Patterns in recent and Holocene Pollen Accumulation Rates across Europe; the Pollen Monitoring Programme Database as a tool for vegetation reconstruction

The collection of modern : , spatially extensive pollen data is important for the interpretation of fossil pollen diagrams. Such :::::::::: assemblages :::: and ::: the :::::::::::: reconstruction :: of :::: past :::::::: vegetation ::::::::::: communities :: in ::::: space :::: and :::: time. ::::::: Modern datasets are readily available for percentage data but lacking for pollen accumulation rates (PAR). Filling this gap has been the motivation of the pollen monitoring network, whose contributors monitored pollen deposition in modified Tauber-traps for several years or decades across European latitudes :::::: Europe. Here we present this monitoring dataset consisting of 351 trap locations with a total of 2742 5 annual samples covering the period from 1981 to 2017. This dataset shows that total PARs ::: PAR : are influenced by forest cover and climate parameters, which determine pollen productivity and correlate with latitude. ::::::: Regional ::::: forest ::::: cover :::::: >80% : is :::::::: indicated ::: by :::::: >3200 ::: tree :::::: pollen :::::: grains ::::: grains ::::: cm-2 ::: y-1. : Pollen traps situated beyond 200 km of the distribution of a given tree species are still collecting ::: still :::::: collect occasional pollen grains of that species. PAR’s of up to 30 grains cm-2 y-1 in fossil diagram should therefore be interpreted as long distance transport from beyond 200 km from the area of distribution. Compar10 isons between modern and fossil PAR from the same regions show comparable :::::: similar values. Comparisons ::: for :::::::: temperate :::: taxa often demonstrate that similar high values for temperate taxa in fossils sites ::: trap :::::: values : are found further south or downhill. While :: we ::: do ::: not :::: find modern situations comparable to high :::: fossil : PAR values of some taxa (e.g. , : Corylus)do not occur, : . CO2 fertilization and land use may cause high modern PAR ’s that are not documented in the fossil record. The modern data is now publically ::::::: publicly available in the Neotoma Paleoecology Database (https://apps.neotomadb.org/explorer/) and serves 15 improving ::: and ::: aids : interpretations of fossil PAR data. Copyright statement. The article is distributed under the Creative Commons Attribution 4.0 License. Unless otherwise stated, associated published material is distributed under the same licence.


