plants Article A Continuous Centennial Late Glacial-Early Holocene (15–10 cal kyr BP) Palynological Record from the Iberian Pyrenees and Regional Comparisons Valentí Rull 1,2, * , Arnau Blasco 3 , Miguel Ángel Calero 4 , Maarten Blaauw 5 and Teresa Vegas-Vilarrúbia 3 1 Botanic Institute of Barcelona, Spanish National Research Council (CSIC), Pg. Migdia s/n, 08038 Barcelona, Spain 2 Institut Català de Paleontologia Miquel Crusafont, Universitat Autònoma de Barcelona, C. Columnes ICTA-ICP Bld., 08193 Cerdanyola del Vallès, Spain 3 Department of Evolutionary Biology, Ecology and Environmental Sciences, Universitat de Barcelona, Av. Diagonal 643, 08028 Barcelona, Spain; arnaublasco@ub.edu (A.B.); tvegas@ub.edu (T.V.-V.) 4 Department of Geography, Universitat de Barcelona, C. Montalegre 6, 08001 Barcelona, Spain; miquel_angel_calero@ub.edu 5 14 CHRONO Centre for Climate, the Environment and Chronology, School of Natural and Built Environment, Queen’s University Belfast, 42 Fitzwilliam St., Belfast BT7 1NN, UK; maarten.blaauw@qub.ac.uk * Correspondence: vrull@csic.es Abstract: This paper presents the first continuous (gap-free) Late Glacial-Early Holocene (LGEH) pollen record for the Iberian Pyrenees resolved at centennial resolution. The main aims are (i) to provide a standard chronostratigraphic correlation framework, (ii) to unravel the relationships be- tween vegetation shifts, climatic changes and fire, and (iii) to obtain a regional picture of LGEH vegetation for the Pyrenees and the surrounding lowlands. Seven pollen assemblage zones were obtained and correlated with the stadial/interstadial phases of the Greenland ice cores that serve as a global reference. Several well-dated datums were also derived for keystone individual taxa that are useful for correlation purposes. Four vegetation types were identified, two of them corresponding to Citation: Rull, V.; Blasco, A.; Calero, conifer and deciduous forests (Cf, Df) and two representing open vegetation types (O1, O2) with no M.Á.; Blaauw, M.; Vegas-Vilarrúbia, T. modern analogs, dominated by Artemisia-Poaceae and Saxifraga-Cichorioideae, respectively. Forests A Continuous Centennial Late dominated during interstadial phases (Bølling/Allerød and Early Holocene), whereas O1 dominated Glacial-Early Holocene (15–10 cal kyr during stadials (Oldest Dryas and Younger Dryas), with O2 being important only in the first half of BP) Palynological Record from the the Younger Dryas. The use of pollen-independent proxies for temperature and moisture allowed the Iberian Pyrenees and Regional reconstruction of paleoclimatic trends and the responses of the four vegetation types defined. The Comparisons. Plants 2023, 12, 3644. most relevant observation in this sense was the finding of wet climates during the Younger Dryas, https://doi.org/10.3390/ which challenges the traditional view of arid conditions for this phase on the basis of former pollen plants12203644 records. Fire incidence was low until the Early Holocene, when regional fires were exacerbated, Academic Editors: Luca Giupponi probably due to the combination of higher temperatures and forest biomass accumulation. These and Annamaria Giorgi results are compared with the pollen records available for the whole Pyrenean range and the sur- Received: 28 September 2023 rounding lowlands within the framework of elevational, climatic and biogeographical gradients. Revised: 19 October 2023 Some potential future developments are suggested on the basis of the obtained results, with an Accepted: 20 October 2023 emphasis on the reconsideration of the LGEH spatiotemporal moisture patterns and the comparison Published: 22 October 2023 of the Pyrenees with other European ranges from different climatic and biogeographical regions. Keywords: Pyrenees; palynology; pollen stratigraphy; climate change; fire; vegetation response; Late Glacial; Early Holocene Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons 1. Introduction Attribution (CC BY) license (https:// In the Iberian Peninsula (IP), the Late Glacial, or the Last Glacial Maximum (LGM)- creativecommons.org/licenses/by/ Holocene transition, is an excellent time interval to study the response of vegetation to 4.0/). natural drivers of ecological change due to the occurrence of conspicuous climatic reversals Plants 2023, 12, 3644. https://doi.org/10.3390/plants12203644 https://www.mdpi.com/journal/plants Plants 2023, 12, x FOR PEER REVIEW 2 of 30 1. Introduction Plants 2023, 12, 3644 2 of 29 In the Iberian Peninsula (IP), the Late Glacial, or the Last Glacial Maximum (LGM)- Holocene transition, is an excellent time interval to study the response of vegetation to natural drivers of ecological change due to the occurrence of conspicuous climatic rever- and sals the andlack the of significant lack human of significant pressure human [1]. This pressure [1].is This especially true fortrue is especially mountain ranges, for mountain where ranges,extensive landscape where extensive anthropization landscape did notdid anthropization begin not until beginthe untilLate theHolocene Late Holocene[2,3]. In [2,3]. In the Iberian Pyrenees, anthropization began during the Bronze Age (~3.7 cal kyr BP) the Iberian Pyrenees, anthropization began during the Bronze Age (~3.7 cal kyr BP) in the southern in the southernlowlands lowlands[4][4]andand progressed progressed northward northward following following an altitudinal an altitudinal trendtrend char- characterized by rates acterized by rates of m of 40 40 elevation m elevation perper century, century, onon average average [5]. [5]. Therefore, Therefore, climatewas climate wasa amajor majorenvironmental environmentalforceforcefor forPyrenean Pyreneanvegetation vegetationduring during the LGM-Holocene the LGM-Holocene transition.transition. Globally, Globally,thetheLate LateGlacial Glacialhas been has beendefined in Greenland defined in Greenland ice cores by a stadial-interstadial ice cores by a stadial-inter- succession that characterized the transition from the LGM to the Holocene, stadial succession that characterized the transition from the LGM to the Holocene, which is known which as is known as the INTIMATE (INTegration of Ice-core, MArine and TErrestrial event the INTIMATE (INTegration of Ice-core, MArine and TErrestrial record) record)stratig- event raphy [6] (Figure stratigraphy 1). The1). [6] (Figure classical sequence, The classical which which sequence, serves serves as a worldwide stratigraphic as a worldwide strati- reference, includes includes graphic reference, the stadial theGS-2.1a stadial or OldestorDryas GS-2.1a Oldest (OD) Dryas(17,500–14,700 cal yr BP), (OD) (17,500–14,700 calthe yr interstadial GI-1 or Bølling/Allerød (B/A) (14,700–12,900 cal yr BP)—which BP), the interstadial GI-1 or Bølling/Allerød (B/A) (14,700–12,900 cal yr BP)—which has has been subdi- vided into minor events GI-1a to GI-1e—and the stadial GS-1 (12,900–11,700 cal yr BP) or been subdivided into minor events GI-1a to GI-1e—and the stadial GS-1 (12,900–11,700 Younger Dryas (YD), which was the last Pleistocene cooling reversal that preceded the Early cal yr BP) or Younger Dryas (YD), which was the last Pleistocene cooling reversal that Holocene (EH) warming. preceded the Early Holocene (EH) warming. Figure 1. Late Glacial-Early Holocene stratigraphy based on Greenland ice core NGRIP2 [6] com- Figure 1. Late Glacial-Early Holocene stratigraphy based on Greenland ice core NGRIP2 [6] compared pared with the speleothem records from Ostolo cave (OST2), on the westernmost Pyrenean end [7], with the speleothem records from Ostolo cave (OST2), on the westernmost Pyrenean end [7], and and Seso cave, on the central Pyrenees [8] (Figure 2). The glacier dynamics, as reconstructed in the Seso Rudacave, on(parallel valley the central Pyrenees to the [8] (Figure Aiguamòg valley,2). The glacier where dynamics, the record as reconstructed analyzed in theisRuda in this study lies), indi- valley (parallel to the Aiguamòg valley, where the record analyzed cated for reference [9]. GA, glacier advance; OD, Oldest Dryas. in this study lies), is indicated for reference [9]. GA, glacier advance; OD, Oldest Dryas. Plants 2023, 12, Plants 2023, 12, 3644 x FOR PEER REVIEW 33of of 30 29 Figure 2. The Pyrenean range within the Iberian Peninsula. (A) The Iberian Maladeta Massif, with Figure 2. The Pyrenean range within the Iberian Peninsula. (A) The Iberian Maladeta Massif, with the highest Pyrenean peak (Aneto; 3404 m) in the center, as seen from near the Bassa Nera (1890 m) the highest Pyrenean peak (Aneto; 3404 m) in the center, as seen from near the Bassa Nera (1890 m) coring site (Photo: V. Rull). (B) Topographic map of the Pyrenees indicating the Bassa Nera coring coring site (BN) site (Photo: and other V. Rull). (B) localities Topographic of the mapthe Pyrenees and of surrounding the Pyreneeslowlands indicating the Late with Bassa Nera coring Glacial pollen site (BN)The records. andlocations other localities of the of the Pyrenees Ostolo (Os), Sesoand theand (Se) surrounding Moulis (Mo) lowlands with Late paleoclimatic Glacial records arepollen indi- records. cated The locations by white diamonds, of the and Ostolo (Os), peak the highest Seso of (Se) theand Moulis(At, Pyrenees (Mo) paleoclimatic Aneto) is markedrecords are by a black indicatedThe triangle. by white black diamonds, and the highest line is the boundary peakSpain between of theand Pyrenees France,(At, Aneto) which is marked represents theby a black approxi- mate boundary triangle. The blackbetween line is the the northern boundaryand southern between Spain Pyrenean sides and France, (AND, which Andorra), represents theand the white approximate line is the boundary boundary between the between northern theand Eurosiberian (EURO) and southern Pyrenean sidesMediterranean (AND, Andorra), (MED)and biogeographical the white line is regions. the boundaryLate Glacial betweensites: Ba, Balcère; Bi, the Eurosiberian Bious;and (EURO) Bl, Mediterranean Banyoles; Bo, Borde; (MED)Br, Barbazan; Bt, Bentué; biogeographical regions. Bu, Búbal; By, Biscaye; Cs, Castet; Ec, Ech; Es, Estanya; Et, Estarrès; Fr, Freyxinède; Go, Gourg- Late Glacial sites: Ba, Balcère; Bi, Bious; Bl, Banyoles; Bo, Borde; Br, Barbazan; Bt, Bentué; Bu, Búbal; Nègre; La, Laurenti; Lo, Lourdes; Ma, Marboré; Mg, Monge; Pc, Parco; Pr, Portalet; Pt, La Pouretère; By, Biscaye; Cs, Castet; Ec, Ech; Es, Estanya; Et, Estarrès; Fr, Freyxinède; Go, Gourg-Nègre; La, Ra, Racou Tr, Tramacastilla (details and references in Appendix A). Sites with charcoal records as Laurenti; fire proxies Lo,are Lourdes; Ma, Marboré; underlined. (C) Map Mg, Monge; of the IberianPc, Parco; Pr,showing Peninsula Portalet;the Pt, location La Pouretère; of theRa, Racou Pyrenees Tr, Tramacastilla (details and references in Appendix A). Sites with charcoal records (red box). The Mediterranean (Med) and Eurosiberian (Euro) biogeographical regions are indicated as fire proxies are underlined. by pale yellow (C)and Map of the green Iberian areas, Peninsula showing the location of the Pyrenees (red box). The respectively. Mediterranean (Med) and Eurosiberian (Euro) biogeographical regions are indicated by pale yellow In theareas, and green Pyrenees and surrounding lowlands, the Late Glacial-Early Holocene (LGEH) respectively. palynostratigraphic and paleovegetational trends are relatively well known on the north- In the ern side butPyrenees and surrounding largely unknown lowlands, on the southern the Late flank Glacial-Early (Figure Holocenefrom 2). A first synthesis (LGEH) the palynostratigraphic and paleovegetational trends are relatively well known on northern French slope [10] reported that the Late Glacial began with the expansion of Bet- the northern sidethe ula, butdecrease largely unknown on the southern in Chenopodiaceae and the flank (Figure 2). A first of decrease/disappearance synthesis from Ephedra at the the in- northern French slope [10] reported that the Late Glacial began with the expansion itial Bølling. In some localities, Juniperus and Pinus also increased. Open cold/dry steppe of Betula, the decrease in Chenopodiaceae and the decrease/disappearance environments were represented by Artemisia, which remained relatively constant, and in- of Ephedra at the initial Bølling. In some localities, Juniperus and Pinus also increased. creasing amounts of Poaceae, Apiaceae and Rumex. In the drier Mediterranean areas (east-Open cold/dry steppe environments were represented by Artemisia, which remained relatively constant, ern sector), steppe assemblages were dominated by Artemisia, whereas in the wetter At- and increasing amounts of Poaceae, Apiaceae and Rumex. In the drier Mediterranean lantic areas (western sector), the dominant were the Poaceae. A manifest Pinus increase areas (eastern sector), steppe assemblages were dominated by Artemisia, whereas in the coeval with a Juniperus decrease was recorded at the B/A transition, when Quercus began wetter Atlantic areas (western sector), the dominant were the Poaceae. A manifest Pinus to establish and Corylus was still absent. During the YD, steppe elements (Juniperus, Arte- increase coeval with a Juniperus decrease was recorded at the B/A transition, when Quercus misia, Chenopodiaceae, Ephedra) expanded again, while forest trees (Betula, Pinus) de- began to establish and Corylus was still absent. During the YD, steppe elements (Juniperus, clined. This YD vegetation opening was more pronounced in the Mediterranean areas due to the occurrence of drier conditions. At the end of the YD, Artemisia declined and Ephedra Plants 2023, 12, 3644 4 of 29 Artemisia, Chenopodiaceae, Ephedra) expanded again, while forest trees (Betula, Pinus) declined. This YD vegetation opening was more pronounced in the Mediterranean areas due to the occurrence of drier conditions. At the end of the YD, Artemisia declined and Ephedra disappeared due to the onset of Holocene warming, characterized by the massive colonization of the Pyrenees by deciduous forests (Betula, Corylus, Quercus). Further high-resolution studies on some localities of the easternmost French Pyrenean side recorded an expansion of steppe elements (Artemisia, Ephedra, Poaceae, Apiaceae, Caryophyllaceae and others) and a decline in Pinus during the OD stadial [11]. The same studies documented a regional expansion of Juniperus, Betula and Salix during the initial B/A and a significant Pinus increase at the end of this interglacial. The YD stadial was characterized by a sudden decrease in Pinus and Betula and the expansion of steppe elements, especially Artemisia. The incoming warmer conditions during the Early Holocene were marked by a significant increase in Betula followed by Quercus and Corylus, which marked the initial colonization of the region by temperate forests. Oscillations in Pinus during Late Glacial times would have been largely due to fluctuations in long-distance pollen transport from low elevations, but the local presence of Pinus forests during the Early Holocene was supported by the presence of stomata of Pinus uncinata [11]. In general, increases in pollen from forest trees are considered to be indicative of forest expansions from neighboring refugia, which would have survived the last glaciation on sites with particular microclimatic conditions. These studies refined the existing chronologies and allowed a more precise dating of significant events such as the maximum of Artemisia (~18.1 cal kyr BP), the rise of Betula (~14.7 cal kyr BP), Quercus (~12.3 cal kyr BP) and Corylus (~11.5 cal kyr BP), or the first decline of Betula (~12.8 cal kyr BP), among others [12]. A more recent study identified similar trends [13], but the low resolution of the LGEH sequence and the existence of a single radiocarbon date in this interval prevent detailed correlations. A more recent multicentennial record identified the following succession: Juniperus-Artemisia (OD), Pinus-Betula-Poaceae (B/A and YD), Pinus-Betula-Quercus-Corylus (Early Holocene) [14]. This work included charcoal analysis and documented