A global compilation of diatom silica oxygen isotope records from lake sediment – trends, and implications for climate reconstruction

. Oxygen isotopes in biogenic silica ( δ 18 O BSi ) from lake sediments allow for quantitative reconstruction of past hydroclimate and proxy–model comparison in terrestrial environments. The signals of individual records have been attributed to different factors, such as air temperature (T air ), atmospheric circulation patterns, hydrological changes and lake evaporation. While every lake will have its own set of drivers of d18O, here we explore the extent to which regional or even global signals 45 emerge from a series of palaeoenvironmental records. For this purpose, we have identified and compiled 71 down–core records published to date and complemented these datasets with additional lake basin parameters (e.g. lake water residence time and catchment size) to best characterize the signal properties. Records feature widely different temporal coverage and resolution ranging from decadal–scale records covering the last 150 years to records with multi–millennial scale resolution spanning glacial–interglacial cycles. Best coverage in number of records (N=37) and datapoints (N=2112) is available for northern 50 hemispheric (NH) extra–tropic regions throughout the Holocene (corresponding to Marine Isotope Stage 1; MIS 1). To address the different variabilities and temporal offsets, records were brought to a common temporal resolution by binning and subsequently filtered for hydrologically open lakes with lake water residence times <100 yrs. For mid– to high–latitude (>45° N) lakes, we find common δ 18 O BSi patterns during both the Holocene and the Common Era and maxima and minima corresponding to known climate episodes such as the Holocene Thermal Maximum (HTM), Neoglacial Cooling, Medieval 55 Climate Anomaly (MCA) and the Little Ice Age (LIA). These patterns are in line with long–term T air changes supported by previously published climate reconstructions from other archives as well as Holocene summer insolation changes. In conclusion, oxygen isotope records from NH extratopic lake sediments feature a common climate signal at centennial (for CE) and millennial (for Holocene) time scales despite stemming from different lakes in different geographic locations and constitute a valuable proxy for past climate reconstructions.


Scientific background
Oxygen isotopes are ubiquitous within the global water cycle and are among the best (hydro)climate proxies worldwide as a result of their potential quantitative interpretability. The most abundant isotope 16 O and the rarer isotope 18 O are subject to fractionation during water phase transformation and transport processes (Fig. 1). As a result, the relative abundance of these 65 isotopes varies across space, time and reservoirs. The IAEA's standard Vienna Standard Mean Ocean Water (VSMOW) serves as a baseline, closely resembling the isotopic composition of modern seawater. Relative abundance of 18 O with regard to 16 O is expressed relative to the VSMOW standard as δ 18 O and given in units of permil.
Measuring δ 18 O in water, vapor, snow and ice allows for quantifying hydroclimatic processes in the global water cycle and has been used to this end for many decades (Dansgaard, 1964;Konecky et al., 2020). In addition to its use in present-day 70 hydrology, δ 18 O in environmental archives also serves as a powerful proxy for past hydrology and, in turn, climate. As such, δ 18 O is a crucial and quantitative tool for both reconstructions of past climate and for comparison of proxy data to climatemodel outputs.
The predictability of isotope fractionation processes in the water cycle due to clear physical constraints allows for using δ 18 O not only as a quantitative tool for past (hydro)climate reconstructions, but also for implementing δ 18 O in global climate models 80 (e.g. Danek et al. 2021). However, challenges persist in understanding and comparing data and model outputs, partly due to complex signal formation in environmental archives (Danek et al., 2021). Lake sediments are prominent and widespread archives in terrestrial environments (Leng, 2006;Biskaborn et al., 2016;Subetto et al., 2017). Like glaciers and ice caps, lakes may record the isotopic signature of past precipitation (Shemesh et al., 2001a).
The signal formation in lakes, however, is fundamentally different from that in glaciers and ice caps with the lake water body 85 acting as a buffer of the enclosed (hydro)climate signal, influenced by factors such as, catchment hydrology, lake ontogeny, and sediment accumulation rates (Bittner et al., 2021). While biogenic silica is also found in phytoliths, sponge spicules and chrysophytes, this work focuses exclusively on diatoms. Diatoms are microscopic algae synthesizing their frustules from lake water and the silica dissolved therein, thereby preserving the isotope signal of past lake water. The oxygen isotope composition of the biogenic silica of diatom frustules (δ 18 OBSi) buried in lake sediment therefore provides valuable insight on past lake 90 water and, in turn, precipitation.
Naturally, the interplay of hydrological and sedimentological processes limits the time span and resolution over which environmental information can be obtained from a lake system, and this varies from lake to lake. Most commonly, δ 18 OBSi has https://doi.org/10.5194/cp-2022-96 Preprint. Discussion started: 24 January 2023 c Author(s) 2023. CC BY 4.0 License. been applied to lake sediments devoid of carbonates. These lakes are especially common in high altitude and high latitude regions (Leng and Barker, 2006). 95