The need for a dataset of modern absolute pollen deposition
Pollen analysis became the widely used method for the reconstruction of the Holocene vegetation. Pollen percentages are a simple representation of pollen analytical results, but have a number of well-known limitations and biases that are often 20 ignored. One of these is separating locally produced from long distance transported pollen (Davis, 2000), which is paramount for mapping past changes in plant distributions. Also reconstructing the position of tree-lines from pollen percentage data may be misleading as local treeless vegetation (e.g., tundra) produces few pollen grains, while distant woodlands (e.g., consisting of boreal trees) produce much pollen. In such situations absolute pollen data are very informative as was already realized by Hesselman (1919) and Malmström (1923). Pollen accumulation rates (PAR) or the number of pollen grains deposited on the sediment surface over a set period of time, are in theory superior to pollen percentages as they do not suffer the closure effect of percentage data. Thus by using absolute data it is possible to differentiate between low amounts of long distance transported versus high amounts of locally produced pollen. In a seminal publication Davis et al. (1964) document the power of using 5 absolute pollen deposition for the interpretation of the spread of trees during the postglacial afforestation around Rogers Lake in Connecticut, USA. Another bias in percentage data is the interdependence of values obscuring the quantification in the amount of change of a single taxon. PAR are therefore required when studying the population dynamics of individual trees (Bennett, 1983). While absolute pollen data do not share the artefacts of percentage data, they are often difficult to obtain and subject to a different set of limitations. 10 One limitation for obtaining reliable PAR from sediment cores is the requirement for accurate chronologies. Lake internal processes such as re-deposition and sediment focusing and also catchment erosion may bias the resulting values (Davis and Brubaker, 1973;1984;Giesecke and Fontana, 2008;Bennett and Buck, 2016). These are some reasons why advances in interpreting PAR have been slow. The other reason is that collecting modern PAR values is not as simple as collecting mosses, soil litter or the top sediment of lakes for obtaining modern pollen percentages for a particular vegetation type. Modern rates 15 of pollen accumulation can be obtained from monitoring pollen deposition using pollen traps (Hicks, 1994), as well as from carefully sampling the top sediment of lakes that are either annually laminated or precisely dated (Matthias and Giesecke, 2014). Due to the high inter-annual variability in pollen production (Andersen, 1980;Haselhorst et al., 2020), it is necessary to conduct pollen monitoring over several years to enable comparisons with estimates from sediment cores (Hicks, 1974;Hicks and Hyvärinen, 1999). 20 For these reasons there are only a few investigations of the pollen vegetation relationship using absolute pollen deposition, while there are numerous studies using percentage data. Nevertheless, investigations using pollen traps yielded invaluable insights into the mode of pollen transport (Tauber, 1967). Also the construction of representation factors for common Europe trees by Andersen (1970), which are still used, was based on pollen monitoring data from pollen traps. In this way pollen monitoring studies have contributed to the development of models of pollen dispersal and deposition (Gaillard et al., 2008). 25 Several aspects of PAR data have not been exhaustively explored in modern comparison studies and here we will focus on the following three: 1) the influence of climate in combination with forest biomass; 2) the application of PAR to indicate the local presence of trees versus long distance transport of pollen; using modern PAR values of single taxa to interpret fossil situations. 30 Recent investigations demonstrate the linear response of absolute pollen deposition to absolute tree abundance (Seppä et al., 2009, Sugita et al. (2009, Matthias and Giesecke (2014)), which may be used to reconstruct past standing tree biomass of different trees. However, at an annual time resolution, variability in PAR can be explained by weather conditions during the time of flowering as well as during the previous year (Hicks, 1999, van der Knaap et al. (2010, Nielsen et al. (2010)). Thus the question arises: If weather is determining annual pollen production, could climate in addition to determining biomass determine average PAR? Comparing Pinus PAR between two pine dominated forest regions in central Sweden and north east Germany shows much higher values in the south, suggesting that PAR may not correspond to tree biomass alone (Matthias and Giesecke, 2014). The relationship between pollen production and weather suggests that more pollen is produced when the primary productivity of the tree is higher. This is also true for fertilization with CO 2 (LaDeau and Clark, 2006). Therefore, 5 climate and even the amount of CO 2 in the atmosphere may determine the pollen productivity of a tree at a given site. Welten (1944) already interpreted the first fossil PAR in this way, suggesting that climate deteriorations may not immediately lead to a decline in forest cover but to the amount of pollen produced. This interpretation of changing PAR was however forgotten. If climate and CO 2 determine pollen productivity than the postglacial increase in PAR at Rodgers Lake could also be due to a change in these parameters. It is, therefore, important to investigate the possible relationship between climate and average PAR 10 in more detail.

Local presence versus long distance transport
Also the initial question on the amount of pollen that may arrive at a site from long distance sources has not been addressed in a systematic way using modern absolute pollen deposition data. Occasional reports of PAR values for the first occurrence of a macrofossil indicating local presence exist (e.g. Giesecke, 2005a), but modern comparisons are lacking. While percentage 15 data are not well suited for detecting distribution limits, a continental scale comparison (Lisitsyna et al., 2011a) provides some guidance on values that can be used for mapping past distribution changes (Giesecke et al., 2017). The PMP dataset presented here provides a continental scale dataset permitting such a comparison with PAR data.

Modern analogies
Spatially extensive modern pollen percentage datasets provide the possibility of searching for modern analogues for fossil 20 pollen proportions and in this way reconstructing past vegetation and environmental conditions assemblages (Davis et al., 2013;Jackson and Williams, 2004;Overpeck et al., 1985). Modern datasets of absolute pollen deposition are hitherto rarely used to reconstruct past tree abundances or environmental conditions. By using a network of pollen traps across the latitudinal tree-line in Finland, Hicks et al. (2001) showed that average modern PAR values can be obtained representing the gradual transition from the boreal forest to tundra. These "modern analogues" were successfully applied to reconstructing Holocene 25 shifts of the latitudinal tree line (Seppä and Hicks, 2006). This idea of building a modern dataset of absolute pollen deposition that can be used as a reference to interpret fossil PAR was the motivation for the establishment of the Pollen Monitoring Programme (PMP, Hicks et al., 1996;.