no fire activity during the OD, low to medium fire incidence during the B/A and the YD, and a sudden fire increase during the Early Holocene. These patterns were explained in terms of climate–vegetation interactions, with the former as the main driver and the latter as a modulator of the amplitude of fire regimes. In this way, the abrupt fire increase at the beginning of the Holocene was linked to the expansion of temperate deciduous woodlands. On the southern Pyrenean side, only four LGEH palynological sequences are available, all of them on the western side (Figure 2). Three of these records (Marboré, Búbal and Tramacastilla) are of multicentennial resolution, and the LGEH stadials/interstadials are not well defined palynostratigraphically. In addition, in the Búbal and Tramacastilla records, chronological correlations among similar palynological zones are problematic despite their proximity (barely 6 km distant), which was attributed to methodological dating issues [15]. The fourth record, Portalet, is well bracketed chronologically and of higher resolution but incomplete, as the YD is missing by a sedimentary gap, which was attributed to the permanent freezing of the Late Glacial paleolake [16]. The lack of highland YD palynological records is the norm on the southern side, where this climatic reversal is represented only in the lowlands by cold-steppe elements (notably Artemisia, Poaceae, Chenopodiaceae and Juniperus); see the review in Ref. [17]. Therefore, a chronologically well-bracketed complete and continuous LGEH palynological sequence of centennial or higher resolution for the Iberian Pyrenees is lacking. In general, increases in pollen from forest trees (Pinus, Betula, Corylus, Quercus) similar to those documented for the northern Pyrenean slopes occurred in the Iberian Pyrenees during warming phases (B/A, EH). These increases are believed to reflect temperate forest expansions from nearby glacial refugia situated in the southern lowlands (<800 m) as part of a highly diverse mosaic landscape composed of conifer forests, mixed pine-oak forests, parklands, grasslands and xerophytic steppes [18,19]. Similar to the northern side, LGEH Pyrenean records from the Iberian side usually lack charcoal records as fire proxies, except Plants 2023, 12, 3644 5 of 29 for the Portalet sequence, where Late Glacial wildfire incidence was lower than in the Early to Mid-Holocene, probably owing to the combination of higher temperatures and increased biomass fuel due to the expansion of deciduous forests since 10 cal kyr BP [20]. Other palynological parameters and proxies unusual on both northern and southern Pyrenean slopes are pollen accumulation rates (PAR), as a proxy for plant cover [21], and non-pollen palynomorphs (NPP), which have been demonstrated to be very useful for environmental reconstruction [22,23]. This paper presents the Bassa Nera record, which is the first well-dated complete and continuous (gap-free) Late Glacial-Early Holocene (~15,200–10,000 cal yr BP) palynological record of vegetation change at centennial resolution from the central Iberian Pyrenees, in- cluding PAR, charcoal and NPP analyses. The main Holocene trends (10 cal kyr BP onward) in the development of the Bassa Nera vegetation have already been studied [24–27], but the LGEH ecological history remains unknown. The main aims of this study are (i) to provide a standard LGEH palynological sequence for the central Pyrenees, which may serve correlation and dating purposes; (ii) to characterize the response of vegetation to local and regional climatic shifts in the absence of human disturbance, which may be useful to disentangle natural and anthropogenic drivers of ecological change; (iii) to document the fire regime history and its potential climatic and vegetational drivers; and (iv) to situate the results obtained in a regional context by comparing them with former studies on the northern and southern Pyrenean sides, including the surrounding lowlands. In addition to providing clues on the origin of ecosystems and their sensitivity to natural forces, knowing the responses of preanthropic vegetation to climate changes is useful to anticipate the potential responses of natural communities to future climatic change. 2. Study Area 2.1. Present Conditions Bassa Nera is a small pond (ca. 100 m long, 60 m wide and 5 m deep) located in a small depression known as Pletiu dera Muntanheta, situated in the headwaters of the Aiguamòg valley (upper Garona basin), at 42◦ 380 1700 N–0◦ 550 2700 E and 1890 m elevation. Although it belongs to Spain (Val d’Aran), the site is located on the northern Pyrenean slope, near the Pyrenean ridge. In some maps, the pond is also called Estanh dera Muntanheta, but we prefer to use Bassa Nera because this is the most utilized name in recent botanical and pale- oecological publications. The pond is in its later infilling stages and is surrounded by a mire. The site is situated in the peripheral protection zone of the National Park “Aigüestortes i Estany de Sant Maurici” (PNAESM). The Pletiu dera Muntanheta is of glacial origin and is excavated on the granodioritic bedrock that is part of the Carboniferous-Permian Maladeta batholith [28,29]. The climate is alpine with oceanic influence, and the regional 0 ◦ C isotherm is estimated to be at 2950 m elevation [30]. The nearest meteorological station is Port de la Bonaigua, situated approximately 6 km NE and at 2072 m elevation, 182 m above Bassa Nera. The annual average temperature for the 1990–2020 period is 3.34 ◦ C, and the total annual precipitation is 1218 mm. Minimum average temperatures (−2.9 to −3.2 ◦ C) are recorded in January and February, and maxima (11.6 ◦ C) correspond to July and August. Total precipitation shows a bimodal pattern with maxima (131–137 mm) in May and November and minima (70–79 mm) in February and July. Biogeographically, the Bassa Nera is located in the southern margin of the Eurosiberian region, characterized by wet temperate Atlantic macrobioclimates and dominated by plant species distributed across central and northern Europe. This contrasts with the Mediter- ranean biogeographic region, under the influence of Mediterranean climates with at least two consecutive arid months during the warmest season (summer). The Mediterranean biome dominates most of the Iberian Peninsula (Figure 2) and is characterized by plant species distributed across southern Europe and northern Africa [31]. A study of plant communities along the Aiguamòg valley ranging between approx- imately 1300 m and 2600 m elevation, which included the Pletiu dera Muntanheta, rec- ognized the following vegetation stages: mixed forests of Pinus sylvestris and Betula Plants 2023, 12, 3644 6 of 29 (1300–1600 Plants 2023, 12, x FOR PEER REVIEW m); conifer forests of Pinus mugo and Abies (1600–2000 m); mosaic 6 of 30 vegeta- tion of Pinus mugo, Rhododendron and Nardus (2000–2250 m); and alpine meadows/bogs with Festuca, Nardus, Carex, Luzula and Juncus [32] (Figure 3). In this transect, the Bassa Nera liesCarex, Nardus, within the Pinus-Abies Luzula forest and Juncus [32] belt,3).