that the observed correlation between envi-
sink and lead to longer-term global climate cooling (Schaller ronmental conditions and reconstructed local sea level is
et al., 2012; Johansson et al., 2018). This is possibly visi- the result of fluctuating ocean circulation patterns. Paleo-
ble in the slight cooling between the middle Nekum Member geographic reconstructions suggest the presence of a sea-
and Meerssen Member (Fig. 5), although the uncertainty in way connecting the Arctic Ocean to this study site during
the Meerssen Member is large. the late Maastrichtian (Engelke et al., 2017; Scotese, 2021;
Fig. 1a). Modeling studies show that increased isolation of
5.2 Ocean circulation, sea level, and/or precipitation
the Arctic made it fresher during the Maastrichtian (Ladant
changes
et al., 2020), suggesting that this boreal water mass was
both cooler in temperature and depleted in δ 18 Osw . As a re-
Paleogeographic reconstructions (Engelke et al., 2017; sult, the Maastrichtian type region would have been under
Scotese, 2021; Fig. 1a) and the presence of seagrass fossils the influence of relatively cool, depleted water masses. Our
(Hart et al., 2016; van der Ham et al., 2017) indicate that study site was situated on a shallow carbonate platform sur-
during the Campanian–Maastrichtian, the study site was lo- rounded by small landmasses, the exact size and position of
cated on a large, shallowly submerged carbonate platform. which are difficult to define (Engelke et al., 2017; Scotese,
Over the Late Cretaceous, eustatic sea level fluctuated multi- 2021; Fig. 1a). During periods of low relative sea level, the
ple times, including one sea level transgression (KMa5) in landmasses of the Rhenohercynian Zone (Rhenish Massif,
the latest Maastrichtian at 66.8 Ma (Haq, 2014). Sea level London–Brabant Massif) would have blocked southern cur-
based on the relative abundance of the palynomorph Parale- rents originating in warmer regions, which would likely be
caniella in ENCI quarry sediments – the same section as our saltier and have higher δ 18 Osw values due to evaporative en-
fossils – indicates local sea level fluctuations that differ from richment (e.g., the semi-restricted Western Tethys; Fig. 1a).
the global eustatic curve (Schiøler et al., 1997; Fig. 6e). Since However, as large parts of these landmasses were relatively
high relative abundances of this acritarch taxon are suggested flat and low-lying during the Upper Cretaceous (Vanden-

berghe et al., 2014), relatively minor increases in sea level substantial landmass, precipitation and precipitation-derived
could potentially breach sills or island barriers, changing runoff are depleted in δ 18 O by 4 ‰–5 ‰ VSMOW relative
both the position of ocean currents and the dominant wa- to seawater (Zhang et al., 2021). Additionally, surface runoff
ter mass in a given area. Indeed, lithostratigraphic studies may contain dissolved terrestrial organic matter depleted in
in the region have shown that during various phases in the δ 13 C, which could explain the correlation between δ 18 Osw
Campanian–Maastrichtian, the structural highs of the nearby and δ 13 C through our study period. However, it is difficult
London–Brabant Massif were fully submerged (e.g., Dusar to explain how local freshwater in a subtropical environment
and Lagrou, 2007). As a result, during sea level lows, the surrounded by low-lying landmasses could be substantially
Maastrichtian type region would have been predominantly colder than local seawater, so we favor the variable water
under the influence of colder, depleted water masses sourc- mass hypothesis over a local freshwater contribution.
ing from the north, while during sea level highs, the influ- Yet another possible interpretation of the covariation be-
ence of warmer, more enriched water masses sourcing from tween fluctuating temperature and δ 18 Osw trends is via
the south likely increased (Fig. 1a). climate-related reorganization of atmospheric circulation,
Similar interactions between cooler and warmer water moving broad precipitation bands north and south over the
masses have been called upon to explain unexpected distri- study region. An enhanced hydrological cycle has been hy-
butions in warm-water ammonite fauna in the Coniacian– pothesized for the latest Cretaceous (e.g., Woelders et al.,
Santonian of Europe (Remin et al., 2016) and remarkably 2017) due to increased zonal surface winds and ∼ 20 %
southerly occurrences of cold-water belemnites in late Ceno- stronger Hadley circulation in the Northern Hemisphere
manian France (Gale and Christensen, 1996). In this case, the (Bush and Philander, 1997). If, during warmer climates, pre-
authors suggest that warm waters flowing east to west from cipitation bands moved such that the Maastrichtian type re-
the proto-Tethys competed with cooler waters flowing south- gion experienced less rainfall, δ 18 O enrichment of surface
ward between Greenland and Scandinavia (see their Fig. 10). waters would coincide with warmer temperatures and vice
We suggest that a similar mechanism applied to late Maas- versa. This is difficult to assess, even in global climate model
trichtian paleogeography and water masses (Fig. 1a). Ostra- simulations, as model proficiency in reconstructing paleo-
cod assemblage changes indicate an increasing contribution precipitation regimes is difficult to validate and/or calibrate.
of warm Mediterranean (Tethyan) waters in the latest Maas- The rough correlation with δ 13 C is also difficult to explain in
trichtian (Bless, 1988). Faunal abundance differences along this scenario.
the US East Coast indicate the presence of an ancient Gulf
Stream-like current as far back in time as the Maastrichtian
(Watkins and Self-Trail, 2005), which would have warmed 6 Conclusions
the Maastricht region above its expected temperature for lati-
tude when bathed in water from the south, as the modern-day This study presents a new clumped-isotope-based paleotem-
Gulf Stream does for western Europe. perature time series from the latest Cretaceous in northwest-
Fluctuations in δ 13 C appear to correlate with changes in ern Europe. The 147 analyses from the Maastrichtian type
temperature and δ 18 Osw (Fig. 6). In this hypothetical model section and surrounding outcrops reveal a mean temperature
of the alternating influence of southern and northern source of 20.4±3.8 ◦ C and an average δ 18 Osw of +0.2±0.8 ‰ VS-
waters, this could indicate different δ 13 C of dissolved inor- MOW, consistent with a subtropical shallow marine environ-
ganic carbon (δ 13 CDIC ) in the two water masses. The exact ment. This average temperature aligns well with other pale-
values of δ 13 CDIC in either water mass are difficult to de- othermometry studies from similar paleolatitudes and with
fine because of overprinting by δ 13 C vital effects in some modeled Maastrichtian paleotemperatures. Increasing 147 -
taxa (Neithia sp., Entollium sp.; Figs. S2 and S3). Looking at based temperatures at ∼ 66.4 Ma, in conjunction with other
the most common taxa (Acutostrea), variations on the order indicators, are interpreted to reflect a regional manifesta-
of 0.5 ‰VPDB – or possibly as much as 1 ‰ – are possi- tion of the global LMWE and directly link pre-KPB warm-
ble between horizons. This is similar to variability seen in ing in northwestern Europe to CO2 emissions from Deccan
the modern Atlantic Ocean (Eide et al., 2017) and is there- Traps volcanism. Simultaneous increases in temperature and
fore plausibly driven by alternating water masses of differing δ 18 Osw during the LMWE mask the timing of warming in
δ 13 CDIC . In this scenario, the colder, northern (Arctic) water δ 18 Ocarb -based temperature reconstructions, which may ex-
mass would be more depleted in δ 13 CDIC than the southern plain a more delayed apparent timing of the LMWE in other
source water. studies. Seen now in two spatially distant records, this de-
This dataset is not capable of distinguishing between the serves further exploration and emphasizes the need to use
above preferred hypothesis and a temporally variable input of paleothermometry tools that take into account variations in
isotopically depleted freshwater. During times of lower sea δ 18 Osw .
level, partial basin restriction may have occurred, allowing Covariations in temperature, δ 18 Osw , and (to a lesser de-
local freshwater inputs to lower δ 18 Osw values. In maritime gree) δ 13 C from horizon to horizon may reflect the vary-
regions such as modern Bermuda, despite lack of elevation or ing influences of different water masses at different times.