Controls on δ 18 OBSi
As a result of the complex signal formation in δ 18 OBSi records, most authors refer to a combination of factors for interpreting a given individual record. A schematic overview of some of the main processes is provided in Fig. 1. Lake water temperature (Tlake) imparts a direct effect on δ 18 OBSi due to temperature-dependent fractionation during biosynthesis of biogenic silica. The temperature-dependent fractionation coefficients were initially determined for marine environments 105 (Leclerc and Labeyrie, 1987;Matheney and Knauth, 1989) and later adapted for lacustrine sediments, as well. The effect of Tlake amounts to ca. -0.2‰/°C during biosynthesis (Dodd and Sharp, 2010;Moschen et al., 2005). However, the applicability of these calibrations to sedimentary records is debated due to possible disequilibrium between diatom frustules and lake water, as well as taphonomic and diagenetic processes (Leng and Barker, 2006;Ryves et al., 2020;Smith et al., 2016;Tyler et al., 2017). 110 More commonly, changes in δ 18 OBSi are attributed to δ 18 O of lake water (δ 18 Olake) and, with the lake as a buffer, to the isotopic composition of precipitation (δ 18 Oprecipitation). Locally, δ 18 Oprecipitation is primarily influenced by air temperature (Tair) with a fractionation of 0.7‰/°C on global average (Dansgaard, 1964). On a broader scale, parameters such the moisture source region and atmospheric circulation patterns are important. Air masses transported to a lake are linked to their trajectories and source regions. Hence, changes in δ 18 OBSi records may be indicative of changes in atmospheric circulation patterns, also influencing 115 the local ratio of precipitation to evaporation (P/E). The parameter P/E refers to the balance between the amount of precipitation and evaporation within a lake and its catchment. Typically, water loss through evaporation leads to higher δ 18 Olake values due to evaporative enrichment, and therefore higher δ 18 OBSi. Yet, precipitation amount and seasonality (e.g., rain vs. snow) will also strongly impact the isotope signal of the lake water. Disentangling the effects of precipitation and evaporation in δ 18 OBSi records is thus a major challenge and therefore some authors commonly refer to P/E-changes in conjunction with other factors 120 (Hernandez et al., 2013;Rosqvist et al., 2013;Broadman et al., 2020a). Shemesh et al. (2001a) have attributed the δ 18 OBSi signal changes of Lake 850 in Swedish Lapland to different contributions of main source regions of precipitation, namely the Atlantic and Arctic Oceans. Indeed, for this location, changes in atmospheric circulation patterns imply changes in both, Tair and humidity.
Insolation is not strictly speaking an environmental parameter which can be inferred from δ 18 OBSi records. It is, however, 130 commonly used as an explanatory variable and underlying driver of Holocene climate change. Insolation can be calculated for any given place throughout time (Laskar et al. 2004), which allows for direct comparison with δ 18 OBSi time series. Insolation primarily influences Tair and, indirectly, the P/E balance and atmospheric circulation patterns. Moreover, hydrological processes within a lake's catchment may substantially impact δ 18 Olake and hence, δ 18 OBSi. Some of these processes are closely linked to climate and precipitation patterns described above. Most notable are variations in the 135 amount of snow melt (Mackay et al., 2013;Rosqvist et al., 2013;Meyer et al., 2022) and glacial influx (Meyer et al., 2015a) reaching the lake. Glaciers or snow fields in the hinterland, may provide more melt (with lower/depleted δ 18 O) in warm phases -countering the influence of warming on δ 18 Oprecipitation (Meyer et al., 2015a).
Other hydrological processes, however, are only indirectly linked to climate. These include the formation and closure of outflows to the lake which change the lake's hydrological setting (Hernandez et al., 2008;Vyse et al., 2020;Bittner et al., 140 2021). Associated changes in mass-balance of a lake and, thus, P/E-balance may lead to a different δ 18 OBSi signal.
In order to interpret a given δ 18 OBSi, information about the present hydrology is important as it provides insight on which processes influence δ 18 OLake and, in turn, δ 18 OBSi. Naturally, as the hydrology of a given lake can only offer snapshots of present-day constraints, caution has to be applied when extrapolating these into the past. The isotopic composition of three components is of interest for assessing the signal properties of δ 18 OBSi records: (1) δ 18 O lake water, (2) δ 18 O of inflows to the 145 lake and (3) δ 18 O of precipitation. Lake water composition can provide insight on whether lake water is isotopically homogenous, both spatially and at different depths of the lake basin.
Offsets of δ 18 OLake from the Local Meteoric Water Line (LMWL) give hints on whether or not lake evaporation may have had a significant effect on lake water isotopic composition. δ 18 OLake can also be used in conjunction with Tlake and the most recent δ 18 OBSi value for testing the temperature-dependent fractionation and, in turn, the applicability of the proxy for paleoclimate 150 reconstructions. Complementing the lake water isotopic composition with the isotopic composition of inflow (rivers) provides further information on specific hydrological constraints. For larger lakes and catchments, the different inflows and their effect on the lake's isotopic mass balance have been used for inferring changing hydrology or precipitation regimes in different parts of the catchment (e.g. Mackay et al. 2011).
While the isotopic composition of precipitation provides useful information, it is often not available for field studies and few 155 monitoring studies of precipitation and lake water isotope compositions in combination with δ 18 OBSi do exist (Kostrova et al., 2019;Hernandez et al., 2010). A possible workaround can be derived from the GNIP database (IAEA/WMO, 2022) and spatial interpolations thereof, e.g. The Online Isotopes in Precipitation Calculator (Bowen, 2017).

Sample preparation and measurement 160
Extracting pure diatom valves (frustules) from lake sediments is a complex process and depends on a sufficiently high diatom concentration in the sediment. Various cleaning procedures for diatom extraction exist, which are mainly based on the use of H2O2 for removing organic matter from the sediment and HCl for the removal of carbonate (Chapligin et al., 2012a). Detrital components are often separated by being centrifuged in heavy liquid solution, i.e. Sodium Polytungstate (Morley et al., 2004;Chapligin et al., 2012a). Different procedures in sample preparation, i.e. different temperatures, may result in offsets of 165 measured δ 18 OBSi (Swann et al., 2010;Chapligin et al., 2012b;Tyler et al., 2017). Impurities due to incomplete removal of https://doi.org/10.5194/cp-2022-96 Preprint. Discussion started: 24 January 2023 c Author(s) 2023. CC BY 4.0 License. detrital components may also affect the measured δ 18 OBSi (Lamb et al., 2005;Chapligin et al., 2012b). To account for this, the purity of samples is usually assessed prior to measurement by visual inspection or direct measurement of the amount of contaminants of a given sample (Bailey et al., 2018;Broadman et al., 2020a). Fewer studies determine the isotopic composition of detrital contaminants to apply a correction to the measured δ 18 OBSi values (Brewer et al., 2008;Mackay et al., 2011;Wilson 170 et al., 2014a;Bittner et al., 2021;Kostrova et al., 2021). Consequently, caution has to be applied when comparing absolute values of different individual δ 18 OBSi records, as offsets between individual records may result from both different preparation and measurement techniques and different isotopic signals of the corresponding lake water. The potential and challenges associated with data stemming from different preparation and measurement techniques have already been addressed (Chapligin et al., 2011;Mackay et al., 2011). 175

Aim of this work
A comprehensive compilation and assessment of the lake sediment δ 18 OBSi records published to date is missing. Due to the crucial role of lake and catchment hydrology in signal formation, such a compilation needs to include individual lake basin parameters, such as lake volume and lake water residence time (tres), in order to provide reliable constraints for interpreting 180 and comparing the individual records. Proxy data compilations have already addressed a lack of standardized metadata, data availability and data uniformity of paleo-data (Pfalz et al., 2021). Such compilations, however, generally do not include δ 18 OBSi-records and do not provide all relevant lake basin parameters  or have a more limited temporal focus (Konecky et al., 2020). Consequently, no study has yet empirically linked the signal properties of δ 18 OBSi records and lake basin parameters in a harmonized dataset. 185 In order to overcome these gaps, this paper aims at providing a comprehensive compilation and combined statistical evaluation of the lake sediment δ 18 OBSi-records published to date. We accomplish these objectives by means of the following working steps: 1) collecting available lake sediment δ 18 OBSi-records published to date, 2) complementing these records with the individual lake basin parameters and 3) assessing the signal properties of δ 18 OBSi records with regard to lake basin parameters, in order to 4) identify common spatio-temporal patterns and trends in the δ 18 OBSi signals. This effort shall lead to a better 190 understanding of the general constraints for interpreting lake sediment δ 18 OBSi-records and make these records more readily usable also for proxy-model comparisons. It is, hence, a contribution to bridge the gap between modelling and isotope geochemistry approaches in paleoclimate science.