~350 (Figure m transect, In this below thethetree line, Bassa which Nera is situated at lies within 2250 the Pinus-Abies forest belt, ~350 m below the tree line, which is situated at 2250 m. Aseven m. A further study specifically of the Pletiu dera Muntanheta identified fur- plant associations [33]. The pond ther study specifically of the margins areMuntanheta Pletiu dera characterized by Sphagnum identified seven plantpeats, Carex-Potentilla associations [33]. The pond associations, ormargins are characterized Sphagnum-Carex bydepending carpets, Sphagnum peats, Carex-Potentilla on the local features.associations, The surrounding areas are covered by marshes and grasslands dominated by Carex, Scirpusareas or Sphagnum-Carex carpets, depending on the local features. The surrounding are or Succisa, with covered by marshes and grasslands dominated by Carex, Scirpus or Succisa, scattered Sphagnum hummocks. Pinus-Abies forests with a Rhododendron-dominated under- with scattered Sphagnum story grow hummocks. on encirclingPinus-Abies forests slopes. The same with study a Rhododendron-dominated highlighted the uniqueness understory of the site in grow on encircling slopes. The same study highlighted the uniqueness of the site in terms terms of plant richness in comparison with other Pyrenean mires and attributed this fact to of plant richness in comparison with other Pyrenean mires and attributed this fact to the the combination of hydrological and water chemistry variability. combination of hydrological and water chemistry variability. Figure3.3. Modern Figure Modern pollen pollenspectra along spectra alongan an elevational transect elevational including transect the Bassa including theNera coring Bassa Nerasite. coring site. Only major components (>2% of the total) are shown. Forest trees are in green, and shrubs/herbs are Only major components (>2% of the total) are shown. Forest trees are in green, and shrubs/herbs are in orange. Cyperaceae (blue) were not included in the pollen sum. Simplified from Ref. [32]. in orange. Cyperaceae (blue) were not included in the pollen sum. Simplified from Ref. [32]. 2.2. Previous Studies 2.2. Previous Studies The Pletiu dera Muntanheta was included in a modern-analog study developed on The Pletiu an altitudinal dera (ca. transect Muntanheta 870–2600 mwas included elevation) inthe along a modern-analog Aiguamòg valley, study developed using sur- on face sediments and moss polsters. This survey was aimed at establishing qualitative and using an altitudinal transect (ca. 870–2600 m elevation) along the Aiguamòg valley, surface sediments quantitative and moss relationships polsters. between extantThis surveypatterns vegetation was aimed at establishing and modern qualitative pollen sedi- and quantitative mentation, relationships which may between be useful for extant vegetation the interpretation patterns of past pollen and records in modern terms of pollen sedimentation, vegetation changewhich [32].may The be useful lower for thevegetation montane interpretation of past pollen belt (800–1300 m) wasrecords in terms of character- ized by Alnus, vegetation Betula, change deciduous [32]. The lowerQuercus, Fraxinus montane and Salixbelt vegetation pollen, which is (800–1300 m)absent in the was characterized upper by levels Alnus, (Figure Betula, 3). In the middle deciduous and Fraxinus Quercus, upper montane levelspollen, and Salix (1300–1600 m), the which pollen in the is absent assemblages were almost exclusively composed of Pinus, with lower amounts upper levels (Figure 3). In the middle and upper montane levels (1300–1600 m), the pollenof Corylus. The subalpinewere assemblages belt (1600–2300 m; below the almost exclusively tree line) was composed completely of Pinus, with dominated by Pinus, lower amounts of Corylus. with low percentages of grass and sedge pollen, whereas the pollen of deciduous trees The subalpine belt (1600–2300 m; below the tree line) was completely dominated by Pinus, with low percentages of grass and sedge pollen, whereas the pollen of deciduous trees from montane belts was virtually absent. The situation was similar in the alpine belt Plants 2023, 12, 3644 7 of 29 (2300–2600 m; above the tree line) but with higher amounts of herbaceous plants, which may be locally more abundant than Pinus. The best altitudinal indicators were Fraxinus and Salix, whose pollen is restricted to the montane and subalpine belts, respectively. Betula and Corylus pollen was restricted to the montane belt, together with deciduous Quercus, Tilia, Alnus, Buxus and Brassicaceae. Pollen types almost restricted to the alpine belt were Plantago, Calluna, Poaceae, Asteraceae and Cyperaceae, which may serve as tree line indicators. Interestingly, the pollen of Olea and evergreen Quercus have their maxima in the subalpine belt, although their parent plants are characteristic of lowlands and were not found in the vegetation gradient analyzed, suggesting efficient upward wind dispersal. In the same transect, the modern sedimentation of NPP was studied to unravel their potential as paleoecological indicators [26]. NPP assemblages resulted to be influenced not only by elevation but also by local moisture conditions, landscape openness and human pressure, especially grazing. Previous studies in the Pletiu dera Muntanheta based on cores retrieved in the mire surrounding Bassa Nera recorded the vegetation shifts of the last ~10,000 years. These studies demonstrated that human pressure was low until the Late Bronze Age. In- deed, the first evidence of grazing dates to 7300 cal yr BP, and the oldest cereal records were found at ~5190 cal yr BP, but general landscape anthropization did not occur until 3150–2650 cal yr BP [25]. No archaeological evidence of human occupation of the PNAESM prior to the Holocene is available, and Neolithization began by ~7250 cal yr BP at lower elevations (<1600 m) and ~5350 cal yr BP at higher elevations [34]. No paleoecological studies are available for the LGEH, but a recent glaciological survey near Bassa Nera provides the paleoclimatic context for the interpretation of the palynological studies presented in this paper. LGEH chronology and paleoclimatology were resolved in detail in the adjacent Ruda valley using 10 Be cosmic-ray exposure dating of moraine boulders and polished surfaces, combined with glacier equilibrium-line altitude calculations [9]. After the LGM, a general glacier retreat occurred that exposed areas at elevations of approximately 1860–1900 m between 15 and 14 cal kyr BP. The Bassa Nera lies within this elevational range; thus, it can be inferred that the Pletiu dera Muntanheta was covered by glaciers until ~15 cal kyr BP, when our pollen record begins. Therefore, this record may be the oldest possible to be retrieved in the Bassa Nera sediments and represent periglacial environments. The deglaciation trend continued until the YD and the Holocene but was interrupted by two glacier advances/stillstands at 13.5 and 13.0 cal kyr BP, represented by moraine systems situated at 2080 m and 2190 m elevation, which means that the glacier fronts were 190 m and 300 m above Bassa Nera, respectively. The deglacial trend exposed the surface of the highest cirque floors of the Ruda valley (2300–2500 m elevation) by 12.7–12.6 cal kyr BP, at the onset of the YD [9]. 3. Methods 3.1. Coring and Dating The Bassa Nera core used in this study (PATAM12-A14) was retrieved in July 2014 with a 50 mm diameter Livingstone corer. This core consisted of 10 drives (D1 to D10) for a total length of 845 cm (1102 cm once corrected for compression). This study is focused on the basal sequence (drives D7 to D9), which contains the LGEH interval. Eight samples for radiocarbon dating were taken in this interval and submitted to acid/base digestion (KOH, HCl, HF) to remove the rejuvenating effect of eventual incorporation of younger plant material and/or humic acids percolating through the peat profile [35]. AMS radiocarbon dating was performed at the Radiochronology Lab of the Université Laval (Canada). Calibration was performed with CALIB 8.2 and IntCal 20 [36]. The Bayesian age-depth model was built with the software Bacon 3.1.0 [37]. 