We hypothesize that colder waters with lower δ 18 Osw and Acknowledgements. We thank Ashling Neary (SCIPP lab), An-
δ 13 CDIC may have originated from the Arctic and reached the gela Dial (MEAL), Lora Wingate, and Kacey Lohmann (SIL) for
study site via a passageway between Greenland and Scandi- assisting in sample preparation and measurements at the Univer-
navia (Fig. 1a) during periods of lower sea level when emer- sity of Michigan. The GSG sample was collected as part of a field
gent landmasses to the south acted as barriers. Warmer wa- trip related to the International Conference on Paleoceanography in
2014, and we thank the organizers and co-participants.
ters with higher δ 18 Osw and δ 13 CDIC may have been sourced
from a proto-Tethys to the southeast or proto-Atlantic to the
southwest (Fig. 1a) and flushed the study region during inter-
Financial support. This research has been supported by the Geo-
vals of higher sea level.
logical Society of America (grant no. 12956-20) and the University
Overall, the fossils of the Maastricht Formation are effec- of Michigan Earth and Environmental Science Department’s Turner
tive paleotemperature archives that reveal linkages between Research Award to Heidi E. O’Hora, the Research Foundation Flan-
local sea level changes, Deccan Traps volcanism, and ocean ders (grant no. 12Z6621N) to J. Vellekoop, and the National Sci-
temperature on the European continent in the lead-up to the ence Foundation Ocean Sciences Postdoctoral Research Fellowship
end Cretaceous. (NSF grant no. 1420902) to Sierra V. Petersen.
Data availability. All data presented in this paper are available in Review statement. This paper was edited by Denis-
the EarthChem database at https://doi.org/10.26022/IEDA/112046 Didier Rousseau and reviewed by two anonymous referees.
(O’Hora et al., 2021). Atmospheric CO2 estimates for the Late Cre-
taceous can be found in Zenodo (Hoenisch, 2021).
Further data are available in the Supplement as follows.
References
– Fig. S1: oxygen isotope values of shell specimens and bulk
carbonate Anderson, N. T., Kelson, J. R., Kele, S., Daëron, M., Bonifa-
– Fig. S2: carbon isotope values of shell specimens and bulk car- cie, M., Horita, J., Mackey, T. J., John, C. M., Kluge, T.,
bonate Petschnig, P., Jost, A. B., Huntington, K. W., Bernasconi,
– Fig. S3: carbon isotopic composition vs. oxygen isotopic com- S. M., and Bergmann, K. D.: A unified clumped isotope
position in well-preserved shells and bulk matrix thermometer calibration (0.5–1000 ◦ C) using carbonate-based
– Fig. S4: reconstructed seawater oxygen isotopic composition standardization, Geophys. Res. Lett., 48, e2020GL092069,
through time https://doi.org/10.1029/2020GL092069, 2021.
Barnet, J. S. K., Littler, K., Westerhold, T., Kroon, D., Leng, M.
– Fig. S5: covariation of oxygen isotopic composition of carbon-
J., Bailey, I., Röhl, U., and Zachos, J. C.: A High-fidelity ben-
ate, water, and temperature
thic stable isotope record of Late Cretaceous–Early Eocene cli-
– Table S1: published paleotemperature data for other studies re- mate change and carbon-cycling, Paleoceanogr. Paleoclimatol.,
constructing Maastrichtian marine temperatures 34, 672–691, https://doi.org/10.1029/2019PA003556, 2019.
– Table S2: age model of Vellekoop et al. (2022) Bernasconi, S. M., Mueller, I., Bergmann, K. D., Breitenbach,
– Table S3: preservation data for all samples, including SEM S. F. M., Fernandez, A., Hodell, D. A., Jaggi, M., Meckler,
preservation determinations and trace element concentrations; A.N., Millan, I., and Ziegler, M.,: Reducing uncertainties in car-
– Table S4: summary of in-house carbonate standards bonate clumped isotope analysis through consistent carbonate-
based standardization, Geochem. Geophy. Geosy., 19, 2895–
2914, https://doi.org/10.1029/2017GC007385, 2018.
Bernasconi, S. M., Daëron, M., Bergmann, K. D., Bonifacie,
Supplement. The supplement related to this article is available
M., Meckler, A. N., Affek, H. P., Anderson, N., Bajnai, D.,
online at: https://doi.org/10.5194/cp-18-1963-2022-supplement.
Barkan, E., Beverly, E., and Blamart, D.: InterCarb: A com-
munity effort to improve interlaboratory standardization of
the carbonate clumped isotope thermometer using carbonate
Author contributions. SVP, JV, and HEO’H designed the study. standards, Geochem. Geophy. Geosy., 22, e2020GC009588,
SVP and JV collected the samples. HEO’H, SVP, MMJ, and SRS https://doi.org/10.1029/2020GC009588, 2021.
measured and analyzed the data. HEO’H and SVP prepared the pa- Bless, M. J. M.: Possible causes for the change in ostracod as-
per, with contributions from all co-authors. semblages at the Maastrichtian-Palaeocene boundary in southern
Limburg, The Netherlands, Meded. Werkgr. Tert. Kwart. Geol.,
25, 197–211, 1988.
Competing interests. The contact author has declared that none Bowman, V. C., Riding, J. B., Francis, J. E., Crame, J. A.,
of the authors has any competing interests. and Hannah, M. J.: The taxonomy and palaeobiogeography
of small chorate dinoflagellate cysts from the Late Creta-
ceous to Quaternary of Antarctica, Palynology, 37, 151–169,
Disclaimer. Publisher’s note: Copernicus Publications remains https://doi.org/10.1080/01916122.2012.750898, 2013.
neutral with regard to jurisdictional claims in published maps and Brand, U. and Morrison, J. O.: Biogeochemistry of fossil marine
institutional affiliations. invertebrates, Geosci. Can., 14, 85–107, 1987.

