NALPS19: Sub-orbital scale climate variability recorded in Northern Alpine speleothems during the last glacial period

Sub-orbital-scale climate variability of the last glacial period provides important insights into the rates that the climate can change state, the mechanisms that drive such changes, and the leads, lags and synchronicity occurring across different climate zones. Such short-term climate variability has previously been investigated using δO from speleothems (δOcalc) that grew along the northern rim of the Alps (NALPS), enabling direct 15 chronological comparisons with δO records from Greenland ice cores (δOice). In this study, we present NALPS19, which includes a revision of the last glacial NALPS δOcalc chronology over the interval 118.3 to 63.7 ka using eleven, newly-available, clean, precisely-dated stalagmites from five caves. Using only the most reliable and precisely dated records, this period is now 90 % complete and is comprised of 16 stalagmites from seven caves. Where speleothems grew synchronously, the timing of major transitional events in δOcalc between 20 stadials and interstadials (and vice versa) are all in agreement on multi-decadal timescales. Ramp-fitting analysis further reveals that, except for one abrupt change, the timing of δO transitions occurred synchronously within centennial-scale dating uncertainties between the NALPS19 δOcalc record and the Asian Monsoon composite speleothem δOcalc record. Due to the millennial-scale uncertainties in the ice-core chronologies, a comprehensive comparison with the NALPS19 chronology is difficult. Generally, however, we find that the 25 absolute timing of transitions in the Greenland Ice Core Chronology (GICC) 05modelext and Antarctic Ice Core Chronology (AICC) 2012 are in agreement on centennial-scales. The exception to this is during the interval 100 to 115 ka, where transitions in the AICC2012 chronology occurred up to 3,000 years later than in NALPS19. In such instances, the transitions in the revised AICC2012 chronology of Extier et al. (2018) are in agreement with NALPS19 on centennial scales, supporting the hypothesis that AICC2012 appears to be considerably too young 30 between 100 to 115 ka. Using a ramp-fitting function to objectively identify the onset and the end of abrupt transitions, we show that δO shifts took place on multi-decadal timescales in the North Atlantic-sourced regions (N. Alps and Greenland), whereas shifts in the monsoon were on multi-centennial timescales. Given the near-complete record of δOcalc variability during the last glacial period in the northern Alps, we also offer preliminary considerations regarding the controls on mean δOcalc for given stadials and interstadials. We find 35 that as expected, δOcalc values became increasingly lighter with distance from the oceanic source regions, and increasingly lighter with increasing altitude. Exceptions were found for some high-elevation sites that locally display δOcalc values that are heavier-than-expected in comparison to lower-elevation sites, possibly caused by a summer bias in the recorded signal of the high-elevation site, or a winter bias in the low-elevation site. Finally, we propose a new mechanism for the centennial-scale stadial-level depletions in δO such as ‘precursor’ events 40

further reveals that, except for one abrupt change, the timing of δ 18 O transitions occurred synchronously within centennial-scale dating uncertainties between the NALPS19 δ 18 Ocalc record and the Asian Monsoon composite speleothem δ 18 Ocalc record. Due to the millennial-scale uncertainties in the ice-core chronologies, a comprehensive comparison with the NALPS19 chronology is difficult. Generally, however, we find that the 25 absolute timing of transitions in the Greenland Ice Core Chronology (GICC) 05modelext and Antarctic Ice Core Chronology (AICC) 2012 are in agreement on centennial-scales. The exception to this is during the interval 100 to 115 ka, where transitions in the AICC2012 chronology occurred up to 3,000 years later than in NALPS19. In such instances, the transitions in the revised AICC2012 chronology of Extier et al. (2018) are in agreement with NALPS19 on centennial scales, supporting the hypothesis that AICC2012 appears to be considerably too young 30 between 100 to 115 ka. Using a ramp-fitting function to objectively identify the onset and the end of abrupt transitions, we show that δ 18 O shifts took place on multi-decadal timescales in the North Atlantic-sourced regions (N. Alps and Greenland), whereas shifts in the monsoon were on multi-centennial timescales. Given the near-complete record of δ 18 Ocalc variability during the last glacial period in the northern Alps, we also offer preliminary considerations regarding the controls on mean δ 18 Ocalc for given stadials and interstadials. We find 35 that as expected, δ 18 Ocalc values became increasingly lighter with distance from the oceanic source regions, and increasingly lighter with increasing altitude. Exceptions were found for some high-elevation sites that locally display δ 18 Ocalc values that are heavier-than-expected in comparison to lower-elevation sites, possibly caused by a summer bias in the recorded signal of the high-elevation site, or a winter bias in the low-elevation site. Finally, we propose a new mechanism for the centennial-scale stadial-level depletions in δ 18 O such as 'precursor' events 40 Greenland Stadial (GS)-16.2, , as well as the 'within-interstadial'  cooling event. Our new high-precision chronology shows that each of these δ 18 O depletions occurred in the decades and centuries following rapid rises in sea level associated with increased ice-rafted debris and southward shifts of the Intertropical Convergence Zone, suggesting that influxes of meltwater from moderately-sized ice sheets may have been responsible for the cold reversals causing the Atlantic Meridional Overturning Circulation 5 to slow down similar to the Preboreal Oscillation and Older Dryas deglacial events.