3.2. Palynological Analysis For pollen analysis, 51 samples (2 cm3 ) were taken in the LGEH interval with a volumetric syringe at regular intervals of approximately 100 years once the age-depth Plants 2023, 12, 3644 8 of 29 model was known. These samples were digested with NaOH for disaggregation, HCl for carbonate removal and HF for silicate removal, and submitted to centrifugation with Thoulet solution (density 2 g/cm3 ) for mineral separation [38]. Two tables of Lycopodium spores (Batch 280521 291; 13,761 ± 448 spores/tablet) were added to each sample before processing [39]. Processing was performed at the Archaeobotany Lab of the Catalan Institute of Human Paleoecology and Social Evolution (IPHES). Pollen/spore identification was based on usual pollen catalogues for the region [40,41] and the author’s (VR) own reference collection, which contains over 700 species/subspecies from the Iberian flora. NPP identification was based on our previous research [26,27] and the NPP Image Database (https://non-pollen-palynomorphs.uni-goettingen.de/; last accessed on 3 May 2023). Most fungal spores were not identified, but special attention was placed on those considered to be indicators of herbivory and, therefore, of human impact, such as Sporormiella, Sordiaria and Podospora [42–45]. Pinus stomata were identified after García Álvarez et al. [46]. Counting was conducted up to a minimum pollen sum of 300 grains, stabilization of confidence intervals and saturation of diversity [47]. In a number of samples, Pinus was superabundant, and its counts were stopped at 100–150 grains and further extrapolated using the ratio between the introduced and the counted Lycopodium spores. In those cases, the above statistical criteria were applied to the remaining pollen types. The pollen sum included all pollen types with the exception of aquatic and semiaquatic plants (Cyperaceae, Typha, Ludwigia). Paleoecological interpretation was based on the available modern-analog studies for both pollen/spores and NPP developed on the elevational transect that includes the study site [26,32]. Pollen accumulation rates (PAR) are used as a proxy for vegetation cover [21]. It is important to stress that pollen is not used as a climatic proxy, as usual in most Pyrenean literature. This is due to the lack of sufficient taxonomic resolution (usually at the genus or family level) and the fact that such a procedure may lead to circularity when analyzing the response of vegetation to changing climates. Paleoclimatic trends were inferred from pollen-independent proxies such as the OST2 (temperature) and Seso (moisture) speleothem records (Figures 1 and 2). Unfortunately, independent paleoclimatic proxies for hydrological balance are available only for the Younger Dryas [8,48]. Charcoal particles above 5 µm were also recorded and used as fire proxies. Two size classes of these particles were considered, above and below 150 µm, as proxies for fires around the coring site and regional fires, respectively [49]. Zonation of pollen diagrams was performed using the optimal splitting by information content (OSIC) and broken-stick methods [50]. Cluster analysis used the Pearson linear correlation coefficient (r) and the unweighted pair-group average (UPGMA) clustering algorithm [51]. Plotting and zonation were performed with Psimpoll 4.27 [52], and statistical analyses were carried out with MVSP 3.22 [53] and Past 4.12 [54]. 4. Results 4.1. Lithology and Chronology The full core was subdivided into five sedimentary units, of which three (U3 to U5) were present in the LGEH interval studied here. These units are characterized by intercalated clay, silt and sand layers of varied appearance and composition. U3 was composed of gray clays and fine sands with plant macroremains, whereas in U4, the alternating laminae consisted of brown clays and ochre silts/sands with low organic matter content, and U5 was dominated by brown-greenish clay layers with no apparent organic matter and occasional gravel and silt/sand beds (Figure 4). Radiocarbon dating (Table 1) showed that the interval studied contained a complete and continuous (i.e., free from sedimentary gaps) LGEH sequence that lasted approximately 5000 years, from 15.3 to 10.3 cal kyr BP, and included the GS-2 (OD) stadial, the GI-1 (B/A) interstadial, the GS-1 (YD) stadial and the EH warming (Figure 1). The average sediment accumulation rate (r) was 0.63 mm yr−1 , but it was far from homogeneous, showing minimum to intermediate values (0.29–0.82 mm yr−1 ) in U3 and U5 and maximum values in U4, especially in its [36]. Lab Code Drive Sample Depth (cm) 14C yr BP Cal yr BP (2σ) ULA-10114 D7 63–64 802 9260 ± 20 10,300–10,557 ULA-10115 D7 104–105 854 9785 ± 20 11,191–11,241 Plants 2023, 12, 3644 ULA-10116 D8 19–20 874 10,040 ± 20 11,358–11,727 9 of 29 ULA-10117 D8 51–52 913 10,410 ± 20 12,102–12,482 ULA-10118 D8 94–95 964 10,635 ± 20 12,621–12,716 ULA-10119 D9 32–33 990 10,915 ± 20 12,757–12,882 −1 ). This acceleration event occurred in the first part of the YD lower part (0.90–1.60 ULA-10120 D9mm y 76–77 1047 11,710 ± 25 13,493–13,602 (12.8–12.3 cal kyr BP) ULA-10121 D9 and is abbreviated 1103 119–120 here as YDsa. 12,715 ± 25 15,049–15,281 Figure 4. Lithology, Bacon age-depth model [37] and sediment accumulation rates (SAR) of the core Figure 4. Lithology, Bacon age-depth model [37] and sediment accumulation rates (SAR) of the core analyzed in this work (PATAM12-A14). Dated samples are in blue, and the model is represented by analyzed dotted redin(mean) this work (PATAM12-A14). and gray Dated samples (95% confidence interval) lines. Theare in blue, stadial/interstadial INTIMATE and the model isstra-represented by dotted(Figure tigraphy red (mean) and 1) is also gray (95%for represented confidence interval) reference. CS, lines. core scan; The INTIMATE U3–U5, stadial/interstadial sedimentary units; YDsa, Younger Dryas stratigraphy sedimentation (Figure acceleration. 1) is also represented for reference. CS, core scan; U3–U5, sedimentary units; YDsa, Younger Dryas sedimentation acceleration. Table 1. Radiocarbon dates of the basal part of core PATAM12-A14 using pollen residues [35]. Dating: Radiochronology Lab, Université Laval, Canada (ULA). Calibration: CALIB 8.2 and IntCal 20 [36]. Lab Code Drive Sample Depth (cm) 14 C yr BP Cal yr BP (2σ) ULA-10114 D7 63–64 802 9260 ± 20 10,300–10,557 ULA-10115 D7 104–105 854 9785 ± 20 11,191–11,241 ULA-10116 D8 19–20 874 10,040 ± 20 11,358–11,727 ULA-10117 D8 51–52 913 10,410 ± 20 12,102–12,482 ULA-10118 D8 94–95 964 10,635 ± 20 12,621–12,716 ULA-10119 D9 32–33 990 10,915 ± 20 12,757–12,882 ULA-10120 D9 76–77 1047 11,710 ± 25 13,493–13,602 ULA-10121 D9 119–120 1103 12,715 ± 25 15,049–15,281 4.2. Palynology Globally, a total of 56 palynomorph types (33 pollen, 8 pteridophyte spores and 15 NPP) were identified. Total counts averaged 914 (804 pollen, 18 pteridophyte spores and 92 NPP), and the average pollen sum was 792. The pollen diagram was subdivided into seven pollen zones, which are described in chronological order. Pollen and spore percentages are shown in Figure 5, NPP percentages are depicted in Figure 6 and influx values are displayed in Figure 7. Boundary ages, in 95% confidence intervals, were estimated using the age-depth model of Figure 4. The most abundant NPP are depicted in Figures 8–10. Plants 2023, 12, 3644 10 of 29 Zone BN1 (1100 to 1061 cm; 14,961–15,300 to 13,635–14,445 cal yr BP; late GS-2/OD). Codominance of arboreal and nonarboreal elements. The dominant tree was Pinus (>40%), followed by the herbs Artemisia and Poaceae (15–20% each); all other elements were absent or below 1% of the pollen sum. These dominance relationships were due to the general scarcity of pollen, as indicated by minimum PAR values, rather than to high abundances of major types. Low percentages of fungal spores (~5%). NPP also scarce and dominated by Pediastrum (alga) and Type BN-II as the main animal remain. Minimum charcoal influx. Zone BN2 (1061 to 1001 cm; 13,635–14,445 to 12,821–13,191 cal yr BP; late GS-2/OD to GI-1/BA). Arboreal elements largely dominant (70–80%) due to the increase in Pinus (60–80%) and the decline in Artemisia and Poaceae (<10%). This dominance shift was due to an increase in Pinus influx, which was responsible for a general increase of PAR values, whereas Artemisia and Poaceae remained at values similar to Zone BN1. General percentage decrease of NPP except fungi spores, which remained similar to the former zone. Algae (especially Botryococcus, which peaked at the end of the zone) and charcoal influx slightly increased upward. Zone BN3 (1001 to 960 cm; 12,821–13,191 to 12,493–12,701 cal yr BP; early GS-1/YD). Transitional zone representing a gradual reversal to values similar to BN1, except for