Brand, W. A., Assonov, S. S., and Coplen, T. B.: Correction for the on 13 C–18 O bond ordering in bivalve mollusks, Biogeosciences,
17 O interference in d13C measurements when analyzing CO 10, 4591–4606, https://doi.org/10.5194/bg-10-4591-2013, 2013.
2
with stable isotope mass spectrometry (IUPAC Technical Re- Eide, M., Olsen, A., Ninnemann, U. S., and Johannessen,
port), Pure Appl. Chem., 82, 1719–1733, 2010. T.: A global ocean climatology of preindustrial and mod-
Brinkhuis, H. and Schiøler, P.: Palynology of the Geulhemmer- ern ocean δ 13 C, Global Biogeochem. Cy., 31, 515–534,
berg Cretaceous/Tertiary boundary section (Limburg, SE Nether- https://doi.org/10.1002/2016GB005472, 2017.
lands), Geol. Mijnbouw, 75, 193–213, 1996. Eiler, J. M.: Paleoclimate reconstruction using carbonate clumped
Bush, A. B. G. and Philander, S. G. H.: The Late Creta- isotope thermometry, Quaternary Sci. Rev., 30, 3575–3588,
ceous: Simulation with a coupled atmosphere-ocean gen- https://doi.org/10.1016/j.quascirev.2011.09.001, 2011.
eral circulation model, Paleoceanography, 12, 495–516, Engelke, J., Linnert, C., Mutterlose, J., and Wilmsen, M.:
https://doi.org/10.1029/97PA00721, 1997. Early Maastrichtian benthos of the chalk at Kronsmoor,
Caldeira, K. and Rampino, M. R.: Carbon dioxide emis- northern Germany: implications for Late Cretaceous en-
sions from Deccan volcanism and a K/T boundary vironmental change, Palaeobio. Palaeoenv., 97, 703–722,
greenhouse effect, Geophys. Res. Lett., 17, 1299–1302, https://doi.org/10.1007/s12549-017-0283-2, 2017.
https://doi.org/10.1029/GL017i009p01299, 1990. Felder, W. M.: Lithostratigrafie van het Boven-Krijt en het Dano-
Came, R. E., Brand, U., and Affek, H. P.: Clumped isotope signa- Montien in Zuid-Limburg en het aangrenzende gebied, in:
tures in modern brachiopod carbonate, Chem. Geol., 377, 20–30, Toelichting bij geologische overzichtskaarten van Nederland,
https://doi.org/10.1016/j.chemgeo.2014.04.004, 2014. edited by: Zafwijn, W. H. and van Staalduinen, C. J., 63–72,
Clyde, W. C., Ramezani, J., Johnson, K. R., Bowring, S. A., and Haarlem (Rijks Geologische Dienst), 1975.
Jones, M. M.: Direct high-precision U–Pb geochronology of Felder, W. M. and Bosch, P. W.: Geologie van de St. Pietersberg bij
the end-Cretaceous extinction and calibration of Paleocene as- Maastricht, Grondboor & Hamer, 52, 53–63, 1998.
tronomical timescales, Earth Planet. Sc. Lett., 452, 272–280, Fendley, I. M., Mittal, T., Sprain, C. J., Marvin-DiPasquale, M.,
https://doi.org/10.1016/j.epsl.2016.07.041, 2016. Tobin, T. S., and Renne, P. R.: Constraints on the volume and
de Winter, N. J., Vellekoop, J., Vorsselmans, R., Golreihan, A., rate of Deccan Traps flood basalt eruptions using a combi-
Soete, J., Petersen, S. V., Meyer, K. W., Casadio, S., Speijer, R. P., nation of high-resolution terrestrial mercury records and geo-
and Claeys, P.: An assessment of latest Cretaceous Pycnodonte chemical box models, Earth Planet. Sc. Lett., 524, 115721,
vesicularis (Lamarck, 1806) shells as records for palaeosea- https://doi.org/10.1016/j.epsl.2019.115721, 2019.
sonality: a multi-proxy investigation, Clim. Past, 14, 725–749, Font, E., Adatte, T., Sial, A. N., de Lacerda, L. D., Keller, G.,
https://doi.org/10.5194/cp-14-725-2018, 2018. and Punekar, J.: Mercury anomaly, Deccan volcanism, and
Defliese, W. F., Hren, M. T., and Lohmann, K. C.: the end-Cretaceous mass extinction, Geology, 44, 171–174,