Methods 195
This study follows a three-stage approach: the first stage (data acquisition) comprises identifying lacustrine δ 18 OBSi datasets and publications published to date. A second stage includes acquiring the actual δ 18 OBSi datasets and archiving them in a standardized format. Where appropriate, the identified published datasets were arranged into longer continuous records. This includes records from the same sediment core published in different manuscripts and records from different sediment cores from the same lake but similar locations. The third and final stage (record analysis) interprets the actual lacustrine δ 18 OBSi 200 isotope records with respect to their hydrological and geographical constraints in order to identify possible common trends and signal properties of all isotope records or smaller spatially or temporally constrained datasets.
Identification and acquisition of the datasets and publications in this work followed an additive approach, i.e. thorough literature survey. We have chosen this approach due to the limited number of laboratories and working groups worldwide measuring δ 18 OBSi. In this study, we focused exclusively on down-core records from lacustrine sediments. Identified 205 publications and datasets were entered into a uniform metadata table giving one entry to each publication and each dataset. If more than one publication was written about the same sediment core, each of these publications was given a separate entry.
Likewise, if one publication presents data from more than one sediment core, each of the sediment cores was given a separate entry to the database.
For each of these entries, metadata on coring procedure, hydrological setting and chronology were supplemented. Data were 210 extracted from the corresponding publication(s), from public repositories or directly from the authors. Hydrological parameters such as catchment area, average depth and tres were additionally obtained from the HydroLAKES database (Messager et al., 2016) and stored as separate variables. For a detailed description on how the values in the HydroLAKES database were created, we refer to Messager et al. (2016). This procedure was necessary because original publications do not always specify all of these parameters and because parameters supplied in original publications may not be consistent. For further analysis and for 215 linking the individual isotope records with lake basin parameters, the HydroLAKES database entries were given preference.
Original publications were only used for lakes where no data were available from the HydroLAKES database. The parameter "maximum water depth" was always taken from the original publications since this parameter is not provided by the HydroLAKES database. While the geo-statistical approach of the HydroLAKES database may not provide the most precise values for individual lake basin characteristics, it ensures comparability among the different lake basins analyzed in this study. 220 It also prevents potential issues arising from different authors providing conflicting values for the same lake basin (i.e. Lake Baikal, Lake El'gygytgyn).
Isotope datasets for each of the entries were archived in separate tables specifying depth, age and measured isotope values.
Datasets were taken directly from the respective publication, whenever possible. If datasets were not available, within the original publications or as supplements thereof, public repositories (such as www.pangaea.de; ncei.noaa.gov) were searched. 225 In case datasets were unavailable from repositories, the lead authors were contacted directly. In case of published data unavailable from repositories or authors, plots of the original publications were digitized, if possible (Rietti-Shati et al., 1998;https://doi.org/10.5194/cp-2022-96 Preprint. Discussion started: 24 January 2023 c Author(s) 2023. CC BY 4.0 License. . Where digitizing plots was not feasible, records were excluded from further consideration; this was the case for Chondrogianni et al. (2004) and .
Chronologies were adapted from the original publications, where available and stored in cal yrs BP format (relative to 1950 230 CE). Sample ages given in different formats (e.g. b2K or CE) were converted to cal yrs BP. This procedure also applies to radiocarbon-based chronologies. Chronologies were not recalculated because the effect of different 14 C-calibrations and different age model approaches is supposedly minor with regard to the aim of this study. This is especially true when considering the fact that high-resolution datasets (annual to decadal resolution) are the exception among all datasets identified.
The error introduced by the different age models is, thus, considered minor when compared to the resolution of the datasets. 235 For applications requiring more precise chronologies and better comparability of datasets, we provide the radiocarbon measurements of respective datasets as well. This enables future users to create tailor-made chronologies if needed. Datasets without chronologies were stored using depth notation only.
For further analysis, datasets were partly regrouped and combined to generate longer continuous δ 18 OBSi datasets, henceforth referred to as «records». This comprised treating datasets with data from several coring sites or outcrops from the same lake 240 as one single record, in agreement with the author's original interpretation (Quesada et al., 2015;Swann et al., 2018). Data stemming from the same sediment core but published in different manuscripts were combined into single continuous records.
Likewise, data from different cores presented in different publications but stemming from identical or reasonably similar coring 245 sites were combined to single records. This applies to data from Lake Nar (Dean et al., 2018) and Vuolep Allakasjaure (Rosqvist et al., 2004;Jonsson et al., 2010). Lakes consisting of several basins were generally treated as one; this includes Lake Baikal and several smaller lakes.
For trend analysis and comparison of records with regard to Holocene and Common Era climate, we focused on records from northern hemisphere (NH) extratropic lakes (45-90° N) because these time periods and geographical focus are the only options 250 with a significant number of records which makes an interpretation departing from case studies feasible.
In a first step, records were binned to 1 kyr intervals (for the Holocene) and 200 year intervals (for the CE), respectively. These intervals were chosen based on the temporal resolutions of the original records in order to ensure continuous binned records with no empty bins. Choosing higher resolutions would have resulted in many records having empty bins with no datapoints, thus leading to different temporal resolutions even after binning. The binning was done by calculating the mean value of all 255 samples within the respective age interval for each individual record. For the analysis of the Holocene period, datapoints with ages <150 yrs BP were excluded to eliminate any possible effects of recent warming. Analysis of Common Era climate on the other hand makes use of these datapoints. For an assessment of the timing of Holocene maxima and minima, records covering less than 10 bins were discarded. For trend analysis of Holocene and Common Era climate, records covering less than seven bins were discarded. 260 https://doi.org/10.5194/cp-2022-96 Preprint. Discussion started: 24 January 2023 c Author(s) 2023. CC BY 4.0 License.
In order to eliminate offsets between individual binned records, a mean removal was performed by subtracting the mean of a given binned record from the respective record. After mean removal, two subsequent filtering steps were performed in order to exclude records most prone to secondary lake evaporation. First, a subset consisting only of records from open lakes was created, discarding records from semi-closed lakes, closed lakes and paleo-lakes. In a second step, this open-lakes-subset was filtered by discarding all records with tres>100 years. This threshold was chosen in accordance with previous works 265 classifying lakes and their isotopic signals with respect to hydrologic setting and tres (Leng and Marshall, 2004).
After applying these filters, combined trends for geographical regions (NH, Eurasia, North America) were created by calculating the mean of all records for each bin.