PAR, which showed a maintained increase and peaked at the end of the zone. In this case, however, the increase in relative abundance of Artemisia, Poaceae and Chenopodiaceae was due to the actual increase in their pollen influx, reaching a maximum at the end of the zone. Among NPP, fungi displayed an increasing trend and algae significantly increased in both percentage and influx, especially Zygnema and Pediastrum, whereas Botryococcus decreased. Charcoal influx continued its slightly ascending trend. Zone BN4 (960 to 930 cm; 12,493–12,701 to 12,164–12,524 cal yr BP; middle GS-1/YD). Dominance of nonarboreal elements (~70%) and significant decline in PAR values after the ending-BN3 peak. Apparent decrease in Pinus, and slight decline in Artemisia, Poaceae and Chenopodiaceae, in both percentage and influx. The situation was similar to BN3 but Cichorioideae, Cyperaceae and Saxifraga experienced a slight influx increase, which resulted in a meaningful percentage increase, followed by other Asteraceae, Caryophyllaceae and Apiaceae. Fungal spores showed their maximum percentages (although this was not due to a conspicuous influx increase), whereas algae and animal remains remarkably declined reaching their minimum values, in both influx and percentage. Charcoal slightly decreased. Zone BN5 (930 to 892 cm; 12,164–12,524 to 11,584–12,105 cal yr BP; late GS-1/YD). Another transitional zone showing trends opposite to BN3, that is, increasing arboreal and decreasing nonarboreal pollen. PAR values did not recover completely and remained below BN3. First significant increase in Betula at the middle of the zone and further increase in both percentage and influx. Nonarboreal elements recovered the situation of BN3 due to the strong decline of Cichorioideae, Cyperaceae, Saxifraga, other Asteraceae, Caryophyllaceae and Apiaceae. Fungi declined to former BN3 percentage values and all algae, especially the dominant Zygnema, underwent significant increases, peaking at the end of the zone. Among animal remains, only Type BN-II increased reaching values similar to BN1. Charcoal slightly declined to intermediate values. Zone BN6 (892 to 842 cm; 11,584–12,105 to 10,792–11,182 cal yr BP; Early Holocene). Arboreal and nonarboreal pollen types returned to values similar to BN2 but this time the trees were represented by Pinus (up to 60%) and Betula (up to 20%). Nonarboreal elements did not show evident influx decreases and their decline in percentage was due to the influx increase of arboreal elements, which determined a significant increasing trend in PAR. Evergreen Quercus began to appear in a consistent fashion, and the first appearances of deciduous Quercus and Corylus were recorded at the end of the zone. Fungi did not show significant changes and algae declined again to values similar to BN2, in both percentage and influx, as well as in composition. Among animals remains, Cladocera showed relevant abundances for the first time. Charcoal significantly increased and peaked by the middle of the zone. Plants 2023, 12, x FOR PEER REVIEW 12 of 30 Plants 2023, 12, 3644 11 of 29 Figure 5. Percentage diagram (Bassa Nera, core PATAM12-A14) of pollen and pteridophyte spores showing the pollen zones (BN1 to BN7) established using the Figure 5. Percentage diagram (Bassa Nera, core PATAM12-A14) of pollen and pteridophyte spores showing the pollen zones (BN1 to BN7) established using the OSIC and broken stick methods [50]. Solid lines are x10 exaggeration. Arboreal pollen is subdivided into conifer forest (Cf), deciduous forest (Df) and sclerophyll OSIC and broken stick methods [50]. Solid lines are ×10 exaggeration. Arboreal pollen is subdivided into conifer forest (Cf), deciduous forest (Df) and sclerophyll forest (Sf) elements; nonarboreal pollen is subdivided into shrubs (Sb) and herbs (Hs). Σp, pollen sum. Lithological units U3 to U5 as in Figure 4. Boundary ages forest (Sf) elements; of pollen zones arenonarboreal expressed aspollen is subdivided 95% confidence into shrubs intervals (Sb) obtained and Bacon using herbs (Hs). Σp, pollen age-depth sum.(Figure modelling Lithological 4). units U3 to U5 as in Figure 4. Boundary ages of pollen zones are expressed as 95% confidence intervals obtained using Bacon age-depth modelling (Figure 4). Plants 2023, 12, 3644 12 of 29 Plants 2023, 12, x FOR PEER REVIEW 13 of 30 Figure 6. 6. Figure Percentage diagram Percentage diagram(Bassa (BassaNera, Nera,core corePATAM12-A14) PATAM12-A14)ofofnon-pollen non-pollenpalynomorphs palynomorphs(NPP) (NPP) using using the the pollen pollen zones zones defined defined in in Figure 5. Solid Figure 5. Solid lines are x10 lines are exaggeration. ×10 exaggeration. Lithological units U3 to U5 as in Figure 4. The most abundant NPP are depicted in Figures 8–10. Pollen zones and boundary ages as in Figure5. Lithological units U3 to U5 as in Figure 4. The most abundant NPP are depicted in Figure 8-10. Pollen zones and boundary ages as in Figure 5. Plants Plants 2023, 12, 2023, x FOR 12,PEER 3644 REVIEW 13 of 29 14 of 30 Figure 7. Summary Figure influx 7. Summary diagram influx (Bassa diagram Nera, (Bassa core Nera, PATAM12-A14) core PATAM12-A14)displaying the most displaying the mostabundant abundantpollen pollen types, types, thethe NPPNPP groups groups andand charcoal. charcoal. SolidSolid lines lines are are x10 exaggeration. Lithological units U3 to U5 as in Figure 4. Pollen zones and boundary ages as in Figure 5. PAR, pollen accumulation rates. ×10 exaggeration. Lithological units U3 to U5 as in Figure 4. Pollen zones and boundary ages as in Figure 5. PAR, pollen accumulation rates. Plants 2023, 12, x FOR PEER REVIEW 15 of 30 Plants 2023, 12, 3644 14 of 29 Figure 8. Algae remains from Bassa Nera (core PATAM12-A14). (1–3) Pediastrum; (4–6) Botryococcus; Figure 8. Algae remains (7–10) Zygnema; from Bassa (15–18) (11–14) Cosmarium; Nera (core PATAM12-A14). (1–3) Pediastrum; (4–6) Botryococcus; Euastrum. (7–10) Zygnema; (11–14) Cosmarium; (15–18) Euastrum. Plants 2023, 12, 3644 15 of 29 Plants 2023, 12, x FOR PEER REVIEW 16 of 30 Figure 9. Animal remains from Bassa Nera (core PATAM12-A14). (1–2) Chironomids (mandibles); Figure 9. Animal remains from Bassa Nera (core PATAM12-A14). (1,2) Chironomids (mandibles); (3) Type BN-I; (4) Type BN-IIa; (5) Type BN-IIb; (6) Acari (legs); (7-8) Cladocera (mandibles). (3) Type BN-I; (4) Type BN-IIa; (5) Type BN-IIb; (6) Acari (legs); (7,8) Cladocera (mandibles). Zone BN7 (842 to 789 cm; 10,792–11,182 to 9578–10,362 cal yr BP; Early Holocene). Zone characterized by the full dominance of arboreal pollen types (Pinus, Betula, Corylus and both Quercus), which reached values over 95%. Nonarboreal taxa were negligible, and only Poaceae attained values of ~1%. This was due to a meaningful influx increase in Plants 2023, 12, 3644 16 of 29 tree elements, which resulted in maximum PAR values, and a slight influx decrease in nonarboreal taxa. All algae types declined to negligible values except Cosmarium, which slightly increased. Animal remains did not show significant changes, and Pinus stomata attained significant values throughout the zone. Charcoal influx remained high but highly fluctuating, and large particles (>150 µm) showed maximum values. Individually, the first significant increases in some important elements, such as Be- tula (~900 cm; 11.7–12.2 cal kyr BP), deciduous Quercus (~850 cm; 11.0–11.2 cal kyr BP), Corylus (~840 cm; 10.7–11.2 cal kyr BP) and Ulmus (~820 cm; 10.4–11.0 cal kyr BP), are worth noting. Also important are the last significant appearances of relevant elements such Plants 2023, 12, x FOR Saxifraga asPEER REVIEW and Cichoridoideae (~940 cm; 12.2–12.6 cal kyr BP), Chenopodiaceae 17 of 30 (~910 cm; 12.0–12.3 cal kyr BP), and Artemisia and Poaceae (~890 cm; 11.5–12.1 cal kyr BP). Figure 10. Plant macroremains from Bassa Nera (core PATAM12-A14). (1) Characeae (oospore); (2) Figure 10. Plant macroremains from Bassa Nera (core PATAM12-A14). (1) Characeae (oospore); Fragment of epidermis of a pine leaf with stomata and an isolated stoma (upper right corner). (2) Fragment of epidermis of a pine leaf with stomata and an isolated stoma (upper right corner). 