The Pollen Monitoring Programme (PMP)
The programme was launched in August 1996 at a meeting in Finland, bringing mainly European researchers together. Mem- 30 bers of the network changed over the years and monitoring experiments were discontinued or newly started. Although pollen monitoring studies were and are carried out in other continents (e.g. Jantz et al., 2013), the PMP had little success in attracting researchers working outside Europe. The standardisation of the monitoring protocol allowed for easy comparisons between the results in different regions, which were discussed at INQUA in 1999 and led to a special volume published in 2001 (Tinsley andHicks, 2001) collecting results based on several initial time series (van der Knaap et al., 2001;Koff, 2001;Tinsley, 2001;Tonkov et al., 2001), as well as a first comparative study (Hicks et al., 2001). More individual results were 5 published in the following years (e.g. Gerasimidis et al., 2006;Giesecke and Fontana, 2008;Hättestrand et al., 2008;Jensen et al., 2007;Kvavadze, 2001;Pidek, 2007) and comparative studies followed in a second special volume published in 2010 . The data produced by contributors to the PMP were analysed for different questions, including weather parameters determining the amount of pollen production (van der  and its correlation to masting years in Fagus . The programme established a database collecting the original data for individual years, as well as 10 general information on the pollen traps installed in the different regions (Fig. 1). The database was developed offline and was thus difficult to access by individual researchers. The paleoecology database Neotoma (Williams et al., 2018) offers a platform to store the PMP data and make it available to researchers worldwide, allowing them to interrogate the data and potentially identify modern analogues to interpret fossil pollen accumulation rates.
The overall purpose of this manuscript is to present an overview of the data in the PMP database and to interrogate this 15 continental scale dataset of modern PAR with the following aims: a) Examine the hypothesis that climate as well as regional forest cover explain the variability in PAR. b) Study the absolute amount of long distance dispersed pollen encountered in pollen traps beyond the known distribution limits of the parent trees. c) Compare modern and fossil pollen accumulation rates by collecting fossil datasets with estimates of PAR from the same 20 regions where the pollen traps were installed. For the most abundant pollen types we explore how the modern situations can provide a reference for the interpretation of the fossil data.

Study area
Sites in the PMP database were divided into 7 'trap regions' according to their latitude and altitude. These regions were further

Data collection
The pollen traps used in the PMP network generally consist of a bucket or bottle large enough to contain the annual surplus in precipitation on a surface of usually 19.6 cm 2 (5 cm diameter opening) or similar. Many traps had a sloping collar inspired by the design of pollen traps by Tauber (1974), although few collars were truly aerodynamic. The collection of the trap content was generally carried out annually and any special circumstances potentially affecting the annual pollen deposition were noted 5 and stored in the database. For the analyses presented here and data overview, we excluded traps where the pollen record is 2 years or less, as averages may be affected by high inter-annual variability. The only exception is the trap situated in Spitsbergen where there is a two-year record. Pollen accumulation from the two-year record shows little variation and, being the only analogue for a truly arctic and treeless environment, provides important information on long distance pollen transport. We also excluded annual samples with records shorter than 8 months and, in addition, traps or years with spurious values due to 10 particular events or local conditions (Table S1).
Most of the traps in the PMP network were placed in the open vegetation or in forest openings in order to avoid an unrepresentative contribution of individual trees e.g. due to anthers dropping into the traps. Traps were generally installed at ground level mimicking collection conditions relevant for sedimentary archives. Consequently, tall herbs or grasses might have overgrown or covered some of the traps potentially leading to higher pollen deposition. Traps not equipped with a mesh oc- 15 casionally trapped pollen-collecting insects, leading to enormous counts of insect pollinated taxa e.g. Calluna, Erythranthe guttata. The presence of insects in the traps is usually noted for the collection year so that careful evaluation of the information in the database can also inform on herbaceous pollen types (Jensen et al., 2007). Including this information in comprehensive database queries is currently not possible and a manual screening of datasets is required when analysing herbaceous pollen types. This problem does not seem to occur in tree pollen taxa. The occurrence of phytophagous insects in the traps were not 20 accompanied by unusual peaks in tree pollen taxa, indicating that the insects inadvertently trapped were primarily collecting pollen from the herbaceous vegetation around the traps. Comprehensive database queries restricted to tree pollen, Poaceae and Cyperaceae should therefore not be affected by the occurrence of insects in the trap and mainly represent pollen transport via wind, the rainout of pollen from the atmosphere and the gravity component (Tauber, 1967).
Concentrating the content of the traps was carried out either using filter paper or centrifugation and decanting the supernatant. 25 In many cases the trap content was washed onto a paper filter, which was later digested using acetolysis. Pollen quantity was assessed by adding Lycopodium spore tablets (Stockmarr, 1971) to each trap before processing. Pollen concentration was obtained from the ratios between pollen grains counted to Lycopodium spike counted and Lycopodium spike added. Details about Lycopodium spike data, as well as details of the pollen trap such as the exact size of the opening are stored in the database. The PMP database was created in the PostgreSQL database system. Names of pollen taxa were unified using the