Compositional and temperature effects of phospho- https://doi.org/10.1130/G37451.1, 2016.
ric acid fractionation on 147 analysis and implications Gale, A. S. and Christensen, W. K.: Occurrence of the belemnite
for discrepant calibrations, Chem. Geol., 396, 51–60, Actinocamax plenus in the in the Cenomanian of SE France and
https://doi.org/10.1016/j.chemgeo.2014.12.018, 2015. its significance, B. Geol. Soc. Denmark, 43, 68–77, 1996.
Dennis, K. J., Affek, H. P., Passey, B. H., Schrag, D. P., and Eiler, Gao, Y., Ibarra, D. E., Rugenstein, J. K. C., Chen, J., Kukla, T.,
J. M.: Defining an absolute reference frame for ‘clumped’ iso- Methner, K., Gao, Y., Huang, H., Lin, Z., and Zhang, L.: Ter-
tope studies of CO2 , Geochim. Cosmochim. Ac., 75, 7117–7131, restrial climate in mid-latitude East Asia from the latest Cre-
https://doi.org/10.1016/j.gca.2011.09.025, 2011. taceous to the earliest Paleogene: A multiproxy record from
Dennis, K. J., Cochran, J. K., Landman, N. H., and Schrag, D. P.: the Songliao Basin in northeastern China, Earth-Sci. Rev., 216,
The climate of the Late Cretaceous: New insights from the appli- 103572, https://doi.org/10.1016/j.earscirev.2021.103572, 2021.
cation of the carbonate clumped isotope thermometer to Western Golovneva, L. B.: The Maastrichtian (Late Cretaceous) climate in
Interior Seaway macrofossil, Earth Planet. Sc. Lett., 362, 51–65, the northern hemisphere, Geol. Soc. London Spec. Publ., 181,
https://doi.org/10.1016/j.epsl.2012.11.036, 2013. 43–54, https://doi.org/10.1144/GSL.SP.2000.181.01.05, 2000.
Dessert, C., Dupré, B., François, L. M., Schott, J., Gaillardet, J., Haq, B. U.: Cretaceous eustasy revisited, Global Planet. Change,
Chakrapani, G., and Bajpai, S.: Erosion of Deccan Traps deter- 113, 44–58, https://doi.org/10.1016/j.gloplacha.2013.12.007,
mined by river geochemistry: impact on the global climate and 2014.
the 87 Sr/86 Sr ratio of seawater, Earth Planet. Sc. Lett., 188, 459– Hart, M. B., FitzPatrick, M. E., and Smart, C. W.: The Cretaceous/-
474, https://doi.org/10.1016/S0012-821X(01)00317-X, 2001. Paleogene boundary: Foraminifera, sea grasses, sea level change
Dortangs, R. W., Schulp, A. S., Mulder, E. W., Jagt, J. W., Peeters, and sequence stratigraphy, Palaeogeogr. Palaeocl., 441, 420–429,
H. H., and De Graaf, D. T.: A large new mosasaur from the Up- https://doi.org/10.1016/j.palaeo.2015.06.046, 2016.
per Cretaceous of The Netherlands, Neth. J. Geosci., 81, 1–8, Henehan, M. J., Ridgwell, A., Thomas, E., Zhang, S., Alegret,
https://doi.org/10.1017/S0016774600020515, 2002. L., Schmidt, D. N., Rae, J. W., Witts, J. D., Landman, N. H.,
Dusar, M. and Lagrou, D.: Cretaceous flooding of the Brabant Mas- Greene, S. E., and Huber, B. T.: Rapid ocean acidification and
sif and the lithostratigraphic characteristics of its chalk cover in protracted Earth system recovery followed the end-Cretaceous
northern Belgium, Geol. Belg., 10, 27–38, 2007. Chicxulub impact, P. Natl. Acad. Sci. USA, 116, 22500–22504,
Eagle, R. A., Eiler, J. M., Tripati, A. K., Ries, J. B., Freitas, P. S., https://doi.org/10.1073/pnas.1905989116, 2019.
Hiebenthal, C., Wanamaker Jr., A. D., Taviani, M., Elliot, M., Henkes, G. A., Passey, B. H., Wanamaker Jr., A. D., Gross-
Marenssi, S., Nakamura, K., Ramirez, P., and Roy, K.: The in- man, E. L., Ambrose Jr., W. G., and Carroll, M. L.: Carbonate
fluence of temperature and seawater carbonate saturation state clumped isotope compositions of modern marine mollusk and