5. Discussion 5. Discussion This section is aimed at interpreting the above results in terms of vegetation dynam- This section ics is aimed and the at interpreting potential theofabove influence climaticresults drivers,innotably terms temperature of vegetationanddynamics moisture. The and the potentialaccount influence of climatic is subdivided drivers, into two notably subsections: thetemperature and moisture. first aims to identify The the vegetation types that could account is subdivided intohave two produced the pollen subsections: theassemblages first aims found, and the to identify second the aims to reconstruct vegetation types the temporal that could have produced thetrends of these pollen vegetationfound, assemblages types inand relation to paleoclimatic the second aims toshifts and fire. reconstruct the temporal trends of these vegetation types in relation to paleoclimatic shifts and fire. 5.1. Vegetation Types 5.1. Vegetation Types Modern pollen spectra around Bassa Nera are dominated by Pinus (~80%) with small percentages of grasses (<10%), Corylus and deciduous Quercus (<1%) (Figure 3). A similar Modern pollen spectrahas assemblage around not beenBassa Nera found are pollen in our dominated by Pinus sequence, which (~80%) indicateswith small that the LGEH percentages of grasses (<10%), vegetation Corylus and was significantly deciduous different from theQuercus present.(<1%) This is(Figure 3). A similar not surprising, as previous glaciological assemblage has not been found studiesindemonstrated that the sitewhich our pollen sequence, was near the snowline indicates that during the LGEH the Late Glacial [9]. However, no modern pollen analogs exist on the uppermost vegetation was significantly different from the present. This is not surprising, as previous sites close to the present snowline (~2800 m), which indicates that Late Glacial vegetation was also different glaciological studies demonstrated that the site was near the snowline during the Late from the present, even at the highest vegetation belts. Glacial [9]. However,Tonounravel modern pollen analogs exist on the uppermost sites close to the the LGEH vegetation types from Bassa Nera, a cluster analysis was per- present snowlineformed (~2800onm), pollen taxa,indicates which avoiding that scarceLate Glacial with components vegetation was also only sporadic differentFour occurrences. from the present,assemblages even at thewere highest vegetation identified, belts. two representing open (nonforested) associations (O1 and To unravel the O2) LGEH and twovegetation representingtypes forestsfrom (F1 andBassa Nera,11). F2) (Figure a cluster O1 was analysis dominatedwas per- by Artemisia formed on pollenand Poaceae, taxa, avoidingand O2 was dominated scarce components by Saxifraga with onlyandsporadic Cichorioideae. Forest associations occurrences. Four assemblages were showed patternstwo identified, similar to the present: representing open F1 (nonforested) was dominated associations by deciduous trees (O1 and(notably Betula and Corylus) and coincided with Df, whereas F2 had a single component, Pinus, O2) and two representing forests (F1 and F2) (Figure 11). O1 was dominated by Artemisia thus coinciding with Cf. The main difference with extant vegetation types was the Late and Poaceae, andGlacial O2 was dominated abundance of O1byandSaxifraga O2 taxa, and Cichorioideae. especially Forest and Artemisia, Saxifraga associations Cichorioideae showed patterns(butsimilar to the present: F1 was dominated by deciduous also Chenopodiaceae, other Asteraceae, Caryophyllaceae, Ephedra, Apiaceaetrees (notablyand Betula and Corylus) and coincided Plantago), with Df, whereas which is unparalleled in modernF2 pollen had a assemblages single component, Pinus, and deserves thuscon- further sideration. Plants 2023, 12, 3644 17 of 29 coinciding with Cf. The main difference with extant vegetation types was the Late Glacial abundance of O1 and O2 taxa, especially Artemisia, Saxifraga and Cichorioideae (but also Plants 2023, 12, x FOR PEER REVIEW 18 of 30 Chenopodiaceae, other Asteraceae, Caryophyllaceae, Ephedra, Apiaceae and Plantago), which is unparalleled in modern pollen assemblages and deserves further consideration. Figure 11. Cluster analysis using pollen taxa above 0.5% of the total. Dominant taxa of each associ- Figure 11. Cluster analysis using pollen taxa above 0.5% of the total. Dominant taxa of each ation are in bold. association are in bold. In the Mediterranean region, the indicator character of Artemisia pollen in past rec- In the Mediterranean region, the indicator character of Artemisia pollen in past records ords has long been debated, which is linked to the difficulty of identifying the species has long involved been using debated, only pollen which morphology.is linked Theto the difficulty pollen of this genusof identifying the species has traditionally been involved using used only pollenglacial to characterize morphology. and other The pollen cold of this climatic genus phases, buthas traditionally according been to Subally used to char- and acterize Quétzel [55],glacial and other cold the approximately climatic 30 extant phases, species but according of Artemisia living in to theSubally and Quétzel [55], Mediterranean the approximately region bear a wide range 30ofextant environmental Artemisiaand species ofconditions, living in theglacial a strictly Mediterranean character can-region bear a not be guaranteed, wide which is supported range of environmental by increases conditions, and in this pollen a strictly during glacial interglacial character and be guaran- cannot other teed,warming which isphases. In addition, supported the concurrence by increases in this pollenof Artemisia, Chenopodiaceae during interglacial and warming and other Plantago phases. pollen has beenthe In addition, considered concurrencestrongof evidence for regional Artemisia, grazing inand Chenopodiaceae the Pyrenees, Plantago pollen has whereas Cichorioideae been considered has been strong linked for evidence to local grazing regional activities grazing in the[56].Pyrenees, In our case, how- Cichori- whereas ever, the absence of archeological and paleoecological evidence for human impact during oideae has been linked to local grazing activities [56]. In our case, however, the absence of the period of study, along with the lack of fungal spores from coprophilous fungi as indi- archeological and paleoecological evidence for human impact during the period of study, cators of grazing activities (Sporormiella, Sordiaria, Podospora), strongly argues against this along with the lack of fungal spores from coprophilous fungi as indicators of grazing possibility in the Bassa Nera Late Glacial. activities (Sporormiella, In the Pyrenees, Sordiaria, Late Glacial Podospora), pollen strongly assemblages arguesbyagainst dominated Artemisia,thisPoaceae possibility in the Bassa Nera Late Glacial. and Chenopodiaceae have traditionally been interpreted in terms of cold and arid steppe- In the Pyrenees, like vegetation on the basis Late Glacial pollen of similarities assemblages with modern pollendominated spectra fromby Artemisia, this vegetation Poaceae and type in eastern Europe [10,15]. This interpretation was further extended to the adjacentsteppe-like Chenopodiaceae have traditionally been interpreted in terms of cold and arid vegetation Pyrenean on the lowlands basis [57] and theof similarities whole Iberian with modern Peninsula pollen [17,19], spectraoffrom regardless this vegetation elevation. Intype someincases, however, eastern Europesteppe vegetation [10,15]. Thisisinterpretation not mentioned,was and the association further extendedArtemisia- to the adjacent Chenopodiaceae Pyrenean lowlands (along[57] withand several Caryophyllaceae the whole and Cichorioideae, Iberian Peninsula among others) [17,19], regardless of elevation. In is