Investigated taxa, climate and forest cover
We selected the common tree and shrub taxa of Europe. Pollen taxa generally refer to all the species within the genus. Pollen taxa allowing higher taxonomical resolution, which were consistently separated and excluded from the genus in the whole dataset are marked as "excl.". Pollen taxa potentially including pollen grains from another genus are indicated by "incl.": Abies, Alnus (excl. A. viridis), Betula (excl. B. nana-type), Carpinus-type (incl. C. orientalis/Ostrya-type), Corylus, Fagus, 5 Fraxinus (incl. F. ornus), Juniperus-type (incl. Cupressus, Tetraclinis, Thuja), Picea, Pinus (excl. P. cembra-type), Tilia, Quercus (incl. Q.robur-type, Q. cerris-type and Q. ilex-type). Pollen accumulation rates of trees and shrubs were summed as arboreal pollen accumulation (hereafter as "tree PAR"). We also included pollen from the plant families Cyperaceae and Poaceae (excluding cereals). For the purpose of the analysis in this paper we refer to the sum of tree PAR plus Cyperaceae and Poaceae as "total PAR". 10 The climate parameters Mean Annual Temperature (MAT) and Annual Precipitation (APrecip) for the trapping locations were obtained from WorldClim 2 (Fick and Hijmans, 2017). For site altitude we used the information supplied by the individual investigator. Comparisons between PAR and forest cover were conducted using the data of the Forest Map of Europe (Kempeneers et al., 2012), which has a grid resolution of 1 km 2 . Forest cover was extracted as a mean of all grid cells within a 10 km radius. We used regression analysis to explore whether individual or combinations of these environmental parameters 15 describing the trapping location can explain the variance in average pollen accumulation of total and tree PAR. To balance the contribution of high and low pollen producers in the assessment of the PAR, we applied correction factors (Table S2, Andersen, 1970). Average modern PAR have a large variance of values between traps of the same region, with the smaller numbers often being the focus of information. For this reason, we often log-transformed PAR values in the different analyses.

20
Pollen deposition beyond the distribution area of the parent plant was studied by merging the distribution maps of the relevant species included in each of the pollen types listed above (Caudullo et al., 2017;San-Miguel-Ayanz et al., 2016). These comparisons were not possible for Alnus, Betula, Cyperaceae, Juniperus, Pinus and Poaceae as these taxa are widely distributed in Europe and few traps are located beyond their distribution area.
For each trap location and each pollen taxon we calculated the distance to the nearest area of distribution using GIS (GRASS 25 Development Team, 2018). Initial observations showed that PAR dropped rapidly away from the distribution of the parent tree and did not decline at the same rate at larger distances. We therefore compared distance to the decadic logarithm of PAR, applying linear regression to explore thresholds of long-distance transport (hereafter also as "LDT") at 200 km from their mapped distribution limits. This distance was chosen as a compromise accounting for uncertainties in the information on distribution limits and available pollen trap data. Pollen traps in the UK are situated beyond the natural distribution limits of 30 several of these trees but were excluded from the comparison as the target taxa may be planted in the area.
We compare LDT with the characteristic radius for the same taxa. Characteristic radius is a useful measure showing pollen transport predicted by pollen dispersal models. It represents the theoretical proportion of pollen loading at different distances from the source plants (Prentice, 1988) and can thus be easily compared with our empirical values. We used Gaussian Plume model with wind speed 3 m.s -1 (Prentice, 1985).