Published datasets 270
Following a thorough literature survey, a total number of 71 published down-core datasets of δ 18 OBSi from 64 sites has been identified. Ever since the first records of δ 18 OBSi from lake sediments have been published (Rietti-Shati et al., 1998;Shemesh and Peteet, 1998), there has been a growing research interest, manifest in an increasing number of publications (Fig. A1). Only recently has there been a stagnation of the number of records published. A detailed overview of the identified publications and the records presented therein is provided in Tab. A2. 275 Most publications give the δ 18 OBSi data as time series, as is usual in paleoenvironmental studies. The chronologies rely on a wide variety of dating methods (Fig. A2), with 14 C (used on 46 records) being by far the most frequent. For subrecent time periods and shorter time scales, 210 Pb and 137 Cs are also used extensively (on 21 and 12 records, respectively). Globally or regionally correlated time markers such as tephra layers (Heyng et al., 2015) or high-resolution methods such as varve counts (Rozanski et al., 2010) are used relatively sparsely as they are not available for all time periods and/or lake basins. 280

Combined records
We combined separate records stemming from the same lake sites to longer, continuous composite records and yielded a total of 54 combined δ 18 OBSi records, seven of which stem from paleolakes (three with and four without chronology) while 49 records stem from present, still existing lakes (44 with and five without chronology). An overview of these combined records 285 is provided in Tab. A1, complemented by metadata of the records and the corresponding lake basins. Each individual record received a number (#X) which is used consistently throughout the text.
due to the availability of lake sediments, diatom abundances in a lake and/or scientific foci of the research groups. By far the highest number of records is available for MIS 1 (N=37) and especially for the last 2 kyrs (N=48). MIS 2 is still covered by 19 records, whereas towards MIS 3 (N=5) and beyond fewer records have been generated. However, single records cover even time periods in MIS 11 (at Lake Baikal; #2) and beyond (MIS G1/104, #1), outlining the applicability of this proxy across a 295 wide range of time scales. A paleo lake from the Baringo-Bogoria Basin (record #1), is the oldest lacustrine δ 18 OBSi record and has been dated to the onset of MIS 104 in the late Pliocene (Wilson et al., 2014b).
Regarding the number of published records by continent (Fig. 3), there is a clear focus on Asia and Europe with 21 and 15 records published, respectively. In Asia, there is a strong regional focus on Siberia, whereas most other parts of the continent are not yet investigated for lake sediment δ 18 OBSi. A regional focus also occurs in Africa and South America. While a sizeable 300 number of records stemming from these continents have been published (N=10 and N=11, respectively), there are pronounced regional foci in the East African Rift and the Andes, respectively. Another focus region is Alaska, where most (N=6) of the published North American records (N=7) are located. δ 18 OBSi work has also been carried out in Oceania with published records from lakes in New Zealand (#42) and South Georgia (#19).
In summary, the distribution of available δ 18 OBSi data displays, thus, a bias towards the mid-to high latitudes of the northern 305 hemisphere. Additionally, this distribution is also indicative of the site accessibility and (regional) research foci of the working groups. This has a rather pronounced effect on the spatial distribution of available δ 18 OBSi records. The geographical distribution is also indicative of cold regions devoid of carbonates where biogenic silica is the most promising archive to obtain oxygen isotope records from lake sediments.
This pattern also affects the latitudinal distribution of the records (see also Tab. A1) with records ranging from 54.17 °S (# 19) 310 to 69° N (# 39). Records stemming from low-latitude lakes (particularly in Africa and South America) are often located at high altitudes above 3000 m asl (above sea level; e.g., #11, 17, 18, 21). Lake Simba Tarn (record #34) features the maximum altitude for an individual δ 18 OBSi record of 4959 m asl. Altitudes below 1000 m asl, however, are most common (N=36), especially for high-latitude lakes. In total, 12 lakes are located below 100 meters asl (e.g., #29, 35). Many of these low-altitude lakes are located in maritime locations in immediate proximity to the coast (e.g., #19, 30) or have even had marine intrusion 315 stages in the past (a postglacial transgression at Nettilling Lake; #41). Extremely continental environments are tackled in central and eastern Siberia and the Lake Baikal region (e.g., #4, 8, 9, 10, 25).

Spatial coverage and resolution of combined records
Varying spatial resolution among the compiled isotope records is linked to the fact that lakes and their corresponding 330 catchments are not points in space but areas. Consequently, lake water and, thus, the related δ 18 OBSi-signal represent a spatial average, integrating over the catchment size of the respective lake. Lake basin sizes range from small ponds of <1 km² (e.g. #19, 11, 23, 33, 34) to some of the largest lakes in the world, including lake Malawi and Lake Baikal covering 29,544 km² and 31,968 km², respectively (Fig. 4A). Large and voluminous lakes are the exception, however, and most of the lakes represented here are <10 km² in size (N=25). Note that in case of paleo lakes, such parameters are not applicable. While authors sometimes 335 provide estimates on paleo lake extent, these figures are not consistent with the values determined for present lakes and are https://doi.org/10.5194/cp-2022-96 Preprint. Discussion started: 24 January 2023 c Author(s) 2023. CC BY 4.0 License.
therefore not included in the evaluation of the dataset. Catchment sizes vary by several orders of magnitude as well. For small lakes, catchments are often <10 km² (N=13, e.g. #22, 33, 37), whereas the largest lakes, Lake Malawi and Lake Baikal feature catchments of 128,727 km² and 569,176 km², respectively (Fig. 4A). Therefore, these lakes integrate the environmental signal over a large and potentially diverse region in terms of hydrology. While most of the lakes compiled in this study are primarily 340 fed by surface runoff and/or precipitation (according to the original publications of records), groundwater influx may also play a pivotal role. Groundwater influx may introduce a large memory effect of past precipitation due to the generally long residence times of aquifers. This may have an impact on records especially when looking at short timescales. Groundwater input, however, is usually not accounted for and, thus, beyond the scope of our study. In summary, both lake sizes and catchment sizes vary by several orders of magnitude ( Fig. 4A) among the sites with existing δ 18 OBSi records, and span from local signals 345 to regional averages. However, most of the records (N=18) stem from lakes with catchments <100 km², suggesting rather local signals. While single local signals represent small areas, different local signals may well correlate on continental and hemispheric scales. to more than one records. Note that records from Paleo-lakes and lakes with incomplete information on lake and catchment size were not considered for this figure (N=33). B) Depiction of corresponding sampling intervals and tres. Note that records from lakes without information on tres were not considered for this figure (N=37).

Temporal coverage and resolution of combined records
Temporal resolution of the records and the resulting signal properties are determined by both the lake basin itself 355 (i.e. accumulation rates and preservation of diatom silica) and the sampling routine applied to the sediment core (i.e. the properties could not be investigated in detail in this manuscript. We therefore focus on tres and sampling frequency of a core 360 as a means of characterizing the records and their temporal resolution. The tres is closely linked to the size of lake basins and varies accordingly. It ranges from several weeks for small lakes (e.g., #49, 50) to centuries (219 and 375 years for Lake Malawi and Lake Baikal, respectively). On the whole, tres of sub-annual to annual scale are most common (N=17) among the lakes considered in this study (Fig. 4B). It should be noted that longer tres leads to averaging of the signal over a longer time period. Likewise, different tres may also have an effect on absolute δ 18 OBSi values 365 and variabilities. Lakes with very long tres (tres>100 years) are tendentially more susceptible to lake evaporation, and may effectively behave like closed-system lakes with regard to the δ 18 OBSi signal even if they are hydrologically open (i.e. have outflows). These are still referred to as hydrologically open in this work and hydrological settings and tres are addressed separately in the discussion.
Also the sampling frequency varies by several orders of magnitude (Fig. 4B). It stretches from annual to multi-millenial 370 timescales. Consequently, the signal properties and the recorded climatic and/or hydrological forcings of the identified records can be expected to vary accordingly. Most records, however, plot in the top left half of graph of Fig. 4B, which indicates that the temporal offset between two sediment samples exceeds tres. This suggests that the sampling routine is the limiting factor of the temporal resolution of these records. Only three records from Lake Baikal plot in the lower right half of Fig. 4B, suggesting in these cases (#4, #20, #51) tres to be the dominant factor in determining the record's resolution, at least with respect to inflow-375 related changes.
We thus conclude that the sampling resolution is the main factor determining the temporal resolution of most records with tres acting as an additional smoothing mechanism. A lake with a centennial-scale tres can, therefore, display centennial-scale changes of climate and hydrology to their full amplitude. Decadal-scale changes, on the other hand, can be expected to be attenuated by an order of magnitude in a lake with centennial-scale tres. Based on tres and sampling resolutions of the records, 380 comparing records on a centennial or millennial scale is the most promising approach for assessing common patterns.