Introduction
Speleothems from the northern rim and central European Alps have provided a number of important, highresolution, precisely 230 Th-dated records of both orbital-and millennial-scale climate variability during the last glacial and interglacial periods (Spötl and Mangini, 2002;Boch et al., 2011;Moseley et al., 10 2014;Luetscher et al., 2015;Moseley et al., 2015;Häuselmann et al., 2015). The oxygen isotopic signature of such records (herein referred to as δ 18 Ocalc) has helped improve fundamental understanding of the effect that changes in atmospheric (Luetscher et al., 2015) and North Atlantic circulation (Moseley et al., 2015) have on European climate, whilst the robust chronologies have provided important information about the timescales upon which the climate can change in this well-populated region Moseley et al., 2014). 15 Furthermore, the pattern and timing of excursions in δ 18 Ocalc of northern Alpine speleothems during the last glacial cycle have been shown to be synchronous within dating uncertainties Moseley et al, 2014) with the sawtooth-pattern of changes in the δ 18 O of Greenland ice cores (herein referred to as δ 18 Oice) (known as Dansgaard-Oeschger cycles;Johnsen et al., 1992;Dansgaard et al., 1993), thus reflecting the shared North Atlantic moisture source and integrated climate system . The sawtooth pattern of δ 18 O is 20 generally interpreted in both Greenland and the northern Alps as being caused by a rapid increase in temperature and humidity leading into a mild climate state (interstadial), followed by a gradual cooling leading into a cold and dry glacial state (stadial). In total, 25 such cycles of rapid warming and gradual cooling, as well as many other smaller centennial-and decadal-scale events, are recognised as having occurred during the last glacial period (Dansgaard et al., 1993;NGRIP Project members, 2004;Capron et al., 2010a). This has resulted in a new 25 stratigraphic framework (INTIMATE event stratigraphy) for abrupt climate changes in Greenland, in which shorter-scale events that occur within the 25 main stadials and interstadials are designated "a to e" (Rasmussen et al., 2014). This nomenclature will be used in the remainder of this article.
When considering the timing of the transitions in δ 18 O between stadial and interstadial states, the largest offsets between the northern Alps speleothem chronology (NALPS) and Greenland Ice Core Chronology (layer-counted 30 GICC05, 0 to 60 ka; Svensson et al., 2008 and modelled GICC05modelext, 60 to 122 ka;Wolff et al., 2010) are 767 years in Marine Isotope Stage (MIS) 3 (Moseley et al., 2014) and 1,060 years in MIS 5 . The former is associated with the warming transition into Greenland , and the latter with the cooling transition into Greenland Stadial-22 (GS-22;Rasmussen et al., 2014). The timing for both of these transitions in the NALPS chronology was constrained from speleothems high in detrital thorium (Boch et al., 35 2011;Moseley et al., 2014). Since one of the prerequisites for reliable 230 Th dating is that minimal 230 Th is incorporated into the calcite at the time of deposition (Ivanovich and Harmon, 1992;, it is reasonable to question the accuracy of the age of these two transitions. In the case of the MIS 3 sample (Moseley et al., 2014), the correction for the initial incorporation of daughter nuclides was well constrained by isochron methods (Ludwig and Titterington, 1994;, however, in the case of the MIS 5 sample , the detrital Th was corrected for using an a priori assumption that the contaminant phase had the same composition of the silicate bulk earth (Wedepohl, 1995). Furthermore, the accuracy of the GICC05modelext chronology is questionable in the vicinity of GI-22 to 21 (Capron et al., 2010b;Vallelonga et al., 2012).
Specifically, the duration of GS-22 appears to be underestimated, probably as a result of an overestimation of the 5 annual layer thickness by the ss09sea06bm ice flow model (Johnsen et al., 2001) upon which GICC05modelext is based in the portion of the record older than 60 ka (Wolff et al., 2010;Vallelonga et al., 2012). Vallelonga et al. (2012) thus revised the duration of GS-22 from 2,620 years to 2,894 ± 99 years using annual layer-counting of seasonal cycles in the chemical impurities in the ice. Given the uncertainties in the chronologies for both the NALPS speleothems and NGRIP ice core during GI-22 to GI-21, it is thus difficult to determine the reliability 10 and extent of the leads, lags and synchronicity at this time. In addition to the complexities around GS-22, the chronology of events between GI-25 to 23 are also poorly constrained. This is visible when comparing the GICC05modelext chronology (Wolff et al., 2010) with the Antarctic Ice Core Chronology (AICC)2012 chronology (Veres et al., 2013), which differ by up to 2,700 years, and also when comparing the pattern of the δ 18 O shifts during GS-24 in NALPS and NGRIP (Boch et al., 2011). 15 Here, we revisit the NALPS speleothem chronology over the interval 63.7 to 118.3 thousand years ago (ka)  using new samples that are low in detrital thorium and/or have a more pronounced δ 18 Ocalc signal, with the aim of improving the chronology such that better informed conclusions about leads/lags and synchronicity in the climate system may be made. The original record was discontinuous, with coverage of 76 % of the 54.6 ka interval. Gaps in the record were present between 111.6 and 110.0, 94. 5 and 89.7, 84.7 and 83.0, 20 77.5 and 76.0, 75.5 and 72.0 ka (Boch et al., 2011). With the addition of new speleothems, we extend the coverage to 90 %, improve the accuracy and precision of some climate transitions, and designate the revised chronology "NALPS19".

Regional Climate
The European Alps, situated between 44 and 48 °N, are a 950 km-long mountain range running ENE-WSW 25 located close to the southern fringe of the European mainland. The highest peaks, reaching over 4,000 m in elevation, are situated in the Western Alps of France and Switzerland, whilst the Eastern Alps, located in Austria, are on average 1,000 m lower. Across the whole of the Alps, the average elevation is c. 2,500 m above sea level (a.s.l.), thus this mountain range forms a major topographic barrier between the North Atlantic and Mediterranean climate zones (Wanner et al., 1997). Today, the Alps are located to the south of the extra-tropical 30 westerlies, which bring precipitation sourced from the North Atlantic to the northern and western flanks, particularly during winter and spring (Wanner et al., 1997;Sodemann and Zubler, 2010). Lagrangian backtrajectory studies have shown that for the period 1995-2002, the North Atlantic contributed c. 40% of the annual mean moisture to the Alps, whilst the Mediterranean contributed 23%, the Arctic, Nordic and Baltic seas 16%, and the European land masses 21% (Sodemann and Zubler, 2010). Contributions to the northern versus southern 35 side of the Alps, however, displayed considerable seasonal differences. Throughout the year, the North Atlantic contributes more moisture to the northern Alps as compared to the Southern Alps, and this is especially pronounced in winter and spring (Sodemann and Zubler, 2010). During summer, Central European land masses are the dominant moisture source across the entire Alps, though the North Atlantic still makes some contribution to the northern flanks, and the Mediterranean to the southern flanks. In autumn, the northern Alps receive comparable quantities of moisture from both the North Atlantic and Mediterranean, whilst the Southern Alps are dominated by moisture from the Mediterranean (Sodemann and Zubler, 2010). On longer, multi-decadal timescales, moisture sources and trajectories to the Alps have been shown to be highly variable. In particular, the phase of the North Atlantic Oscillation (NAO), which is especially pronounced in winter (Wanner et al, 1997), 5 exhibits one of the strongest controls. During the positive phase, when positive sea-surface temperature and airpressure anomalies build up in the southwestern North Atlantic, and negative ones in the north, the associated temperature gradient across the western North Atlantic is high. This leads to an intensification of the North Atlantic polar front jet stream, which creates a high pressure zone over the Alps and Mediterranean causing higher temperatures and less precipitation (Wanner et al., 1997). Conversely, during a negative NAO phase the 10 air pressure decreases over the Alps and Mediterranean leading to lower air temperatures and higher precipitation.