brachiopod shells, Geochim. Cosmochim. Ac., 106, 307–325, Li, L. and Keller, G.: Abrupt deep-sea warming at the end of the
https://doi.org/10.1016/j.gca.2012.12.020, 2013. Cretaceous, Geology, 26, 995–998, 1998.
Henkes, G. A., Passey, B. H., Grossman, E. L., Shenton, B. Liebau, A.: Paläobathymetrische und paläoklimatische Veränderun-
J., Pérez-Huerta, A., and Yancey, T. E.: Temperature lim- gen im Mikrofaunenbild der Maastrichter Tuffkreide, Neues
its for preservation of primary calcite clumped isotope pa- Jahrb. Geol. P. A., 157, 233–237, 1978.
leotemperatures, Geochim. Cosmochim. Ac., 139, 362–382, Luijendijk, E., Van Balen, R. T., Ter Voorde, M., and Andriessen,
https://doi.org/10.1016/j.gca.2014.04.040, 2014. P. A. M.: Reconstructing the Late Cretaceous inversion of
Herngreen, G. F. W., Schuurman, H. A. H. M., Verbeek, J. W., the Roer Valley Graben (southern Netherlands) using a new
Brinkhuis, H., Burnett, J. A., Felder, W. M., and Kedves, M.: model that integrates burial and provenance history with fission
Biostratigraphy of Cretaceous/Tertiary boundary strata in the track thermochronology, J. Geophys. Res.-Sol. Ea., 116, p. 19,
Curfs quarry, Netherlands Institute of Applied Geoscience TNO, https://doi.org/10.1029/2010JB008071, 2011.
61, 1–58, 1998. Maastricht Climate: https://en.climate-data.org/europe/
Hoenisch, B.: Paleo-CO2 data archive (Version 1), Zenodo [data the-netherlands/limburg/maastricht-893/#temperature-graph,
set], https://doi.org/10.5281/zenodo.5777278, 2021. last access: 14 June 2021.
Hull, P. M., Bornemann, A., Penman, D. E., Henehan, M. J., Mackey, T. J., Jost, A. B., Creveling, J. R., and Bergmann, K.
Norris, R. D., Wilson, P. A., Blum, P., Alegret, L., Baten- D.: A Decrease to Low Carbonate Clumped Isotope tempera-
burg, S. J., and Bown, P. R.: On impact and volcanism across tures in Cryogenian Strata, AGU Advances, 1, e2019AV000159,
the Cretaceous-Paleogene boundary, Science, 367, 266–272, https://doi.org/10.1029/2019AV000159, 2020.
https://doi.org/10.1126/science.aay5055, 2020. Meckler, A. N., Ziegler, M., Millaìn, M. I., Breitenbach, S. F. M.,
Huyghe, D., Daëron, M., de Rafelis, M., Blamart, D., Sébilo, and Bernasconi, S. M.: Long-term performance of the Kiel car-
M., Paulet, Y. M., and Lartaud, F.: Clumped isotopes in mod- bonate device with a new correction scheme for clumped iso-
ern marine bivalves, Geochim. Cosmochim, Ac., 316, 41–58, tope measurements, Rapid Commun. Mass Sp., 28, 1705–1715,
https://doi.org/10.1016/j.gca.2021.09.019, 2022. https://doi.org/10.1002/rcm.6949, 2014.
Jagt, J. W., Donovan, S. K., Fraaije, R., Mulder, E. W., Nieuwen- Meyer, K. W., Petersen, S. V., Lohmann, K. C., and Winkel-
huis, E., Stroucken, J., van Bakel, B., and van Knippenberg, P.: stern, I. Z.: Climate of the Late Cretaceous North Ameri-
Remarkable preservation of selected latest Cretaceous macrofos- can Gulf and Atlantic Coasts, Cretaceous Res., 89, 160–173,
sils from the Maastrichtian type area (the Netherlands, Belgium), https://doi.org/10.1016/j.cretres.2018.03.017, 2018.
Foss. Rec., 4, 75–78, 2016. Meyer, K. W., Petersen, S. V., Lohmann, K. C., Blum, J. D.,
Jagt, J. W. M. and Jagt-Yazykova, E. A.: Stratigraphy of the type Washburn, S. J., Johnson, M. W., Gleason, J. D., Kurz,
Maastrichtian – a synthesis, Scr. Geol., 8, 5–32, 2012. A. Y., and Winkelstern, I. Z.: Biogenic carbonate mercury
Johansson, L., Zahirovic, S., and Müller, R. D.: The interplay be- and marine temperature records reveal global influence of
tween the eruption and weathering of large igneous provinces Late Cretaceous Deccan Traps, Nat. Commun., 10, 1–8,
and the deep-time carbon cycle, Geophys. Res. Lett., 45, 5380– https://doi.org/10.1038/s41467-019-13366-0, 2019.
5389, https://doi.org/10.1029/2017GL076691, 2018. Miller, K. G., Wright, J. D., and Browning, J. V.: Visions of
Keutgen, N.: A bioclast-based astronomical timescale for ice sheets in a greenhouse world, Mar. Geol., 217, 215–231,
the Maastrichtian in the type area (southeast Nether- https://doi.org/10.1016/j.margeo.2005.02.007, 2005.
lands, northeast Belgium) and stratigraphic implications: Molina, E., Alegret, L., Arenillas, I., Arz, J. A., Gallala, N., Hard-
the legacy of PJ Felder, Neth. J. Geosci., 97, 229–260, enbol, J., Salis, K. von, Steurbaut, E., Vandenberghe, N., and
https://doi.org/10.1017/njg.2018.15, 2018. Zaghbib-Turki, D.: The global boundary stratotype section and
Kim, S. T. and O’Neil, J. R.: Equilibrium and nonequilibrium point for the base of the Danian stage (Paleocene, Paleogene,
oxygen isotope effects in synthetic carbonates, Geochim. Cos- “Tertiary”, Cenozoic) at El Kef, Tunisia-original definition and
mochim. Ac., 61, 3461–3475, https://doi.org/10.1016/S0016- revision, Episodes, 29, 263–273, 2006.
7037(97)00169-5, 1997. Möller, P. and Kubanek, F.: Role of magnesium in nucleation pro-
Ladant, J. B. and Donnadieu, Y.: Palaeogeographic regulation of cesses of calcite, aragonite and dolomite, Neues Jb. Miner. Abh.,
glacial events during the Cretaceous supergreenhouse, Nat. Com- 126, 199–220, 1976.
mun., 7, 1–9, https://doi.org/10.1038/ncomms12771, 2016. Nava, A. H., Black, B. A., Gibson, S. A., Bodnar, R. J., Renne, P.
Ladant, J.-B., Poulsen, C. J., Fluteau, F., Tabor, C. R., MacLeod, K. R., and Vanderkluysen, L.: Reconciling early Deccan Traps CO2
G., Martin, E. E., Haynes, S. J., and Rostami, M. A.: Paleogeo- outgassing and pre-KPB global climate, P. Natl. Acad. Sci. USA,
graphic controls on the evolution of Late Cretaceous ocean circu- 118, e2007797118, , https://doi.org/10.1073/pnas.2007797118,
lation, Clim. Past, 16, 973–1006, https://doi.org/10.5194/cp-16- 2021.
973-2020, 2020. O’Hora, H. E., Petersen, S. V., Vellekoop, J., Jones, M. M.,
Land, L. S.: Diagenesis of skeletal carbonates, J. Sediment and Scholz, S. R.: Clumped isotope data from Maastrichtian-
Res., 37, 914–930, https://doi.org/10.1306/74D717D5-2B21- aged bivalves from the Maastrichtian type region, Ver-
11D7-8648000102C1865D, 1967. sion 1.0, Interdisciplinary Earth Data Alliance (IEDA) [data set],
Leloux, J.: Numerical distribution of Santonian to Da- https://doi.org/10.26022/IEDA/112046, 2021.
nian corals (Scleractinia, Octocorallia) of southern Lim- Percival, L. M., Jenkyns, H. C., Mather, T. A., Dickson, A.
burg, the Netherlands, Geol. Mijnbouw, 78, 191–195, J., Batenburg, S. J., Ruhl, M., Hesselbo, S. P., Barclay,
https://doi.org/10.1023/A:1003743301625, 1999. R., Jarvis, I., and Robinson, S. A.: Does large igneous
province volcanism always perturb the mercury cycle? Com-