Common Era Climate
The Common Era subset of records including only records stemming from sites north of 45° N (Fig. 5A, N=19), comprises 385 460 datapoints. The records display considerable offsets with δ 18 OBSi values ranging from +19‰ (#34) to +33‰ VSMOW (#35). These offsets can be linked to their individual environmental setting, e.g. latitudinal and continentality effects, as well as to their potential to be prone for lake evaporation. Amplitudes in δ 18 OBSi of these records vary from 0.4‰ (#24) to circa 9‰ https://doi.org/10.5194/cp-2022-96 Preprint. Discussion started: 24 January 2023 c Author(s) 2023. CC BY 4.0 License. VSMOW (#35). This variability might correspond to their different hydrological settings. Closed lakes (such as Sunken Island Lake; #26) have a tendency towards higher δ 18 OBSi amplitude likely due to lake evaporation leading to a more enriched isotope 390 signature in the lake (Broadman et al., 2020b).
Binned records (Fig. 5B, N=14) do not display any common pattern and feature δ 18 OBSi-values ranging from +20‰ to +31‰ VSMOW. However, binning of datapoints and exclusion of shorter records reduces the δ 18 OBSi amplitudes of individual records, to less than 5‰. Offsets between individual records may be linked to site-specific characteristics of individual lakes and catchments, as well as latitudinal and continentality effects. 395 After mean removal (Fig. 5C)  The combined Common Era records (Fig. 5F) comprise 8 records and show a general δ 18 OBSi decrease over the last 2 kyr 400 amounting to ca. 2‰ VSMOW. The δ 18 OBSi amplitude within individual bins, however, might exceed this number, most notably at 100, 1300 and 1900 yrs BP, which show the highest amplitudes of up to 5‰ for the Common Era. This is in line with differences between the timing of δ 18 OBSi maxima and minima between North American and Eurasian records. At centennial scale, these differences might also be linked to dating uncertainties. Moreover, the record of Lake Bolshoye Shchuchye (#16) features a much larger amplitude. Its exceptional δ 18 OBSi variability has been attributed by the authors to 405 variations of snow and snow melt in the lake's catchment and therefore does represent a different kind of precipitation-based signal compared to most other records. The combined trend features its highest δ 18 OBSi values at 1900 yrs BP, followed by a decrease leading to a relative minimum at 1100 and 1300 yrs BP. This phase is followed by a second δ 18 OBSi peak at 700 and 900 yrs BP, after which a decrease can be observed. The decrease at 300 and 500 yrs BP features the least deviation between individual records. The most recent bin shows again a much larger δ 18 OBSi variability. While most records display the lowest 410 values at this time, Lake Kotokel (#8) for instance shows an increase compared to the previous bins. This increase may or may not be related to recent warming; and due to the complex hydrology of the lake even lake evaporation cannot be ruled out. The majority of the records however does not indicate a recent δ 18 OBSi maximum, indicative for any impact of recent warming. It has to be noted that there is bias of data points towards the older end of the most recent bin. An absence of recent warming in 415 https://doi.org/10.5194/cp-2022-96 Preprint. Discussion started: 24 January 2023 c Author(s) 2023. CC BY 4.0 License. the data therefore might be linked to this bias and does not rule out a reaction of lake systems to recent warming. As Eurasian records constitute the majority of records (N=7), the Eurasian trend is similar to the NH picture (Fig. 5H).

Figure 5. Common Era Northern hemispheric records (45-90 N) compiled in this study. A) Original records, B) data binned to 200 year intervals, only showing records covering at least 7 bins, C) binned records with mean of individual records removed, D) filtered for records from open lakes only, E) filtered for records from lakes with tres<100 yrs, F) NH trend, calculated as mean of all records in each bin. Shadings show 1 and 2 standard deviations, respectively. G) North American and H) Eurasian
North American records (N=2) do not show a consistently decreasing trend (Fig. 5G), but slightly higher values at 1700 and 1900 yrs BP compared to the most recent bins (100 and 300 yrs BP, respectively). They do, however, show lower δ 18 OBSi 425 values between 900 and 1300 yrs BP, followed by a δ 18 OBSi maximum at 500 yrs BP. As there are only two records available after filtering, caution has to be applied in interpreting this pattern.
Common patterns among Eurasian, and possibly NH, δ 18 OBSi records do suggest a common NH signal throughout the Common Era. The observed maxima and minima correspond to previously described climatic events, notably the Roman Warm Period from 2000 yrs BP to about 1600 yrs BP (Ljungqvist, 2010), the Dark Ages Cold Period from 1600 to 1200 yrs BP (Büntgen 430 et al., 2016;Helama et al., 2017) the Medieval Climatic Anomaly (MCA) from 1200 yrs BP to 800 yrs BP (Bradley et al., 2003;Mann et al., 2009) and the Little Ice Age from 700 yrs BP to 100 yrs BP (Matthews and Briffa, 2005). The good accordance of our data with these previously described warm and cold phases suggests that δ 18 OBSi records presented are influenced by Tair change, either directly or via other parameters which are linked to Tair. Recent research has rejected the global nature of these climatic events and suggested that they are regionally constrained (Neukom et al., 2019). However, the 435 accordance of our data -when assessed at centennial scale -with these climatic events makes sense because they were initially described for North America and Europe.
Amplitudes of the individual Holocene records vary from less than 5‰ (#3) to circa 10‰ VSMOW (#20, #35). Lake 445 El'gygytgyn (#3) is a voluminous, deep hydrologically open lake with little lake evaporation, whereas Heart Lake (#35) is a much smaller, less voluminous lake. Lake Baikal (#20) is a deep voluminous lake like Lake El'gygytgyn, however with a centennial-scale tres making it effectively closed and thus potentially subject to secondary evaporation. Consequently, differences in their isotopic variability might correspond to their different hydrological settings. Closed lakes (such as Sunken Island Lake; #26) have a tendency towards higher δ 18 OBSi amplitude likely due lake evaporation leading to a more enriched 450 isotope signature of lake water (Broadman et al., 2020b). https://doi.org/10.5194/cp-2022-96 Preprint. Discussion started: 24 January 2023 c Author(s) 2023. CC BY 4.0 License. Binned records (Fig. 6B, N=14) do not display any common pattern and feature δ 18 OBSi-values ranging from +20‰ to +35‰