Comparison between modern and Holocene PAR
To enable the comparison of modern with fossil PAR values the pollen trap data was extracted from the PMP database with above described constraints and all annual samples were averaged within traps. For each trap region we selected at least one and 5 a maximum of three Holocene PAR records (Table 1). This distance was chosen as a compromise accounting for uncertainties in the information on distribution limits and available pollen trap data. Holocene PAR estimates often show high variation between samples due to changes in the sedimentary environment. To reduce this effect in this comparison, Holocene data were averaged in 500-year bins. Site and sample compilation resulted in a fossil dataset containing 354 Holocene samples.
We compared trap and fossil PAR datasets in two ways. For a general comparison of the distribution of fossil and modern 10 values, we plotted the frequency of log-transformed PAR values over all regions and traps per taxon. Per region and taxon we compared the frequency distribution of fossil and modern PAR values using a t-test at 5% level of alpha to identify situations where modern and fossil values are comparable. Secondly, we compared trap and fossil PAR at the level of individual sites or trap areas. To facilitate this comparison average trap and fossil PAR values per taxon were submitted to one-dimensional clustering using the R-package Ckmeans.1d.dp (Song and Wang, 2011). This method splits the univariate data in the way that 15 the total of within-cluster sums of squares is always minimum. These classes helped us compare trap and fossil data and to link the highest fossil PARs with the trap PARs. We dealt only with the highest class in each fossil sequence because maximum abundance of several our target taxa was used as a stratigraphic marker of the Holocene period and thus their timing is well known. However, time windows did not smooth out all spuriously high values variation. In order to remove this remaining variation of individual time windows, we ignored some high fossil values (Table S3). Thus, we aimed to find modern analogues 20 for fossil situations represented by several bins (more than 500 years). We linked these periods containing high fossil PAR to the closest pollen trap, using a matrix of geographical distances between fossil sites and pollen traps. All statistical analysis and data visualizations were produced in R (R Core Team, 2019).

Results and interpretations
3.1 Overview of the PMP database and the environments sampled 25 The PMP database version 02.02.2020 contains data from 351 trap locations with a total of 2742 annual samples covering the period from 1981 to 2017. Considering the trap records with 3 years and more we obtained 271 mean trap assemblages.
Trapping sites cover a range of altitudes from 0 to 3000 m a.s.l. with annual precipitation ranging from 402 to 1549 mm. Mean annual temperatures (MAT) for the sites fall between -5.7 to 14.1°C. The forest cover within a 10 km radius of the trapping sites ranges from 0 to 98%. This range of environmental situations has yielded tree pollen accumulation rates from 5 to 86000 30 grains cm -2 y -1 , with a median value of 5400 grains cm -2 year -1 (Fig. 2). An overview of the taxonomic composition of the  (Jahns, 2004) traps ( Vegetation history at the rest of the fossil sites shows a more dynamic development (Fig. S2).

Dependence of variation in PAR on regional forest cover and climate
Total PAR is generally lower at high latitudes, with the lowest values in the Arctic/Alpine region (trap area Spitsbergen), where no trees can grow. However, the highest absolute values of tree PAR are not from the southernmost traps but from the Lowland 10 Temperate region (trap area Tver; Fig. 2). Latitude alone explains about 11% of the variance in log-transformed tree PAR, while MAT and forest cover within 10 km explain 21% and 19% respectively. In combination, these three variables explain 37% of the variation in log-transformed absolute tree pollen deposition. The addition of elevation increased the amount of variance explained to 50% (Table S4a).
Large differences in the pollen productivity between different trees affect this relationship. Adjusting the PAR from individual taxa by Andersen factors reduces the bias of differential pollen production between different plants and makes it possible to consider the total amount of pollen deposition including grasses. This adjustment increases the amount of variance explained 5 by the regression model with all 4 explanatory variables to 56% (Table S4b). Due to the inclusion of grasses, the explanatory power of forest cover within 10 km is reduced, while latitude alone explains 37% of the Andersen adjusted log-transformed total PAR (Fig. 3a, Table S4b).
The regression models consider the full range of the data while, due to local factors, there is often a spread of average trap values for different traps in the same region. The traps with the highest regional values do not follow a latitudinal pattern, so the 10 distribution of the minimum average trap values is more informative (Fig. 3a). These lower values closely follow a latitudinal trend. The average PAR south of 62°latitude and below the altitudinal treeline or close to forests is generally higher than 1000 grains cm -2 y -1 . An area with low PAR in the south is the coastal grassland in northern Bulgaria. The generally low PAR in this area can be explained by the sparse vegetation cover on thin rendzina soils formed on limestone rock. Adjusting the PAR values by Andersen factors increases the values for this region so that they fit the general latitudinal trend (Fig. 3a). Traps with 15 minimum average PAR values per region also correspond well to the forest cover within 10 km (Fig. 3b). Exploring the data showed that a 3% wide bins of the forest cover, traps with the lowest PAR per each bin of the forest cover provide a regression model predicting a tree PAR of 3200 grains cm -2 year -1 at 80% forest cover within 10 km of the trap.