Cave Sites and Speleothems
Previous NALPS studies include MIS 2 in Luetscher et al. (2015) (though this was not branded as 'NALPS'), 15 MIS 3 in Moseley et al. (2014), and MIS 4 to MIS 5 in Boch et al. (2011). The MIS 4/ MIS 5 chronology (which is the part revised here), was constructed from seven speleothems from four cave sites including St. Beatus caves, Große Baschg cave (Baschg cave for short), Klaus-Cramer cave and Schneckenloch .
In this study, two additional samples from Baschg cave and one from Schneckenloch were analysed, plus one sample from Hölloch im Mahdtal (Hölloch cave for short), one from Grete-Ruth-Shaft, and six from Gassel 20 Tropfsteinhöhle (Gassel cave for short). All cave sites are located on the northern rim of the European Alps ( Fig.   1) and have small catchments of less than a few km 2 . The distance between the most westerly and easterly caves is c. 475 km. Details of the speleothems analysed in this study and their respective caves are given in Table 1 whereas images of the respective samples are given in SI Fig. 1.

Analytical Methodology 25
The eleven stalagmites were cut in half along their growth axis and polished by a professional stone mason. Pilot dating studies guided the sample size that was needed for high precision ages. Sub-samples of between 20 to 150 mg were hand-drilled using a handheld-drill fitted with carbide burr-tipped drill bits of diameter 0.5 to 0.8 mm in a laminar-flow hood. The cleanest, densest growth layers were targeted for sampling.
Chemical procedures and aliquot measurements were undertaken in the Trace Metal Isotope Geochemistry 30 Laboratory at the University of Minnesota. Separation and purification of U and Th aliquots from the subsamples was undertaken using standard methods (Edwards et al., 1987) in a clean air environment. Samples were spiked with a dilute mixed 229 Th-233 U-236 U tracer to allow for correction of instrumental fractionation and calculation of U and Th concentrations and ratios. Procedural chemistry blanks were on the order of 5 -83 ag for 230 Th, 2 -523 fg for 232 Th, 73 to 171 ag for 234 U, and 0.2 to 1.6 pg for 238 U. Aliquots of U and Th were analysed 35 on a ThermoFisher Neptune multi-collector inductively coupled plasma mass spectrometer (MC-ICPMS) in peak-jumping mode on the secondary electron multiplier (Shen et al., 2012) Stable isotopes (δ 18 Ocalc and δ 13 Ccalc) were typically micro-milled at a spatial resolution of 250 µm (with the exception of GAS22=200 µm and BA7=500 µm) from the central axis of each sample (SI Fig. 2). In total 5,000 new measurements were made for this study at the University of Innsbruck on a ThermoFisher Delta plus XL isotope ratio mass spectrometer linked to a Gasbench II. Analytical precisions are 0.08‰ and 0.06‰ for δ 18 Ocalc and δ 13 Ccalc, respectively (1σ) . All isotope results are reported relative to the Vienna PeeDee 5 Belemnite standard. In addition to the main isotope track along the central axis, Hendy tests (Hendy, 1971) were also prepared for each sample as a first-order assessment of whether the respective stalagmite was deposited under conditions of isotopic equilibrium, though the preferred approach in recent years has been to reproduce the data in a second stalagmite (Dorale and Liu, 2009). Under the 'Hendy test' criteria,δ 18 Ocalc values should remain constant along a single growth layer, and there should be no correlation between δ 18 Ocalc and δ 13 Ccalc that might 10 otherwise indicate kinetic fractionation. Bayesian age models were constructed for all eleven samples using OxCal version 4.2 for Poisson-process depositional models ('p sequence') and a variable 'k parameter' of 0.001 to 10 mm a-1 (Bronk Ramsey, 2008;Bronk Ramsey and Lee, 2013).

Results
The results of the U-Th MC-ICPMS measurements and associated age calculations can be found in SI Table 1. 15 Age modelling results including growth rates can be seen in SI Fig. 3. The correlation between δ 18 Ocalc and δ 13 Ccalc is shown in SI Fig. 4, whereas the results of Hendy tests are shown in SI Table 2. The key features of all of these results are summarised in Table 2 and will be discussed briefly here. Generally, all speleothems have 238 U concentrations of c. 250 to 1,500 ng g -1 , which are values typical of common alpine dripstones. The cleanest samples, as indicated by high 230 Th/ 232 Th ratios, are from Grete-Ruth Shaft (HUN14) and Gassel cave (GAS12,20 13,22,25,27,29). Correction of final ages for detrital Th contamination in these samples is therefore negligible (SI Table 1). The samples from Baschg, Schneckenloch, and Hölloch caves are all variably contaminated with detrital Th. In the case of BA5, this results in corrections to younger ages of 57-135 years, which are within the levels of dating uncertainty (c. 300 to 400 years) (SI Table 1). BA-7 is the 'dirtiest' of the samples analysed here.
Of the 16 U-Th ages used in the age model, nine are shifted less than 1,000 years to younger ages (SI Table 1). 25 SCH6 has varying levels of detrital Th contamination, being very clean in the older part between c. 75.9 to 77.9 ka, but moderately dirty in the younger section between 74.4 to 75.5 ka (SI Table 1). The majority of the age models is thus constructed from clean samples. The internal morphology of HÖL19 is variable and contains sections of clean calcite, dirty calcite, and calcified loam layers (SI Fig. 1). The youngest part of the stalagmite dates to the late Holocene and the late glacial (SI Table 1) and thus is outside the time frame for this study. 30 Between c.95 mm and c.160 mm from the top, the stalagmite is rich in calcified loam layers and thus is not suitable for dating. Below 160 mm there are a number of sections of clean and dirty calcite. Here we have concentrated on the cleanest part between 187.25 and 226.75 mm from the top. Correction of these ages for detrital Th results in a shift to younger ages of 64 to 213 years, which is within the c. 300-400 years range of dating uncertainty (SI Table 1). Linear regression analysis between δ 18 Ocalc and δ 13 Ccalc, which is used as a test 35 for isotopic equilibrium (Hendy, 1971;Dorale and Liu, 2009), yields extremely low R 2 values below 0.3 for the majority of samples (Table 2) suggesting that kinetic fractionation did not occur. Only GAS22 has an R 2 of 0.4 and GAS27 an R 2 of 0.6 indicating a minor correlation. Variation in δ 18 Ocalc across single growth layers is also generally low, with the exception of one out of five tests in GAS27 yielding a range of 0.8 ‰ (Table 2).