Figure 6. MIS1 Northern hemispheric records (45-90 N) compiled in this study. A) Original records, B) data binned to 1-kyr intervals, only showing records covering at least 7 bins, C) binned records with mean of individual records removed, D) filtered for records from open lakes only, E) filtered for records from lakes with tres<100 yrs, F) NH trend, calculated as mean of all records in
VSMOW, similar to the original records. However, binning of datapoints and exclusion of shorter records results in smoothed amplitudes of less than 5‰ VSMOW. Early Holocene binned records seem to have δ 18 OBSi values (up to 2.5‰) higher than 460 the Holocene mean, whereas late Holocene values are generally up to 2‰ below the Holocene mean. However, some records do not follow this general pattern after mean removal (Fig. 6C). This could be indicative of the fact that the dataset still features lakes with different hydrological settings and tres which have a substantial impact on the recorded signal.
Filtering for hydrologically open lakes (Fig. 6D, N=13) and tres<100 yrs (Fig. 6E, N=10)  Based on this subset of hydrologically open lakes with tres<100 yrs, a combined millennial scale NH trend was calculated. This 470 combined Holocene trend (Fig. 6F) shows a decrease throughout the Holocene amounting to ca. 2‰ VSMOW. The absolute maximum is observed for the Early Holocene bin at 11-12 kyr BP. A decrease at the beginning of the Holocene between 12 kyr BP and 10 kyr BP is followed by a relatively stable middle Holocene until 6 kyr BP and subsequent stronger decrease towards the absolute δ 18 OBSi minimum in the youngest bin (0-1 kyr BP).
When considered separately, North American (N=2) and Eurasian (N=8) records show different patterns and have been 475 described and interpreted differently in the case studies published to date. While the former are primarily liked to atmospheric circulation changes in Alaska (Bailey et al., 2018;Broadman et al., 2020b;Broadman et al., 2020a), the latter are at least partly interpreted as indicative of Tair and insolation changes (Swann et al., 2010;Mackay et al., 2011;Chapligin et al., 2012b;Kostrova et al., 2013a;Kostrova et al., 2019;Kostrova et al., 2021). Eurasian δ 18 OBSi records (N=8) display a slight decrease from 12 kyr BP to 10 kyr BP and suggest a second maximum at 7 kyr BP followed by a second relatively stable phase. An 480 accelerated decrease starting at ca. 4 kyr BP constitutes a more abrupt change than in the NH trend. North American records do not show a consistently decreasing trend throughout Holocene (Fig. 6G), but rather slightly higher ∆ 18 OBSi values in the first half of the Holocene as compared to the second half. Since there are only two North American records fulfilling the aforementioned criteria, it is difficult to meaningfully go beyond the existing case studies, outlining the necessity for further research in this region. 485 The regional differences between North America and Eurasia are also manifested in the timing of absolute δ 18 OBSi minima and maxima of individual records (Fig. 7). The spatial pattern of the Holocene maxima of binned records shows a different timing of maxima for different regions (Fig. 7). Eastern Eurasian sites feature a pronounced early Holocene maximum (at 12 kyr BP), whereas sites more to the west of Eurasia show tendentially rather a middle Holocene maximum, around 6-8 kyr BP. This suggests that the double maxima of the Eurasian trend (Fig. 6H) is at least partly caused by the regional differences over 490 https://doi.org/10.5194/cp-2022-96 Preprint. Discussion started: 24 January 2023 c Author(s) 2023. CC BY 4.0 License.
Eurasia. There are also individual records, however, which do show two peaks within the Holocene e.g. Lake Bolshoye Shchuchye from the Polar Ural Mountains (#16). Records from Alaska feature later δ 18 OBSi maxima (5 kyr BP and 7 kyr BP, respectively). The timing of Holocene minima is rather homogenous in northern Eurasia with all records reaching their minimum in the last 1 kyr BP or between 1 kyr BP and 2 kyr BP (Fig. 7). While the records generally follow the same decreasing long-term δ 18 OBSi 500 trend, they feature either early or late Holocene minima. Since datapoints of 1850 CE and younger have been removed from the dataset prior to analysis to exclude the industrial era, a potential effect of recent climate change is not covered in this subset of data. As with the Holocene maximum, sites in Alaska also differ with regard to Holocene minimum. They tendentially show absolute minima earlier than Eurasian records (3 and 4 kyr BP, respectively). This underlines once more, the different behaviour of the regions (North America and Eurasia).