Long distance dispersed pollen
The comparison of PAR with the distribution limit of different tree taxa shows that PAR generally declines with distance 20 (Fig. 4). A gradual decline is best documented for Quercus where traps cover different distances from the distribution area.
This analysis also documents the long-distance transport of many tree pollen, including the heavy pollen of Picea. For better comparison of the absolute values between taxa, we fitted a linear relationship, also to compare the amount of pollen at 200 km from the distribution limit (Fig. 4b). This comparison indicates that less than 80 grains cm -2 y -1 of Carpinus, Corylus, Fagus, Fraxinus, Quercus and Tilia are deposited beyond 200 km of the distribution of the parent trees. Only Picea shows less than 25 1 grain cm -2 y -1 at 200 km of the distribution range. In the case of Fagus it has to be noted that one trap beyond 200 km of the distribution of the tree recorded more than 30 grains cm -2 y -1 . The general threshold of 80 grains cm -2 y -1 suggested here is near the detection limit and the PAR value may be biased by the size of the pollen count in cases where only one grain was encountered.
PAR at 200 km from the distribution area represent 0.002-36% of the median PAR within the distribution area (Fig. 4), 30 which is 0.2% of pollen loading in average for those seven taxa (Table S5).  (1) and minimum tree PAR per every 3 % of forest cover (2).

Ranges of modern and fossil PAR values
The comparison of modern and fossil PAR values shows good agreement in tree PAR. The highest frequency of tree PAR values ranges between 2000 and 10000 grains cm -2 y -1 in both datasets (Fig. 5). Maximum PAR of the trap dataset are higher  Table S6). In this regional comparison, Betula shows the best agreement between Corylus has a good overall agreement, the regions with similar modern to fossil data are shifted, with Holocene values in the 10 Boreal region corresponding to modern PAR in the Lowland Temperate region.

Taxa specific linkage of the highest average PAR at fossil sites with individual trap values
To facilitate the comparison of modern and fossil PAR, the combined taxa specific values were submitted to a one-dimensional cluster analysis, which resulted in between 5 and 9 classes of PAR values per taxon. Comparing the highest class of fossil PAR to modern trap data on a site-by-site basis shows that it is possible to find modern comparisons for all fossil situations. We 15 demonstrate the main linkage. Detailed descriptions are presented in the supplement (Fig. S3).
Abies declined in most of the populations, thus Roztoce is the only analogous trap area with PAR for fossil sites in central Europe and similarly the Timfristos trap area is analogous for fossil sites in southern Europe (Fig. 6).   Table 1 for full name) highlighted by the corresponding colour for the class (see b) Note the scale of the x-axis corresponds to the x-axis scale of graph a).
Quercus and Tilia (Fig. 7)         Although data on plant biomass and primary productivity are not available for all trapping locations, the regression analysis indicates that mean annual temperature has an influence on the quantity of pollen deposition. The July temperature of the previous year determines the amount of pollen production in Pinus near the tree-line (Autio and Hicks, 2004;McCarroll et al., 2003). Evidence from other European regions (van der Nielsen et al., 2010) suggests that the growing 10 season warmth and other climate variables also explain the interannual variability of pollen deposition. On a regional scale, PAR corresponds to plant biomass of the parent tree (Matthias and Giesecke, 2014;Seppä et al., 2009). However, differences in forest cover cannot explain the latitudinal gradient in PAR described here, which may, at least in part, result from the latitudinal gradient in primary productivity of trees (Gillman et al., 2015). An increase in primary productivity and pollen production has been shown in a carbon dioxide fertilization experiment (Wayne et al., 2002), which supports the interpretation that average 15 PAR of the same species may vary due to environmental parameters determining its productivity.

Long Distance Transport
Modern PAR from traps near the latitudinal limit of Pinus and Betula have been used previously to reconstruct past changes in the northern distribution limits of these trees (Seppä and Hicks, 2006). Our LDT result 80 grains cm -2 y -1 is slightly lower than the range for Pinus and Betula in arctic-apline zone 100-200 grains cm -2 y -1 (Seppä and Hicks, 2006). Here we evaluated greater distances and, therefore, had to ignore both species, while indicating some general thresholds for other dominant European trees. The suggested threshold of 80 grains cm -2 y -1 may be used to differentiate local from long distance transport for most wind pollinated trees. However, this value may still lead to false negatives as the absence of a plant cannot be proven by the absence of evidence for presence. The value of 80 grains cm -2 y -1 for Picea agrees well with the fossil PAR value for the tree of 50 Picea grains cm -2 y -1 found in a sample at Klotjärnen just after the occurrence of the first Picea bud scale (Giesecke, 2005b). However, these modern thresholds estimated here are likely to depend on the abundance of the parent tree 10 in the larger region rather than properties of the pollen types. A higher threshold would be expected for Corylus compared to Fagus. Corylus has a lighter pollen grain than Fagus, which can travel more easily over large distances (Table S5) the proportion of pollen loading within the area of distribution and at 200 km from the source plants measured empirically by our PAR results (0.002-36%, in average 0.2%) is wider and higher than the theoretical range (0.005-13%, in average 0.05%; Table S6). This mismatch can be caused by too leptokurtic character of the Gaussian Plume model with windspeed 3 m.s -1 and fits with previous indication that it underestimates dispersal of pollen with a large grain (Abraham et al., 2014, Theuerkauf et al. (2016). 20 Pollen rain beyond the area of distribution is analogous with high elevations with sparse vegetation. However, with the increasing altitude, PAR decreases independently from the actual growth density, because of the worse climatic conditions and low pollen productivity (Markgraf, 1980).