Coherence and updates to NALPS19 versus NALPS
The new records produced in this study ( Fig. 2b)  118.3 to 63.7 ka. Within this interval, the record is now 90 % complete, compared to 76 % in Boch et al. (2011).
Where speleothems grew synchronously, major transitional events between stadials and interstadials (and vice versa) are all in agreement within uncertainty, which can be very clearly seen in SI Fig. 5. In the interest of completeness and transparency, we present all δ 18 Ocalc records here, however, some samples are cleaner than others as discussed in section 3 (i.e. low in detrital Th as indicated by a higher 230 Th/ 232 Th ratio) and thus deemed 10 to be more reliably dated. The NALPS19 chronology is therefore constructed from only the most reliably dated records from this study and Boch et al. (2011) (Fig. 2c). Considering the construction of NALPS19 further and generally working from youngest to oldest: samples KC1 and HÖL19 are included on the basis that they are the only records available that cover the transitions into stadials-19 and 20. The transition into interstadial-20 is present in both SCH6 (this study) and BA1b . Both samples have comparable levels of detrital 15 Th, and the dating precision of the transition in both samples is c. 200 to 250 years. Given the comparative cleanliness and dating precision, as well as the reproducibility of the timing of the transition to within c. 50 to 85 years (SI Fig. 5), both samples are included in NALPS19. Samples covering the transition into stadial-21 include GAS12 and 13 (this study) and BA1b . Samples from GAS12 and GAS13 are extremely clean with dating precisions of 250 to 300 years (Table 2), whereas those from BA1b are generally moderate to very 20 clean. Critically though, GAS12 contains six ages and over 60 δ 18 Ocalc measurements within the transition, and GAS13, three ages and over 130 δ 18 Ocalc measurements (SI Fig. 2). On the other hand, BA1b has only three δ 18 Ocalc measurements in the transition, and one age which is quite dirty resulting in an age corrected to younger values by 760 years and a dating precision of 580 years ). Based on the higher resolution and higher precision provided by GAS12 and GAS13, as well as the fact they are reproducible during the transition 25 on sub-and decadal timescales, we therefore include GAS12 and GAS13 in NALPS19 and omit BA1b. EXC4 is then included for the interstadial-21 portion on the basis that it is clean. However, for this section it only contains the interstadial and no transitions, therefore it is excluded from the discussion on transition timing (Section 4.2).
The transition into interstadial-21 is captured in BA1 , BA7 and GAS25 (this study). As discussed above, GAS25 is extremely clean, thus correction for detrital Th is negligible and the dating precision 30 is on the order of 300 to 400 years (SI Table 2). In contrast, BA7 is the dirtiest of the samples with large corrections for detrital Th (SI Table 2), whereas BA1 is moderately dirty resulting in comparable shifts to younger ages . Ideally, the complete transition would be constrained only in GAS25 since this sample is the most reliable and best dated, but unfortunately this record is limited to growth mainly during and just after the transition. We therefore include GAS25 where it is applicable and omit BA1 and BA7, but then 35 keep BA1 and BA7 for the parts of the record where there is no alternative available. The transition into stadial-22 is present in GAS25, BA5 (this study), and BA2 . The situation here is similar to the transition into interstadial-21, where GAS25 is the superior sample with higher dating quality. GAS25 therefore takes priority, whereas BA5 is included to complete the stadial part of the record. BA2 is completely omitted from NALPS19 on the basis that correcting for detrital Th causes shifts in ages of centuries  as 40 compared to decades in GAS25 ( Table 2). The section of the record encompassing interstadial-23, stadial-24, and interstadial-24 is fully covered by GAS22, GAS27, GAS29 and HUN14, which are all extremely clean, well-dated records with typical dating precisions of 300 to 400 years (Table 2). Furthermore, the timing of the transition into interstadial-23 is reproducible to within 60 to 100 years between GAS27 and HUN14. The timing of the transition into stadial-24 is in agreement on the order of 40 to 60 years in GAS22, GAS29 and HUN14. 5 Furthermore, the pattern of δ 18 Ocalc shifts across the whole interstadial 24 to 23 period is remarkably similar in the new speleothems analysed here to the pattern of events in NGRIP δ 18 Oice across the same period. This suggests the new speleothem samples are capturing a bigger-scale climate signal, unlike EXC3 and EXC4 from St. Beatus cave , which show a distinctly different pattern in δ 18 Ocalc across this time period.
The reason for the difference is unknown, and is likely due to some local influence or control at the cave site. We 10 acknowledge that there is still value in the St. Beatus records, but they are not ideal for investigations into leads, lags, and synchronicity when more comparable records exist, thus they are not included in NALPS19. Finally, the new record from HUN14 is used to complete the gap that existed previously at stadial-25.  (4), revision of the interval GI-23.1 to GI-25c, which includes the addition of the previously absent GI-25a and b and a more distinctive 'shape' to GS-24 in-line with NGRIP. (2019). The ramp-fitting function is similar to the one used by Mudelsee (2000), but instead uses probabilistic 25 inference to define a transition via a linear ramp between two constant levels. Such an approach enables a consistent approach to chronological quantification of climate transitions (Mudelsee, 2000), unlike the more subjective approach of taking the first data-point that deviates from the baseline of the previous climate state (e.g. Capron et al., 2010a;Moseley et al., 2014;Rasmussen et al., 2014). Adolphi et al. (2018) applied such a ramp-fitting method to the younger, late glacial portion of the NGRIP δ 18 Oice record (Adolphi et al., 2018), 30