Combined Holocene trend in the hemispheric context
The combined trends of δ 18 OBSi for NH, Eurasia and North America are shown in comparison to other NH proxy records in Fig. 8. As individual δ 18 OBSi records are commonly discussed with respect to insolation, we provide June and December insolation curves for 60°N, calculated using (Laskar et al., 2004). As visible in Fig. 8, Eurasian and NH combined trends show 510 a similarity with the June insolation as all records feature a decreasing trend throughout the Holocene. However, insolation decreases steadily after an early Holocene maximum whereas combined NH and Eurasian δ 18 OBSi trends feature a stable early Holocene and a second peak at 7 kyr BP. This presumably relates to the geographical distribution of the timing of δ 18 OBSi maxima and minima as discussed above. Therefore, eastern Eurasian records are most in line with June insolation which has been regarded as a proxy of summer Tair. Records stemming from sites further west are less directly correlated with June 515 insolation. A striking difference is the accelerated decrease of δ 18 OBSi after 4 ka, co-incident with neo-glacial-cooling which is not mirrored in the insolation curve. The relatively stable insolation during this time suggests that the accelerated decrease visible in δ 18 OBSi records must be driven by other factors.
Conversely, December insolation shows an anticorrelation with the δ 18 OBSi records. However, as both the absolute values and the changes in December insolation are an order of magnitude lower than those of June insolation, a decisive influence of 520 December insolation on δ 18 OBSi records can be ruled out. This is in good agreement with previous works, claiming that the δ 18 OBSi proxy yields a summer-dominated signal (Shemesh et al., 2001a;Kostrova et al., 2021). It has to be noted that the records displayed still stem from different latitudes and the insolation patterns are not identical at all these sites. A decrease of summer insolation, however, is generally the case for high latitude regions throughout the Holocene.   65°N and 80°N) show a pronounced Tair increase of about 5°C at the beginning of the Holocene until 10 kyr BP, followed by a stable phase until 7 kyr BP. While the NH and Eurasian trends do not feature this increase during the early Holocene, they 535 do show a relatively stable phase from 10 kyr BP to 7 kyr BP. After 7 kyr BP, both the NH δ 18 OBSi trend and the NH Tair reconstruction feature a decrease of circa 2°C and 1.5‰ VSMOW, respectively. As the present-day global average Tairdependent fractionation in precipitation amounts to 0.695‰/°C (Dansgaard, 1964), a 1.5‰ δ 18 O decrease would, thus, https://doi.org/10.5194/cp-2022-96 Preprint. Discussion started: 24 January 2023 c Author(s) 2023. CC BY 4.0 License.
correspond to a 2°C cooling, the same Tair-change as found by Vinther et al. (2009). This good agreement of δ 18 OBSi combined trends (NH and Eurasia) and the Tair-reconstructions by Vinther et al. (2009) suggests summer Tair to have a major impact on 540 δ 18 OBSi records on millennial time scales.
Multi-proxy-based temperature reconstructions by Marcott et al. (2013) (regional stack 30°N to 90°N) also show a temperature increase until 10 kyr BP, followed by a stable phase until ca. 7 kyr BP and a decrease of ca. 2°C thereafter. This temperature reconstruction includes a much broader geographical focus as well as different archives. Thus, the stable phase from 10 kyr BP to 7 kyr BP and a maximum around 7 kyr BP are shared features of the δ 18 OBSi trends of this work, the ice core 545 reconstruction from Vinther et al. (2009) and the temperature reconstructions by Marcott et al. (2013). However, the early Holocene maximum of the δ 18 OBSi trends is not represented in reconstructions by Vinther et al. (2009) andMarcott et al. (2013).
This discrepancy might reflect different regional biases of Marcott et al. (2013) and this work, which is also in line with the different timing of Holocene maxima in the δ 18 OBSi data (Fig. 8). Additionally, it might also indicate other influences than Tair on the δ 18 OBSi signal in the frame of a more complex climate system during the early Holocene. Given the discrepancy between 550 temperature reconstructions and δ 18 OBSi records in the early Holocene, it is likely that the influence of factors other than Tair is especially pronounced in the early Holocene. The LR04 stack of the δ 18 O of benthic foraminifera (Lisiecki and Raymo, 2005) features continuously decreasing δ 18 O throughout the Holocene which is in accordance with the combined Eurasian δ 18 OBSi trend.
Pollen records may help to further investigate this issue as they stem from terrestrial environments, often even also from lake 555 sediments. Moreover, they offer a similar temporal and geographical focus compared to our records, extending to high latitudes. For this comparison we have used a pollen data compilation by Herzschuh (2021) comprising data from 864 records from sites north of 45°N for the Holocene. Changes of annual Tair and July Tair relative to modern values were retrieved from the pollen dataset. Annual and July Tair show similar patterns with a pronounced early Holocene increase, followed by middle Holocene maximum and a less pronounced decrease until the present day. Both the middle Holocene maximum and the 560 subsequent decrease of Tair are in good agreement with the combined Eurasian δ 18 OBSi trend (Fig. 8A, 8D). This further supports a substantial influence of Tair on the δ 18 OBSi signal. However, the early Holocene maximum of δ 18 OBSi in our combined NH and Eurasian trends is not reflected by pollen-based reconstructions. Caution has to be applied when using pollen-based climate reconstructions for comparison because vegetation changes may lag behind changes of climatic variables (Herzschuh et al., 2016). In summary, the similarity of our combined δ 18 OBSi trends with NH temperature reconstructions and insolation 565 data allows for deducing a clear link to summer Tair for the Holocene records.
It has to be stressed that this influence of Tair may act both directly and indirectly. Directly by means of the temperaturedependent fractionation during precipitation formation, and indirectly via the impact of temperature on the hydrological cycle and δ 18 O of water compartments. This includes factors such as moisture origin, precipitation intermittency and atmospheric circulation patterns which have been described as key drivers in numerous case studies. This underlines the importance of 570 scale when assessing δ 18 OBSi records, both temporally and spatially. On a millennial and hemispheric scale, Tair can be identified as one main driver of δ 18 OBSi records even though the signal formation is generally complex and challenging to interpret with https://doi.org/10.5194/cp-2022-96 Preprint. Discussion started: 24 January 2023 c Author(s) 2023. CC BY 4.0 License.
individual records. A Tair influence for certain records on the millennial scale is therefore not necessarily a contradiction to the findings of the original publications which may have attributed the record's signal to other factors (e.g., hydrological processes) on shorter timescales. 575

Frontiers and Challenges (Records older than MIS 1)
Further back in time than MIS 1, coverage with δ 18 OBSi records is generally limited. As MIS 1 (N=37) and MIS 2 (N=18) have the highest numbers of records, they offer the best possibility of comparing glacial and interglacial records. Records beyond MIS 2 are very sparse and do not allow for filtering and generating common trends. Here, the interpretation has to rely on 580 comparison of individual datasets. There are few (N=16) records only which cross the MIS 1-MIS 2 boundary and these records cover neither MIS 1 nor MIS 2 entirely. A quantitative comparison of MIS 1 and MIS 2 is therefore difficult. Again, caution has to be applied due to the fact that these records stem from lakes with different hydrological and climatic settings (i.e. maritime Alaskan sites vs. continental Siberian sites).
The most promising approach for investigating MIS 1 and MIS 2 is extending the NH MIS 1 datasets to the past (Fig. 9). Five 585 NH δ 18 OBSi records (#3,8,15,24,25) extend back from MIS 1 into MIS 2 and offer insight into the hydroclimate history of Siberia during the last glacial and the deglaciation. Most of these records display a maximum near the MIS 1-MIS 2 boundary and a decrease of δ 18 OBSi with increasing age in MIS 2, suggesting decreasing Tair. This is the case for all records except #3 which displays a relative δ 18 OBSi maximum at ca. 16 kyr BP. This maximum may also be caused by dry conditions outperforming the effect of lower Tair. Only records #3, #8 and #15 extend beyond 13 kyr BP. Both #3 and #15 reach their 590 absolute minima at 23 kyr BP and 24 kyr BP, respectively. Record #8 does not show such a clear pattern and shows a relative maximum at 25 kyr BP instead. This is likely due to the complex hydrological setting (alternating between open and closed at present) which may have changed over these timescales. Generally, most records suggest a lower δ 18 OBSi in MIS 2 when compared to MIS 1. This is remarkable as glacial and interglacial periods feature different environments, atmospheric circulation patterns and likely 595 hydrological settings (e.g. formation or closure of outflows from lakes, lake level fluctuations). Potentially, lower δ 18 OBSi values would be conform with either lower Tair or more humid conditions. In case of MIS2, generally associated with cold and dry conditions, a lower Tair is the more plausible scenario.
Considering all records covering the MIS 1-MIS 2 boundary, without geographical or hydrological constraints, yields a less clear picture (Fig. 9). The calculated offsets of MIS 1 and MIS 2 datapoints may show both higher and lower δ 18 OBSi values in 600 MIS 2 compared to MIS 1. Again, an MIS 2 climate colder and drier than Holocene (MIS 1) may produce either lower or higher δ 18 OBSi values due to opposing effects of Tair and evaporation. A geographical pattern of either of these effects prevailing is not obvious, which suggests that rather than the climatic background, the individual hydrological settings of the lakes in question play a prominent role in determining the δ 18 OBSi signal. This supports the approach of using lakes with both similar latitudinal and hydrological characteristics.  Further back in time than MIS 2, records become even scarcer, as do lake sediments in general. Additionally, lake sediments 610 covering these time periods often lack sufficient diatoms, especially in cold stages, i.e. Lake Baikal during MIS 4 Mackay et al., 2011;Mackay et al., 2013). Lake El'gygytgyn is a peculiar example of a continuous sedimentation history with δ 18 OBSi showing glacial-interglacial cycles at least back to MIS 9 (Chapligin et al., 2012b), which have been attributed by the authors to Tair changes. In addition to glacialinterglacial cycles, it is also possible to compare δ 18 OBSi of interglacials, i.e. MIS 1, MIS 5, MIS 11, when diatoms during cold 615 stages are absent. In the sedimentary records of Lake El'gygytgyn, Chapligin (2012) investigated the δ 18 OBSi differences between the warm stages MIS 1, MIS 5 and MIS 11 and found MIS 11 as warmest interglacial. This interpretation has been supported by pollen-based Tair reconstructions by Melles et al. (2012) and underlines the applicability of the δ 18 OBSi proxy on glacial-interglacial timescales. Studies on Lake Baikal have also investigated past interglacials and have addressed changes in precipitation intermittency and cooling events during MIS 11 . During MIS 5, millennial-scale variability 620 is suggested to have been more stable than during MIS 1 (Mackay et al., 2013). Dethlingen (Shemesh et al., 2001b;Koutsodendris et al., 2012;Wilson et al., 2014b;Schmidt et al., 2017). These offer the possibility of extending the picture past still existing lakes, however lacking (present) hydrological constraints, which makes their interpretation and comparison to other, better-constrained records more challenging. Further research on δ 18 OBSi is 625 therefore needed in order to complement the picture and provide insight into past climate and environmental conditions in continental regions, particularly valuable for high latitude and high-altitude regions poorly covered by other proxy data.
Hydrologically open lakes with a long, continuous sedimentation history are most promising for extending climate reconstructions further into the past.