paring the records of Oceanic Anoxic Event 2 and the end- the end-Cretaceous mass extinction, Science, 363, 862–866,
Cretaceous to other Mesozoic events, Am. J. Sci., 318, 799–860, https://doi.org/10.1126/science.aau2422, 2019.
https://doi.org/10.2475/08.2018.01, 2018. Schulte, P., Alegret, L., Arenillas, I., Arz, J. A., Barton, P. J., Bown,
Petersen, S. V., Dutton, A., and Lohmann, K. C.: End-Cretaceous P. R., Bralower, T. J., Christeson, G. L., Claeys, P., and Cock-
extinction in Antarctica linked to both Deccan volcanism and ell, C. S.: The Chicxulub asteroid impact and mass extinction at
meteorite impact via climate change, Nat. Commun., 7, 1–9, the Cretaceous-Paleogene boundary, Science, 327, 1214–1218,
https://doi.org/10.1038/ncomms12079, 2016a. https://doi.org/10.1126/science.1177265, 2010.
Petersen, S. V., Tabor, C. R., Lohmann, K. C., Poulsen, C. J., Scotese, C. R.: An Atlas of Phanerozoic Paleogeographic Maps:
Meyer, K. W., Carpenter, S. J., Erickson, J. M., Matsunaga, K. The Seas Come In and the Seas Go Out, Annu. Rev. Earth Pl.
K., Smith, S. Y., and Sheldon, N. D.: Temperature and salinity of Sc., 49, 669–718, https://doi.org/10.1146/annurev-earth-081320-
the Late Cretaceous western interior seaway, Geology, 44, 903– 064052, 2021.
906, https://doi.org/10.1130/G38311.1, 2016b. Sial, A. N., Chen, J., Lacerda, L. D., Frei, R., Tewari, V.
Petersen, S. V., Winkelstern, I. Z., Lohmann, K. C., and Meyer, C., Pandit, M. K., Gaucher, C., Ferreira, V. P., Cirilli, S.,
K. W.: The effects of Porapak™ trap temperature on δ 18 O, and Peralta, S.: Mercury enrichment and Hg isotopes in
δ 13 C, and 147 values in preparing samples for clumped Cretaceous–Paleogene boundary successions: Links to volcan-
isotope analysis, Rapid Commun. Mass Sp., 30, 199–208, ism and palaeoenvironmental impacts, Cretaceous Res., 66, 60–
https://doi.org/10.1002/rcm.7438, 2016c. 81, https://doi.org/10.1016/j.cretres.2016.05.006, 2016.
Petersen, S. V., Defliese, W. F., Saenger, C., Daëron, M., Hunt- Smit, J. and Brinkhuis, H.: The Geulhemmerberg Cretaceous/Ter-
ington, K. W., John, C. M., Kelson, J. R., Bernasconi, S. M., tiary boundary section (Maastrichtian type area, SE Nether-
Colman, A. S., Kluge, T., Olack, G. A., Schauer, A. J., Baj- lands); summary of results and a scenario of events, Geol. Mi-
nai, D., Bonifacie, M., Breitenbach, S. F. M., Fiebig, J., Fer- jnbouw, 75, 283–293, 1996.
nandez, A. B., Henkes, G. A., Hodell, D., Katz, A., Kele, S., Smit, J. and Zachariasse, W.J.: Planktic foraminifera in the Creta-
Lohmann, K.C., Passey, B. H., Peral, M. Y., Petrizzo, D. A., ceous/Tertiary boundary clays of the Geulhemmerberg (Nether-
Rosenheim, B. E., Tripati, A., Venturelli, R., Young, E. D., and lands), Geol. Mijnbouw, 75, 187–191, 1996.
Winkelstern, I. Z.: Effects of improved 17 O correction on inter- Sprain, C. J., Renne, P. R., Clemens, W. A., and Wilson, G.
laboratory agreement in clumped isotope calibrations, estimates P.: Calibration of chron C29r: New high-precision geochrono-
of mineral-specific offsets, and temperature dependence of acid logic and paleomagnetic constraints from the Hell Creek
digestion fractionation, Geochem. Geophy. Geosy., 20, 3495– region, Montana, Geol. Soc. Am. Bull., 130, 1615–1644,
3519, https://doi.org/10.1029/2018GC008127, 2019. https://doi.org/10.1130/B31890.1, 2018.
Pucéat, E., Lécuyer, C., Donnadieu, Y., Naveau, P., Cap- Sprain, C. J., Renne, P. R., Vanderkluysen, L., Pande, K., Self, S.,
petta, H., Ramstein, G., Huber, B. T., and Kriwet, J.: and Mittal, T.: The eruptive tempo of Deccan volcanism in rela-
Fish tooth δ 18 O revising Late Cretaceous meridional upper tion to the Cretaceous-Paleogene boundary, Science, 363, 866–
ocean water temperature gradients, Geology, 35, 107–110, 870, https://doi.org/10.1126/science.aav1446, 2019.
https://doi.org/10.1130/G23103A.1, 2007. Stott, L. D. and Kennett, J. P.: The paleoceanographic and paleo-
Ravizza, G. and Peucker-Ehrenbrink, B.: Chemostrati- climatic signature of the Cretaceous/Paleogene boundary in the
graphic evidence of Deccan volcanism from the ma- Antarctic: stable isotopic results from ODP Leg 113, in: Proceed-
rine osmium isotope record, Science, 302, 1392–1395, ings of the Ocean Drilling Program, Scientific Results, College
https://doi.org/10.1126/science.1089209, 2003. Station, Texas, USA (Ocean Drilling Program), 113, 829–848,
Remin, Z., Gruszczyński, M., and Marshall, J. D.: Changes 1990.
in paleo-circulation and the distribution of ammonite Tabor, C. R., Poulsen, C. J., Lunt, D. J., Rosenbloom, N.
faunas at the Coniacian–Santonian transition in central A., Otto-Bliesner, B. L., Markwick, P. J., Brady, E. C.,
Poland and western Ukraine, Acta Geol. Pol., 66, 107–124, Farnsworth, A., and Feng, R.: The cause of Late Cretaceous cool-
https://doi.org/10.1515/agp-2016-0006, 2016. ing: A multimodel-proxy comparison, Geology, 44, 963–966,
Roep, T. B. and Smit, J.: Sedimentological aspects of the K/T https://doi.org/10.1130/G38363.1, 2016.
boundary at Geulhemmerberg, Zuid Limburg, the Netherlands, Tagliavento, M., John, C. M., and Stemmerik, L.: Tropi-
Geol. Mijnbouw, 75, 119–131, 1996. cal temperature in the Maastrichtian Danish Basin: Data
Schaller, M. F., Wright, J. D., Kent, D. V., and Olsen, P. E.: from coccolith 147 and δ 18 O, Geology, 47, 1074–1078,
Rapid emplacement of the Central Atlantic Magmatic Province https://doi.org/10.1130/G46671.1, 2019.
as a net sink for CO2 , Earth Planet. Sc. Lett., 323, 27–39, Tobin, T. S., Ward, P. D., Steig, E. J., Olivero, E. B., Hilburn, I. A.,
https://doi.org/10.1016/j.epsl.2011.12.028, 2012. Mitchell, R. N., Diamond, M. R., Raub, T. D., and Kirschvink,
Schiøler, P., Brinkhuis, H., Roncaglia, L., and Wilson, G. J. L.: Extinction patterns, δ 18 O trends, and magnetostratigra-
J.: Dinoflagellate biostratigraphy and sequence stratigraphy from a southern high-latitude Cretaceous–Paleogene section:
phy of the type Maastrichtian (Upper Cretaceous), ENCI Links with Deccan volcanism, Palaeogeogr. Palaeocl., 350, 180–
Quarry, The Netherlands, Mar. Micropaleontol., 31, 65–95, 188, https://doi.org/10.1016/j.palaeo.2012.06.029, 2012.
https://doi.org/10.1016/S0377-8398(96)00058-8, 1997. Tobin, T. S., Wilson, G. P., Eiler, J. M., and Hartman, J. H.: Environ-
Schoene, B., Eddy, M. P., Samperton, K. M., Keller, C. mental change across a terrestrial Cretaceous-Paleogene bound-
B., Keller, G., Adatte, T., and Khadri, S. F.: U–Pb con- ary section in eastern Montana, USA, constrained by carbon-
straints on pulsed eruption of the Deccan Traps across ate clumped isotope paleothermometry, Geology, 42, 351–354,
https://doi.org/10.1130/G35262.1, 2014.