Chronological implications
whereas Steffensen et al. (2008) applied another ramp-fitting method through the last deglacial. For this study, ramp fitting was applied to: (1) δ 18 Ocalc of the new NALPS19 record (this study); (2) δ 18 Ocalc of the Asian monsoon composite speleothem record (Cheng et al., 2016); (3) NGRIP δ 18 Oice on the GICC05modelext chronology, which is comprised of a composite layer-counted chronology to 60 ka (Svensson et al., 2008) followed by splicing of the ss09sea-modelled chronology (Johnsen et al., 2001) between 60 to 122 ka onto the 35 younger annual-layer counted chronology (Wolff et al., 2010); (4) NGRIP δ 18 Oice on the AICC2012 chronology, which is constructed using glaciological inputs, relative and absolute gas and ice stratigraphic markers, and Bayesian modelling (Veres et al., 2013), and; (5) NGRIP δ 18 Oice on the AICC2012 chronology updated by aligning δ 18 O of the atmosphere as measured in EPICA Dome C with δ 18 Ocalc of Chinese speleothems (Extier et al., 2018). 40 Results of the ramp-fitting are shown in Table 3, Figs. 4 and 5, and SI Fig. 7. Unfortunately, results are not available for some transitions because their shape is incompatible with the transition model, which requires stable periods before and after the transitions. Where multiple NALPS19 speleothems grew synchronously, excellent agreement is found in the absolute timing of transitions, which show differences from as low as 10 years between GAS12 and GAS13 during the endpoint of the transition into GS-21.1, up to a maximum of only 5 163 years difference between GAS22 and HUN14 during the endpoint of the transition into GS-24.1 (i.e., within the 318 years uncertainty of GAS22 at this point) (Table 3, Fig. 4). Similarly, for the NGRIP δ 18 Oice record on the GICC05modelext chronology, we find that the timing of the start of the respective transitions are in excellent agreement (2 to 119 years) between the ramp-fitting used in this study and the INTIMATE event stratigraphy scheme (Rasmussen et al., 2014) (Table 3). Comparison between the timing of the ramp-fitted transitions in 10 NALPS19 and the Asian monsoon speleothem records also show excellent agreement within centennial-scale uncertainties, with the exception of GS-20, which is older in NALPS19 by c. 900 years (Table 3, Figs. 4 and 5).
The NALPS19 age for GS-20 is, however, in very good agreement on a multi-decadal scale with the GICC05modelext chronology (details below). It should be noted that a comprehensive comparison of the timing of transitions between NALPS19 and NGRIP on the three ice-core chronologies is made difficult because of the 15 large uncertainties associated with AICC2012 (c. 3,000 -3,200 years; Veres et al., 2013) and even the absence of uncertainties associated with GICC05modelext (Wolff et al., 2010). To deal with the absence of uncertainties in GICC05modelext, we take the approach of Abbott et al. (2012) and extrapolate the linear trend in ratio between age and uncertainty from the layer-counted 0-60 ka GICC05 chronology (Svensson et al., 2008), which yields an uncertainty of c. 4.5 % by 120 ka (Table 3, Fig. 4). In reality, the uncertainty is likely to be considerably less 20 since well-dated markers exist in some places (e.g. Guillou et al., 2019). Nevertheless, if only the central age is considered (where + indicates the respective chronology is earlier=older than NALPS19, andis vice versa), excellent agreement in the absolute timing of the transition is displayed between NALPS19 and GICC05modelext for GS-20, which is offset by c. -45 years, and GI-21.1, which is offset by c. +20 to +80 years (Table 3 Fig. 5). The difference in durations for GI-21.1 (54 to 107 years) and GI-23.1 (68 to 138 years) is slightly larger but still comparable on multi-decadal timescales. The duration of GI-23.1 in the Asian monsoon is 5 also comparable at 81 years. The greatest difference between NALPS19 and the Greenland chronologies is displayed for GI-20, which varies between 21 to 114 years. Interestingly, with the exception of GI-23.1, the duration of transitions in the Asian monsoon are considerably longer (on multi-centennial timescales) than for the North Atlantic-sourced NALPS19 and Greenland chronologies (on multi-decadal timescales) (Table3, Fig.   5). 10 The NALPS19 chronology also enables a new consideration of the duration of GS-22, which previously has been the subject of debate given the various different timescales presented in the literature Vallelonga et al., 2012). Here, we use the same strategy as for the previous studies and define the duration of  Table 4). The Vallelonga et al. (2012) layer-counted chronology thus indicated a longer duration for GS-22 than the GICC05modelext chronology (2,620 years) and a shorter duration for 20 the precursor event (300 years) (together 2,920 years; Table 4) (Wolff et al., 2010). The ramp fitting function was not able to constrain the transition into the precursor event (GI-21.2), thus we consider here the duration of the full period from the cooling into GS-22 to the warming into GI-21.1e, which in the NALPS19 chronology is 3,993 ± 155 years (Table 4). This finding is in agreement with the duration from the previous NALPS chronology of 3,660 ± 526 years (Table 4), but is c. 1,000 years longer than in GICC05modelext and 750 years 25 longer than in the layer-counted chronology (395 years if the uncertainties are considered). In contrast, a relatively long duration of 4,122 ± 650 years has been proposed for NGRIP on the EPICA Dronning Maud Land (EDML) Antarctic ice core chronology (Capron et al., 2010b;Vallelonga et al., 2012), which is in agreement with the duration from NALPS19. Additionally, the duration of the same period as estimated from the Asian monsoon composite record is 4,489 ± 960 years. The speleothem δ 18 Ocalc records from both the Alps and China 30 therefore support a longer-duration for the period between the cooling into GS-22 to the warming into GI-21.1e, which is in line with the NGRIP-EDML chronology (Capron et al., 2010b;Vallelonga et al., 2012).