Conclusions
In this study, we have identified and gathered the existing lacustrine δ 18 OBSi records published to date in order to generate an added value by unifying them into a first δ 18 OBSi data compilation. We have identified 53 δ 18 OBSi records derived from 71 635 publications. These records stem from the entire globe with their geographical distribution focusing on high altitude and high latitude lacustrine environments. Diatoms bear the advantage of being available in dilute, non-alkaline lakes common in these environments. Regional clusters of records (e.g. northern Eurasia, East Africa) indicative of the research foci of the individual research groups employing the δ 18 OBSi proxy allow for generating regional subsets of data. Temporal coverage stretches from sub-recent timescales to the Pliocene which highlights the applicability of δ 18 OBSi on different timescales. Best coverage is 640 available in MIS 1 which is linked to the age of lakes, and hence the availability of lake sediments, especially in high latitudes.
Moreover, biogenic silica (here entirely diatom-based) is most abundant during warm periods (interstadials and interglacials) compared to colder periods. The interpretability of δ 18 OBSi records relies on regional setting, hydrological constraints of individual lakes and catchments, and preferably supported by isotope measurements of present lake water. Most δ 18 OBSi records stem from open lakes (N= 41), suggesting for these lakes a negligible influence of lake evaporation. In contrast, closed lakes 645 (N=12) and paleo-lakes (N=9) have been investigated less. It has to be noted that a lake's ontogeny and hydrology may change throughout time and thus, hydrological changes constitute an approach for interpreting these δ 18 OBSi records.
Spatial resolution of the δ 18 OBSi records is determined by the size of the lake and its corresponding catchment with lake water effectively integrating the input signal. While this regional signal may integrate large areas such as in the case of lake Baikal, most catchments (N=18) are <100 km 2 , suggesting a locally confined signal for these lakes. Regarding temporal resolution, 650 most records feature sampling resolutions which by far exceed tres, suggesting sampling resolution to be the decisive factor in determining a record's signal. However, in case of tres >100yrs, lakes may be subject to increased lake evaporation. Accounting for offsets and different temporal resolution of records and filtering for similar hydrological settings (hydrologically open lakes with tres<100 yrs), we find a common pattern throughout the Common Era at the centennial scale, which we attribute to changing hydroclimate conditions. Changes are similar between Eurasia and North America, but they 655 still differ in the timing of δ 18 OBSi maxima. Generally, the combined Eurasian δ 18 OBSi record seems to include major climate episodes during this period, including Roman Climate Optimum, Migration Period Pessimum, Medieval Climatic Anomaly (MCA) and Little Ice Age. An effect of recent warming is not visible in the Common Era data, likely due to a lack of records meeting the filter criteria.
For the entire Holocene (MIS 1) in Eurasia and the NH (for hydrologically open lakes with tres<100 yrs), we find a common 660 decreasing δ 18 OBSi trend on the millennial scale which roughly follows summer insolation. The timing of the absolute δ 18 OBSi maxima of individual records in Eurasia differs and shows an east-west gradient with eastern Eurasian records featuring earlier Holocene maxima compared to western Eurasian records. Holocene minima occur within the last 2000 kyrs for all Eurasian records. North American records diverge from this MIS 1 pattern with later maxima and earlier minima. This behaviour is likely linked to atmospheric circulation patterns as also described by the authors of the individual records. 665 Extending the millennial-scale trend of Eurasian records into MIS 2 shows generally lower δ 18 OBSi values in MIS 2, suggesting lower Tair (and not P/E) in Glacial times, also visible in most other records that act on glacial-interglacial timescales. The applicability of the δ 18 OBSi proxy beyond MIS 2 is generally constrained by the availability of lake sediments and the abundance of diatoms in the respective sediments. Lake El'gygytgyn, as a prime example of hydrological continuity, displays glacialinterglacial cycles in the δ 18 OBSi record, supporting Tair being a prominent driver of δ 18 OBSi on millennial time scales. In Glacial 670 periods, diatoms abundances are low, and this limitation does sometimes not allow for oxygen isotope analysis. However, comparison between interglacials where diatoms are generally more abundant, is feasible, i.e. at Lake Baikal and Lake El'gygytgyn. In summary, we demonstrate the applicability and intercomparability of combined δ 18 OBSi records into a first lacustrine δ 18 OBSi compilation allow for paleoclimate reconstructions, when accounting for regional offsets, different temporal resolutions and hydrological backgrounds. δ 18 OBSi records compiled in this study are, thus, an important tool for reconstructing 675 paleoclimate across the globe and across a variety of time scales. In NH extratropic regions, their Tair-driven, quantitative signal makes them especially usable for use in conjunction with paleoclimate models. Future research would be most valuable in complementing the existing data with longer records (covering MIS 2 and beyond) as well as records from regions underrepresented thus far (especially the southern hemisphere).

Data availability 705
The dataset is being submitted to PANGAEA.

Author contributions
PM, HM, BD and BB designed the research project and structure of the manuscript. PM compiled the data base, carried out statistical analyses and produced all figures and tables and wrote the main part of the manuscript. ML, GR, EB, AH, GS, AM, HM, HB and PB contributed data publicly not available. All co-authors brought in their expertise have made substantial 710 contributions to the manuscript. All authors have written parts of the manuscript, commented on drafts and have approved its final submitted version.