Tobin, T. S., Bitz, C. M., and Archer, D.: Modeling cli- Vellekoop, J., Kaskes, P., Sinnesael, M., Dehais, T., Huygh, J.,
matic effects of carbon dioxide emissions from Dec- Jagt, J., Speijer, R. P., and Claeys, P.: A new age model
can Traps volcanic eruptions around the Cretaceous– and chemostratigraphic framework for the Maastrichtian type
Paleogene boundary, Palaeogeogr. Palaeocl., 478, 139–148, area (southeastern Netherlands, northeastern Belgium), Newsl.
https://doi.org/10.1016/j.palaeo.2016.05.028, 2017. Stratigr., 55, 4, 479–501, https://doi.org/10.1127/nos/2022/0703,
Ullmann, C. V., Campbell, H. J., Frei, R., Hesselbo, S. P., 2022
Pogge von Strandmann, P. A. E., and Korte, C.: Partial di- Voigt, S., Wilmsen, M., Mortimore, R. N., and Voigt, T.: Ceno-
agenetic overprint of Late Jurassic belemnites from New manian palaeotemperatures derived from the oxygen isotopic
Zealand: Implications for the preservation potential of δ 7 Li val- composition of brachiopods and belemnites: evaluation of Creta-
ues in calcite fossils, Geochim. Cosmochim. Ac., 120, 80–96, ceous palaeotemperature proxies, Int. J. Earth Sci., 92, 285–299,
https://doi.org/10.1016/j.gca.2013.06.029, 2013. https://doi.org/10.1007/s00531-003-0315-1, 2003.
Vandenberghe, N., Craen, M. D., and Beerten, K.: Geological Vonhof, H. B. and Smit, J.: Strontium-isotope stratigraphy of the
framework of the Campine Basin, Studiecentrum voor Kernen- type Maastrichtian and the Cretaceous/Tertiary boundary in the
ergie Centre d’étude de l’énergie Nucléaire, 112, 2014. Maastricht area (SE Netherlands), Geol. Mijnbouw, 75, 275–282,
van der Ham, R. W. J. M., van Konijnenburg-van Cittert, J. 1996.
H. A., Jagt, J. W. M., Indeherberge, L., Meuris, R., Deck- Watkins, D. K. and Self-Trail, J. M.: Calcareous nannofos-
ers, M. J. M., Renkens, S., and Laffineur, J.: Seagrass stems sil evidence for the existence of the Gulf Stream dur-
with attached roots from the type area of the Maastrichtian ing the late Maastrichtian, Paleoceanography, 20, PA3006,
Stage (NE Belgium, SE Netherlands): Morphology, anatomy, https://doi.org/10.1029/2004PA001121, 2005.
and ecological aspects, Rev. Palaeobot. Palyno., 241, 49–69, Wilf, P., Johnson, K. R., and Huber, B. T.: Correlated terrestrial and
https://doi.org/10.1016/j.revpalbo.2017.02.001, 2017. marine evidence for global climate changes before mass extinc-
van Hinsbergen, D. J., de Groot, L. V., van Schaik, S. J., Spakman, tion at the Cretaceous-Paleogene boundary, P. Natl. Acad. Sci.
W., Bijl, P. K., Sluijs, A., Langereis, C. G., and Brinkhuis, H.: A USA, 100, 599–604, https://doi.org/10.1073/pnas.0234701100,
paleolatitude calculator for paleoclimate studies, PLOS One, 10, 2003.
e0126946, https://doi.org/10.1371/journal.pone.0126946, 2015. Winkelstern, I. Z. and Lohmann, K. C.: Shallow burial alteration of
Vellekoop, J., Sluijs, A., Smit, J., Schouten, S., Weijers, J. dolomite and limestone clumped isotope geochemistry, Geology,
W. H., Sinninghe Damste, J. S., and Brinkhuis, H.: Rapid 44, 467–470, https://doi.org/10.1130/G37809.1, 2016.
short-term cooling following the Chicxulub impact at the Woelders, L., Vellekoop, J., Kroon, D., Smit, J., Casadío, S., Prám-
Cretaceous-Paleogene boundary, Proceedings of the National paro, M. B., Dinarès-Turell, J., Peterse, F., Sluijs, A., Lenaerts, J.
Academy of Sciences, P. Natl. Acad. Sci. USA, 111, 7537–7541, T. M., and Speijer, R. P.: Latest Cretaceous climatic and environ-
https://doi.org/10.1073/pnas.1319253111, 2014. mental change in the South Atlantic region, Paleoceanography,
Vellekoop, J., Esmeray-Senlet, S., Miller, K. G., Browning, J. 32, 466–483, https://doi.org/10.1002/2016PA003007, 2017.
V., Sluijs, A., Schootbrugge, B. van de, Damsté, J. S. S., and Woelders, L., Vellekoop, J., Weltje, G. J., de Nooijer, L., Reichart,
Brinkhuis, H.: Evidence for Cretaceous-Paleogene boundary G.-J., Peterse, F., Claeys, P., and Speijer, R. P.: Robust multi-
bolide “impact winter” conditions from New Jersey, USA, Ge- proxy data integration, using late Cretaceous paleotemperature
ology, 44, 619–622, https://doi.org/10.1130/G37961.1, 2016. records as a case study, Earth Planet. Sc. Lett., 500, 215–224,
Vellekoop, J., Sluijs, A., and Speijer, R. P.: An acme of the dinoflag- https://doi.org/10.1016/j.epsl.2018.08.010, 2018.
ellate cyst Palynodinium grallator, Gocht, 1970; a marker for the Zhang, J. Z., Petersen, S. V., Winkelstern, I. Z., and Lohmann,
late Maastrichtian warming event at Northern mid-latitudes?, in: K. C.: Seasonally variable aquifer discharge and cooler cli-
6th International Geologica Belgica Congress: 12-14 September mate in Bermuda during the Last Interglacial revealed by suban-
2018 – Leuven, Belgium, edited by: Elsen, J., Hulsbosch, N., nual clumped isotope analysis, Paleoceanogr. Paleoclimatol., 36,
and Stassen, P., Geologica Belgica Conference Proceedings, 3, e2020PA004145, https://doi.org/10.1029/2020PA004145, 2021.
298 p., https://doi.org/10.20341/gbcp.vol3, 2018. Zhang, L., Wang, C., Wignall, P. B., Kluge, T., Wan, X., Wang, Q.,
Vellekoop, J., Woelders, L., Sluijs, A., Miller, K. G., and Spei- and Gao, Y.: Deccan volcanism caused coupled pCO2 and ter-
jer, R. P.: Phytoplankton community disruption caused by lat- restrial temperature rises, and pre-impact extinctions in northern
est Cretaceous global warming, Biogeosciences, 16, 4201–4210, China, Geology, 46, 271–274, https://doi.org/10.1130/G39992.1,
https://doi.org/10.5194/bg-16-4201-2019, 2019. 2018.
Vellekoop, J., van Tilborgh, K. H. V., Knippenberg, P. V., Jagt, Zhao, M., Ma, M., He, M., Qiu, Y., and Liu, X.: Evaluation of
J. W. M., Stassen, P., Goolaerts, S., and Speijer, R. P.: Type- the four potential Cretaceous-Paleogene (K–Pg) boundaries in
Maastrichtian gastropod faunas show rapid ecosystem recov- the Nanxiong Basin based on evidences from volcanic activity
ery following the Cretaceous–Palaeogene boundary catastrophe, and paleoclimatic evolution, Sci. China Earth Sci., 64, 631–641,
Palaeontology, 63, 349–367, https://doi.org/10.1111/pala.12462, https://doi.org/10.1007/s11430-020-9736-0, 2021.
2020.