NALPS δ 18 O variability during the last glacial period (15-120 ka)
Speleothem deposits from the northern rim of the Alps now provide a near-complete record of δ 18 Ocalc variability during the last glacial period ( Fig. 6; Boch et al., 2011;Moseley et al., 2014;Luetscher et al., 2015), which is 35 remarkably similar to δ 18 O variability recorded in the NGRIP Greenland ice core during the same period. It is hypothesised that the moisture source for both regions during the last glacial period was the North Atlantic, with the primary control on the δ 18 O of precipitation in both Greenland and the Alps being temperature . Changes to the transport pathway have, however, been proposed for the northern Alpine speleothem record of the Last Glacial Maximum (LGM) between 26.5 and 23.5 ka induced by a southward shift in the North 40 Atlantic storm track (Luetscher et al., 2015). The change to the transport pathway is, however, only considered to affect the LGM and not the remainder of the glacial (Luetscher et al., 2015).
We now consider the full glacial Alpine speleothem δ 18 Ocalc record in further detail. In addition to the NALPS records of Boch et al. (2011), Moseley et al. (2014) and NALPS19 (this study), we also consider an MIS 5 record from Siebenhengste (SI Fig. 9), a large cave system located on the northern rim of the Alps of 5 Switzerland (Fig. 1), and a record from Kleegruben cave , which is located in the Central Alps of Austria to the north of the main Alpine crest (Fig. 1). A thorough investigation of the controls on the δ 18 O of precipitation would require a sophisticated modelling approach, which is beyond the scope of this paper, thus here we appreciate that our investigation is a first consideration only. Furthermore, given the many different factors that can influence the δ 18 O of precipitation (Dansgaard, 1964;Rozanski et al., 1993;Clark and Fritz, 10 1997), it would be advantageous to have stable isotope information from fluid inclusions. Unfortunately, the speleothems presented here are largely devoid of fluid inclusions (Brandstätter, unpubl. data).
Today, temperature has been shown to have the most dominant control on the δ 18 O of precipitation along the northern rim of the Austrian Alps (Kaiser et al., 2002;Hager and Foelsche, 2015), though other factors such as a changing moisture source, rain-out along the different transport pathways (continental effect), altitude (altitude 15 effect), the North Atlantic Oscillation, and locally also the amount of rain (amount effect) all have some additional control (Ambach et al., 1968;Dray et al., 1998;Kaiser et al., 2002;Hager and Foelsche, 2015;Deininger et al., 2016). To consider the effects of these controls on the δ 18 O of precipitation during the last glacial period, we have first removed from the speleothem records the variability in mean ocean δ 18 O caused by fluctuations in ice volume (Fig. 5) using a rate of 0.012 ‰ m -1 (Rohling, 2013) and the sea-level curve of Grant 20 et al. (2012).
Mean speleothem δ 18 Ocalc values for individual stadials and interstadials in the ice-volume corrected record have been calculated for each sample (Fig. 7a, SI Fig. 8, SI Table 3). Excluding the samples associated with the LGM because of the different transport pathway (Luetscher et al., 2015), the δ 18 Ocalc range in mean interstadial values is 5.0 ‰ (Klaus Cramer (-7.9 ‰) and Siebenhengste (-7.9 ‰) to Kleegruben (-12.9 ‰)), whilst the range in 25 mean δ 18 Ocalc stadial values is slightly larger (but comparable) at 5.4 ‰ (Siebenhengste (-9.5 ‰) to Kleegruben (-14.9 ‰)) ( Fig. 7a). We now consider the controls on δ 18 Ocalc during periods when more than one speleothem was deposited, specifically GI-23.1, GS-23.2, GI-24.1, and GS-24.1. Generally it is considered that the dominant control on the δ 18 O of precipitation in the northern and central Alps during the last glacial period was temperature, and the dominant moisture source was the North Atlantic (as both are today). The correlation 30 between both temperature and distance from the North Atlantic as compared to mean δ 18 Ocalc was investigated to identify potential continental and rainout effects. In all instances, mean δ 18 Ocalc became increasingly lighter with increasing distance from the North Atlantic; a medium correlation is displayed for GI-23.1 (r 2 =0.64, n=4), GS-23.2 (r 2 =0.63, n=3), GS-24.1 (r 2 =0.57, n=4, two samples for Gassel), and a lower correlation during GI-24.1 (r 2 =0.16, n=3). This trend of lighter mean δ 18 Ocalc with increasing distance from the source would be expected 35 with progressive rainout and is consistent with present day observations. Today, spatial variability of the δ 18 O of precipitation in the Austrian Alps is highly dependent on altitude (Hager and Foelsche, 2015). We find that medium to strong correlations between catchment elevation and mean δ 18 Ocalc existed during GI-23.1 (r 2 =0.49, n=4), GI-24.1 (r 2 =0.67, n=3), GS-23.2 (r 2 =0.79 n=3), and GS-24.1 (r 2 =0.74, n=4 (Gassel has 2 samples)) (Fig. 7c). For GI-24.1, the relationship shows that mean δ 18 Ocalc becomes 40 increasingly lighter with increasing elevation (as would be expected for altitudinal controls on δ 18 O of precipitation). In contrast, the other examined time periods show an inverse relationship to what would be expected for altitudinal control, with mean δ 18 Ocalc becoming heavier with increasing elevation (Fig. 7c). Since GI-24.1 is the only event that does not contain the high-elevation Siebenhengste site, the mean δ 18 Ocalc of 7H-12 was removed from the linear regression analysis for the three time periods showing an inverse relationship (Fig.   7d). This resulted in a switch to increasingly lighter mean δ 18 Ocalc with increasing elevation for GI-23.1, GS-23.2 5 and GS-24.1 (Fig. 7d) (i.e. in line with an altitudinal control on δ 18 O of precipitation).
Given that there is such limited availability of multiple speleothem δ 18 O records covering the same time periods, it is difficult to make firm conclusions on the controls of δ 18 Ocalc. Here though we offer some hypotheses based on the available data. We have shown that for a given time period δ 18 Ocalc trends towards lighter values with increasing distance from the North Atlantic (Fig. 7b). Despite this, there is some variability overprinted on top of 10 this trend. For instance, even though Grete-Ruth is closer to the North Atlantic than Gassel cave, mean δ 18 Ocalc values for Grete-Ruth are consistently lighter than for Gassel (Fig. 7b). Since Grete-Ruth is located at a higher elevation than Gassel cave (Fig. 7c), the lighter mean δ 18 Ocalc values are likely a product of the altitude effect and associated cooler temperatures. In comparison, St. Beatus and Siebenhengste caves are located within 10 km of one another, and are the closest caves to the North Atlantic of all the caves investigated here. As expected, mean 15 δ 18 Ocalc values are heavier for St. Beatus and Siebenhengste than for Grete-Ruth or Gassel caves (Fig. 7b). Closer investigation, however, shows that during GI-23.1, mean δ 18 Ocalc of the low elevation St. Beatus is lighter than the high-elevation Siebenhengste (Fig. 7b). Given the close proximity of the two caves, the condensation level (and therefore condensation temperature) would have been approximately the same, thus one must consider the reason for the difference in mean δ 18 Ocalc for these two caves. Since the three caves at lower elevation (St. 20 Beatus, Gassel, Grete-Ruth) follow the expected altitude-induced trend of lighter mean δ 18 Ocalc with increasing elevation (Fig. 7d), it seems the anomaly lies with the high-elevation 7H-12 stalagmite from Siebenhengste. One reason for the heavier-than-expected mean δ 18 Ocalc at Siebenhengste could be that the full annual signal is better preserved at high-elevation sites that are less exposed to evapotranspiration effects during the summer season than in more vegetated catchments. Alternatively, a summer bias towards isotopically heavier δ 18 Ocalc at the 25 high-elevation site could for instance be caused by wind erosion resulting in relocation of the isotopically-light winter snow, a process that has been well-documented at various Alpine sites (Ambach et al., 1968;Bohleber et al., 2013;Hürkamp et al., 2019). Eventually, if firn developed above Siebenhengste during GI-23.1, then this would also limit the input of isotopically-light precipitation causing a summer bias in the recorded signal. At present there is, however, no evidence to either support or reject the hypothesis of firn above Siebenhengste 30 during MIS 5.
In summary, speleothems from the northern rim of the European Alps provide δ 18 Ocalc records for the majority of the last glacial period. As expected, the limited data set shows that mean δ 18 Ocalc for specific stadials and interstadials generally trends towards lighter values with increasing distance from the coast and with increasing altitude. An exception is the high-elevation 7H-12 stalagmite from Siebenhengste, which appears to record a 35 stronger summer signal. Further investigation of the controls on δ 18 Ocalc in the northern Alps requires a more sophisticated modelling approach.