Analogues for vegetation reconstruction
The comparisons of modern and fossil PAR values show that pollen traps can characterize the population density of partic- PAR at Soppensee in northern Switzerland are 12000 grains cm -2 y -1 (Lotter, 1999) and at Meerfelder Maar (Kubitz, 2000) in western Germany 18000 grains cm -2 y -1 . Judging from pollen percentages, even higher Early Holocene values should be found in more oceanic situations and the Corylus PAR at Hockham Mere in eastern England may be as high as 40000 grains cm -2 y -1 for the early Holocene (Bennett, 1983). Modern values in pollen traps from Wales at around 2000 grains cm -2 y -1 are far below these early Holocene figures and it is likely that modern analogues of sites with high Corylus PAR no longer exist in Europe.

5
Conversely, the high modern PAR values for Pinus and Betula from Poland and Latvia are not found in the fossil examples.
Pinus PAR values around 30000 grains cm -2 y -1 were also obtained from 210 Pb dated modern lake sediment samples in north eastern Brandenburg (Matthias and Giesecke, 2014). This study evaluated the PAR for the years 1993 and 2009. The increase in Pinus PAR values between the first and the second sampling period corresponded with an increase in the amount of standing pine volume in the region. Forestry practices aimed at increasing yield could account for the high Pinus values. Pinus was 10 extensively planted after the 1950s, even on soils where trees with a lower pollen production would have grown naturally.
The fertilization due to increased nitrogen deposition (Pers-Kamczyc et al., 2020), as well as increased atmospheric carbon dioxide, increase the pollen production not only of Pinus. A carbon dioxide enrichment experiment of 19-year old Pinus taeda resulted in a twofold probability of reproductive maturity after 3 years (LaDeau and Clark, 2001). The continued experiment also showed that carbon dioxide fertilization increased the number of pollen cones and therefore pollen grains produced per 15 tree (LaDeau and Clark, 2006 examples considered here are consistently below 6000 grains cm -2 y -1 and published early Holocene values rarely exceed 6000 grains cm -2 y -1 (but see Theuerkauf et al., 2014). Pollen diagrams from the forest steppe ecotone in European Russia are often characterized by high Betula percentages (Nosova et al., 2019;Shumilovskikh et al., 2018). However, there are no suitable diagrams with reliable PAR estimates from that region. It is thus difficult to judge whether high modern trap values are associated with recent land-use change or are characteristic of eastern European forests. 25 The comparison of regional PAR between traps and fossil estimates indicates higher fossil PAR of Picea, Fagus and Abies in middle altitudes of the temperate zone (Fig. 5) (Abraham et al., 2016;Carter et al., 2018). Within a 60 km radius of the fossil sites, Picea decreased in abundance from 70% during the Middle Holocene to 43%, compared to modern abundance. Fagus and Abies declined from Late Holocene values of 22% and 3% to currently 20% to 1% respectively (Abraham et al., 2016). The abundance of Abies in the Roztocze region (SE Poland; Fig. 6) provides a good analogue for the past abundance of the tree in Šumava with maximum PAR of 1000-3000 grains cm -2 y -1 . Abies disappeared from the Czech Republic during the Mediaeval Age due to forest management methods (Kozáková et al., 2011), which were not practiced in south-east Poland.
Linking the fossil to modern PAR values facilitates the interpretation of the fossil record of individual sites. Unfortunately, the details cannot be discussed here. However, the central Swedish sites Holtjärnen and Klotjärnen provide excellent examples.