O'Hora Et Al., 2022. Clumped-Isotope-Derived Climate Trends Leading Up To The End-Cretaceous Mass Extinction in Northwestern Europe

Uploaded by

Copyright:

Available Formats

O'Hora Et Al., 2022. Clumped-Isotope-Derived Climate Trends Leading Up To The End-Cretaceous Mass Extinction in Northwestern Europe

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

O'Hora Et Al., 2022. Clumped-Isotope-Derived Climate Trends Leading Up To The End-Cretaceous Mass Extinction in Northwestern Europe

Uploaded by

Copyright:

Available Formats

Clim.

Past, 18, 1963–1982, 2022

Clumped-isotope-derived climate trends leading up to the

Smithsonian Institute, Washington, DC 20013, USA

Correspondence: Sierra V. Petersen ([email protected])

Received: 4 August 2021 – Discussion started: 9 September 2021

Published by Copernicus Publications on behalf of the European Geosciences Union.

Clim. Past, 18, 1963–1982, 2022 https://doi.org/10.5194/cp-18-1963-2022

https://doi.org/10.5194/cp-18-1963-2022 Clim. Past, 18, 1963–1982, 2022

3 Methods Taxa-specific clumped isotope vital effects in bivalves are

Clim. Past, 18, 1963–1982, 2022 https://doi.org/10.5194/cp-18-1963-2022

https://doi.org/10.5194/cp-18-1963-2022 Clim. Past, 18, 1963–1982, 2022

each sample, individual replicates were deemed to be outliers

Clim. Past, 18, 1963–1982, 2022 https://doi.org/10.5194/cp-18-1963-2022

LAV-Psop 66.4 56.75 GT/NA −1.7 0.1 2.9 0.1 – – – – – –

Clim. Past, 18, 1963–1982, 2022

Clim. Past, 18, 1963–1982, 2022

https://doi.org/10.5194/cp-18-1963-2022 Clim. Past, 18, 1963–1982, 2022

Clim. Past, 18, 1963–1982, 2022 https://doi.org/10.5194/cp-18-1963-2022

https://doi.org/10.5194/cp-18-1963-2022 Clim. Past, 18, 1963–1982, 2022

Clim. Past, 18, 1963–1982, 2022 https://doi.org/10.5194/cp-18-1963-2022

https://doi.org/10.5194/cp-18-1963-2022 Clim. Past, 18, 1963–1982, 2022

Clim. Past, 18, 1963–1982, 2022 https://doi.org/10.5194/cp-18-1963-2022

https://doi.org/10.5194/cp-18-1963-2022 Clim. Past, 18, 1963–1982, 2022

Clim. Past, 18, 1963–1982, 2022 https://doi.org/10.5194/cp-18-1963-2022

https://doi.org/10.5194/cp-18-1963-2022 Clim. Past, 18, 1963–1982, 2022

Clim. Past, 18, 1963–1982, 2022 https://doi.org/10.5194/cp-18-1963-2022

https://doi.org/10.5194/cp-18-1963-2022 Clim. Past, 18, 1963–1982, 2022

Clim. Past, 18, 1963–1982, 2022 https://doi.org/10.5194/cp-18-1963-2022

You might also like