Stadial-level centennial-scale cold events
The recognition of centennial-to millennial-scale climate events, such as 'precursors' to interstadials and withininterstadial depletions in δ 18 Oice (Capron et al., 2010a), led to the designation of the INTIMATE event 40 stratigraphy for the Greenland ice cores over the last glacial period (Rasmussen et al., 2014). Typically, a 'precursor-event' is a feature of a stadial-interstadial transition; this includes GS-16.2, 17.2, 21.2 and 23.2 ( Fig.   8) (Capron et al., 2010a;Rasmussen et al., 2014). It is characterised in northern Alpine speleothems and Greenland ice cores by a rapid increase in δ 18 O from stadial to interstadial conditions. The event is short-lived, lasting a maximum of a few centuries before δ 18 O returns to near-stadial conditions for another few decades to 5 centuries, followed by the main transition into the interstadial. [Ca 2+ ] in the Greenland ice cores varies almost simultaneously with these δ 18 Oice changes, where increases in [Ca 2+ ] are associated with depletions in δ 18 Oice and vice versa. Changes in [Ca 2+ ] are interpreted to reflect changes in dust concentration caused by changes in dust source conditions and transport pathways indicative of regional-to-hemispheric-scale circulation changes (Ruth et al., 2007). In comparison, 'within-interstadial' climate perturbations are characterised in general by smaller-10 amplitude depletions in δ 18 Oice that typically do not reach stadial values, and are often of shorter duration than the reversals at stadial-interstadial transitions. [Ca 2+ ] also varies in-tune with 'within-interstadial' δ 18 Oice fluctuations, but similarly does not reach full stadial values. Such characteristics appear to be consistent in δ 18 O records from both Greenland ice cores and northern Alpine speleothems. The exception to such typical 'withininterstadial' cold perturbations is the event at 107.5 ka in the NALPS19 chronology, which is designated GS-15 24.2 in the INTIMATE event stratigraphy scheme (Rasmussen et al., 2014). This drop in δ 18 Ocalc to stadial values occurred 978 years after the start of the interstadial, thus firmly making it a 'within-interstadial' event rather than one associated with a stadial-interstadial transition. At present, the 107.5 ka-event (GS-24.2) is the only centennial-scale δ 18 O event of such extreme amplitude occurring during an interstadial that is recognised in both Greenland and northern Alpine records. Because of this, it has been likened to the 8.2 ka cold event that 20 occurred in the early Holocene (Alley et al., 1997;Capron et al., 2010a). Still, the δ 18 Oice excursion of the 8.2 ka event did not reach near-stadial values in NGRIP as GS-24.2 did, thus highlighting some differences between these two warm-interrupting cold reversals. In addition, Rasmussen et al. (2014) liken the 'within-interstadial' GS-24.2 cold perturbation to stadial-interstadial transition events GS-16.2 and GS-17.2. Both the similarities between GS-24.2 and the 8.2 ka event, as well as with GS-16.2 and GS-17.2, suggest that such abrupt climate 25 variability is not critically influenced by the size of the Greenland ice sheet (Capron et al., 2010a;Rasmussen et al., 2014).
During the deglacial and early Holocene, large-scale meltwater events are widely suggested as being responsible for causing some short-term climate reversals through the weakening of Atlantic meridional overturning circulation (AMOC) (e.g., Broecker et al, 1994;Teller et al., 2002;Clark et al., 2001Clark et al., , 2004. Such cold reversals 30 are thought to be triggered by meltwater events include the Older Dryas at 14 ka (GI-1d, Rasmussen et al., 2014), the Preboreal Oscillation at 11.4 ka (e.g., Johnsen et al., 1992;Björck et al. 1996;Fischer et al., 2002), the 9.3 ka event (Fleitmann et al., 2008;Yu et al., 2010), and the 8.2 ka event (Alley et al., 1997). In contrast though, not all freshwater injections led to cold events, and not all cold events were caused by freshwater injections (Stanford et al., 2006). For instance, both the Younger Dryas and Heinrich events occurred during times of 35 already-colder sea surface temperatures and weakened AMOC, indicating that the input of freshwater from the iceberg armadas was not the initial cause of the AMOC slowdown (e.g., Hall et al., 2006;Henry et al., 2016).
In the case of the centennial-scale cold reversals of   (Fig. 8), a possible mechanism for each of these events could be similar to the meltwater-triggered cold reversals of the deglacial. This hypothesis is supported when considering that events  shortly following episodes of rapid sea-level rise, which were in excess of 12 m ka -1 in the high-resolution record of Grant et al. (2012) (Fig. 8). Such rapid sea-level rise does not appear to have occurred prior to GS-23.2, though closer inspection of the sea-level curve shows that following the rise prior to GS-24.2, sea levels had remained elevated and underwent a series of rapid oscillations that are smoothed out in the rate-of-change curve (Fig. 8). Likewise, GS-16.2 did not occur coincident with an episode of sea-level rise, but did occur shortly after the rise associated with GS-17.2 (Fig. 8). Additionally, the rapid rises in sea level each began at times of 5 increased ice-rafted debris (IRD) in the North Atlantic (McManus et al., 1999, on U-Th timescale), weakened AMOC and increased ice volume as indicated by high benthic δ 13 C and planktic δ 18 O values, respectively, as well as pluvial periods in Brazil caused by a southward shift of the intertropical convergence zone (ITCZ) (Wang et al., 2004) (Fig. 8). In the late glacial, such episodes are associated with Heinrich events (Wang et al., 2004).
Furthermore during glacial terminations, the sequence of events has been shown to include a Heinrich event, 10 followed by short-lived warming, then a millennial-scale return to cold conditions, and finally the transition to the interglacial (Cheng et al., 2009). Though on shorter timescales, the pattern of events during these specific stadial-interstadial transitions is similar to the pattern of events during glacial terminations. The oscillations of .2 at stadial-interstadial transitions can therefore be considered as being akin to the meltwater-triggered Preboreal Oscillation, which occurred shortly following the warming at the end 15 of the Younger Dryas stadial during a time when considerable volumes of ice still existed, similar to the early glacial. These reversals at stadial-interstadial transitions during the early glacial period are therefore not so much warming events that punctuate cold periods (Capron et al., 2010a), but rathermore stadial-interstadial transitions that failed due to freshwater influx. On the other hand, GS-24.2, which occurred nearly 1,000 years after warming occurred, is more similar to the Older Dryas in which a cold event punctuated a warm interval. 20