5
These sites are situated north of the modern distribution of Tilia, Corylus, Quercus and near the limit of Alnus glutinosa. The fossil PAR values are higher for these taxa than those found in pollen traps at or near these lakes (Giesecke, 2005a;Giesecke and Fontana, 2008 (Fig. 5).

Limitations and problems
Nevertheless, there are significant differences between the accumulation of pollen in traps and on peatlands and lakes (Lisitsyna et al., 2011b;Pardoe et al., 2010). Differences in pollen trap design and placement in the landscape will influence the values. 15 Trap values are also affected by modern processes that have no impact on the fossil signal, such as pollen from trapped insects.
These biases appear minor as indicated by the large consistency of the data collected in the PMP database. Also the comparison of values over this large environmental gradient results in the signal being stronger than the noise. Nevertheless, some traps or individual years have unusual values and were removed from the comparison (Table S1). Despite this, the uncertainty of fossil PAR values is much greater than pollen traps, which is primarily due to the added uncertainty coming from sampling a 20 sediment core, combined with the uncertainty of the age model (Maher, 1981). PAR from lake sediment has additional biases due to differential sedimentation of pollen grains in lakes (Davis and Brubaker, 1973), sediment re-deposition, focussing and catchment erosion (Davis et al., 1984;Pennington, 1979). Although we carefully selected the best available fossil sites, PAR especially from lake Suminko and Rõuge Tõugjärv may be biased by lake internal processes and the addition of stream borne pollen respectively. Nevertheless, their fossil PAR estimates are in the range of values found in pollen traps. Where detailed 25 knowledge of the sedimentation process is available, the bias of sediment focussing may be reduced, as in the example of Hockham Mere cited above (see also Bennett and Buck, 2016;Bennett, 1983). Peatlands may thus seem the better choice for obtaining fossil PAR, which may be the case in northern Scandinavia (Barnekow et al., 2007;Finsinger et al., 2013), but frequent changes in the rate of peat growth lead to difficulties assessing the time represented in individual samples at many sites.

30
The problem of traps collecting large amounts of herbaceous pollen brought by insects and small animals was discussed in the method section and for this reason Poaceae and Cyperaceae are the only herbs selected for our analyses. However, pollen from these two families is also often overrepresented in the pollen traps (Lisitsyna et al., 2011b), as the plants may overhang the trap opening and their pollen may fall directly into the trap. Reduced PAR in the trap may be caused by overgrowth of the vegetation or leaves temporally blocking the opening, while proximity to the forest edge would increase values compared to large open peatlands or lakes. These effects have not been systematically evaluated so far.
Detailed comparisons of vegetation data to PAR hold potential for a better understanding of the spatial representation and processes shaping the pollen signal (Matthias and Giesecke, 2014) and allow estimates of absolute pollen productivity (Sugita et al., 2009) or test pollen dispersal models. However, for this continental scale dataset, available vegetation data have limited 5 precision. Forest inventory data with the detail essential for this type of study are not available for all traps. The forest cover data presented here has a resolution of 1 km 2 , which is insufficient as the abundance of trees within hundreds of meters of the traps is important. Moreover, without information on standing volume or age structure, the percentage cover used here is a crude measure of the vegetation producing the pollen. Forestry practices like harvesting trees that start flowering at a later age (e.g. Picea 30-40 years) reduce the number of trees producing pollen (Matthias et al., 2012) and bias the search for modern 10 analogues. Also, the available mapped distribution limits of trees have large uncertainties precluding more detailed assessments of the quantity of long distance transported pollen using this continental dataset.

Conclusions
Comparison of the mean annual PAR from traps and fossil sites showed similar ranges for Abies, Alnus, Betula, Carpinus, Corylus, Fagus, Fraxinus, Picea, Pinus, Quercus and Tilia at the continental scale. This indicates that there are no major biases 15 hampering the application of the PMP Database data as a modern reference to interpret the fossil record. The dataset clearly shows that climate parameters correlate with latitude in determining pollen productivity. The effect of regional forest cover is discernible. Minimum values suggest that an 80% forest cover within 10 km of the trap results in PAR above 3200 tree pollen grains cm -2 year -1 .
Assessment of long-distance transport indicates that values below 80 grains cm -2 y -1 for Carpinus, Corylus, Fagus, Fraxinus, 20 Picea, Quercus and Tilia may originate from beyond 200 km of a sampling site. This number of 80 grains cm -2 y -1 may therefore be used as a general threshold indicating long distance origin of pollen. The application of these threshold values holds potential to refine and adjust reconstructions of tree distributions.