Conclusions
In this paper, we present the most recent chronology, named NALPS19, for δ 18 Ocalc variability as recorded in speleothems that grew during the last glacial period between c. 15 and 120 ka along the northern rim of the Alps.
In particular, we have updated the record between 63.7 to 118.3 ka, using eleven cleaner, more accurately and precisely dated samples from five caves. Over the 63.7 to 118.3 ka interval, the record is now 90% complete. 25 Ramp-fitting analysis of the transitions between stadials and interstadials shows that δ 18 O shifts in the North Atlantic realm occurred on multi-decadal timescales, whereas transitions in the Asian monsoon occurred on multi-centennial timescales. Further, the absolute timing of shifts show good agreement between NALPS19 and Greenland ice core chronologies within the multi-millennial-scale ice core uncertainties, though absolute offsets are often on multi-decadal to multi-centennial scales. Major differences do, however, arise when comparing 30 NALPS to NGRIP on AICC2012 between 100 to 120 ka, suggesting that the AICC2012 chronology is too young by c. 3,000 years in this time period. Additionally, we propose that the duration of the highly-debated GS-22 -GI-21.2 -GS-21.2 interval was 3,993 ± 155 years, which is in closer agreement to the duration of 4,122 ± 650 years in NGRIP-EDML (Capron et al., 2010b) and the 4,489 ± 960 years of the Asian monsoon composite record (Kelly et al., 2006;Kelly, 2010;Cheng et al., 2016). Preliminary investigation of the trends in mean 35 δ 18 Ocalc as recorded in the NALPS speleothems for different interstadiala and stadials reveals that for a given time period, as expected, δ 18 Ocalc becomes lighter with increasing distance from the source and increasing elevation. Exceptions are found at one high-elevation site, which appears to record a stronger summer signal.
Finally, our accurate and precise chronology enables a deeper investigation of centennial-scale cold reversals that occurred either as 'precursor events' (i.e., GS-16.2, GS-17.2, GS-21.2, GS-23.2;Capron et al., 2010a) or during interstadials (i.e. GS-24.2). Each of these events occurred in the decades and centuries following rapid rises in sea level of over 12 m kyr -1 (Grant et al., 2012) that occurred coincident with IRD events (McManus et al., 1999) and shifts in the ITCZ causing speleothem growth in Brazil (Wang et al., 2004). We therefore propose that these centennial-scale cold reversals are products of freshwater discharge into the North Atlantic during 5 times of moderate ice sheet size, which caused a slowdown of the AMOC and associated atmospheric cooling, similar to deglacial events such as the Preboreal Oscillation or Older Dryas.

Data availability
The stable isotope data both on distance along growth axis and OxCal age models are available at both SISAL and the US National Oceanic and Atmospheric Administration (NOAA) data center for paleoclimate 10 (speleothem site) at the following address: TBC

Author contribution
GM undertook the majority of the U-Th analyses, interpreted the data, and wrote the manuscript. CS conceived the project and carried out field work together with GM and partly SB. SB undertook additional U-Th analyses, prepared and ran Hendy tests and stable-isotope samples. TE developed and ran ramp-fitting models. ML 15 provided data. RLE provided analytical U-Th facilities. All authors directly contributed to the manuscript through discussion or writing.

Competing interests
The authors declare that they have no conflict of interest.       (2012)