some considered cases, tohowever, be characteristic steppeofvegetation pioneering vegetation on open mineral is not mentioned, and the soils, whereas Artemisia- association Poaceae is considered(along Chenopodiaceae to be dominant with several in closed herb vegetationand Caryophyllaceae [58].Cichorioideae, Notably, these asso- among others) ciations are similar to our O1 and O2 assemblages, but in our case, no modern analogs for is considered to be characteristic of pioneering vegetation on open mineral soils, whereas these pollen assemblages have been found (Figure 3). O1 differs from alpine meadow as- Poaceae is considered to be dominant in closed herb vegetation [58]. Notably, these semblages by the abundance of Artemisia and the presence of Juniperus, whereas O2 is associations lacking at all. are similar to our O1 and O2 assemblages, but in our case, no modern analogs forAssemblage these pollen O1assemblages could represent have been found herbaceous (Figurewith vegetation 3). O1 differsspecies Artemisia from alpine char- meadow assemblages acteristic by theclimates. of periglacial abundance of Artemisia Presently, and the two species presence of this genus, of A. Juniperus, eriantha andwhereas A. O2 is lackingoccur gabriellae, at all.at the highest elevations of the Pyrenean alpine belt; however, they are Assemblage restricted O1 could to rocky habitats and do represent not occurherbaceous vegetation on alpine meadows [59].with Artemisia The shrubby species char- Juni- perus is represented acteristic in the alpine of periglacial and subalpine climates. Presently,stages two byspecies J. communis subsp. of this nana and genus, J. A. eriantha and sabina, respectively, A. gabriellae, occurbothat typical of relatively the highest warmof elevations andthehumid Pyrenean environments. alpine belt; It seems however, they unlikely that the same are restricted Artemisia to rocky and Juniperus habitats and do species not occurwereon present alpineinmeadows extremely cold [59].Late The shrubby Plants 2023, 12, 3644 18 of 29 Juniperus is represented in the alpine and subalpine stages by J. communis subsp. nana and J. sabina, respectively, both typical of relatively warm and humid environments. It seems unlikely that the same Artemisia and Juniperus species were present in extremely cold Late Glacial conditions. In addition to Poaceae, Plantago, represented by three species (P. alpina, P. media and P. monosperma), is the only component of this assemblage commonly growing on alpine meadows. Apiaceae pollen could not be identified, but it is mostly of the Eryngium type, which is represented in the high mountain meadows by E. alpinum, E. bourgatii and E. campestre. In summary, this pollen assemblage might be representative of uppermost alpine meadows with scattered shrubs with no modern analogs, probably of steppic nature due to the presence and abundance of unknown Artemisia and Juniperus species. With the available evidence, however, it is not possible to know whether this vegetation type is similar to the modern eastern European steppes because of the lack of sufficient taxonomic resolution (genera or families, at most). Therefore, this pollen assem- blage will be considered to represent periglacial herbaceous vegetation with no further biogeographical considerations. Assemblage O2 is even more difficult to interpret. One of the dominant taxa, Saxifraga, is represented in the Pyrenean highlands by numerous species, many of them living in rocky environments close to the snowline, situated at ~2800 m elevation [59]. The codominant Cichorioideae could correspond to several genera with different ecological requirements, which is also true for other Asteraceae and Caryophyllaceae. The Chenopodiaceae are represented mainly by Chenopodium bonus-henricus, a nitrophilous megaphorb growing on forest clearings and ruderal sites. Ephedra is mostly represented by the E. distachya type, which is absent from the Pyrenean highlands and occurs mainly on southern dry lowland steppes. The lack of sufficient taxonomic resolution and the fact that a similar association has not been documented in extant plant communities or pollen assemblages [32] makes O2 difficult to interpret in terms of vegetation type. A number of O2 elements (notably Cichorioideae, Caryophyllaceae, Chenopodiaceae and Ephedra) have been considered in the Pyrenean and Iberian Late Glacial literature as characteristic of cold/dry Eurasian steppes and interpreted in combination with the O1 elements. However, the clear separation of O1 and O2 assemblages in our Bassa Nera record could have some paleoecological significance that remains unknown. Therefore, this pollen assemblage is considered here to represent an unknown high mountain plant association with no modern analogs. 5.2. Climate Changes, Vegetation Shifts and Fire The relationship between the identified pollen assemblages and the independent pale- oclimatic records is analyzed here using the abovementioned Ostolo and Seso speleothem records for temperature and moisture, respectively. The main departures of the Ostolo record with respect to the NGRIP2 standard were a small offset at the end of GS-2.1a (i.e., the Bølling warming commenced a couple of centuries before in OST2) and the ab- sence of minor GI-1d, GI-1c2 and GI-1b coolings, along with the occurrence of a cooling event during the GI-1c1 Greenland warming (Figure 1). The remaining trends are in good agreement with the ice core records, including the GS-1 (YD) and the 11.4 cold event. Regarding moisture, the GS-1/YD showed a conspicuous trend toward wet climates in the Seso record (Figure 1), coinciding with another speleothem record from the French Moulis cave (Figure 1) reporting humid conditions during the same period [48]. The oc- currence of wet climates on Bassa Nera during the GS-1/YD is supported by the increase in freshwater algae influx, a pollen-independent proxy for water volume/level, which remained at lower values before and after this phase (Figure 7). The significant increase in sediment accumulation rates which occurred during the GS-1/YD (YDsa: Figure 4) may also suggest the occurrence of local humid conditions. Given the good agreement between pollen zones and the INTIMATE event stratigraphy (Figure 5), comparisons of the Bassa Nera palynological record with the available paleoenvironmental proxies will be conducted using the corresponding paleoclimatic phases (Figure 12). Plants 2023, 12, 3644 19 of 29 Plants 2023, 12, x FOR PEER REVIEW 21 of 32 Figure 12. Comparison of pollen assemblages defined by cluster analysis (Figure 11) with proxies Figure 12. Comparison for temperature, of pollenand moisture availability assemblages fire (Figure 1defined by 7). and Figure cluster Pollenanalysis (Figure assemblages are 11) with proxies for presented in percentage and influx units (grains cm −2 y−1), sorted chronologically according to their temperature, moisture availability and fire (Figures 1 and 7). Pollen assemblages are presented in respective maxima. BN, pollen zones; GA, glacier−advances in the adjacent Ruda valley (Figure 1); percentage and influx units (grains cm 2 y−1 ), sorted chronologically according to their respective PAR, pollen accumulation rates; SAR, sediment accumulation rates. maxima. BN, pollen zones; GA, glacier advances in the adjacent Ruda valley (Figure 1); PAR, pollen 5.2.1. GS-2 (Oldest accumulation Dryas) rates; SAR, sediment accumulation rates. Initial deglaciation of the Pla dera Muntanheta under still cold, although no longer 5.2.1. glacial,GS-2 (Oldest climates Dryas) and incipient sedimentation in a proglacial (seasonal?) pond in front of the receding moraines. Periglacial landscapes with dominant permafrost conditions with a poorInitial plantdeglaciation cover of smallof scattered the Pla dera grassyMuntanheta under Artemisia stands (O1) still cold, although no longer on suitable glacial, microenvironments. Pine forests were downslope, at elevations significantly lower thanpond in front of the climates and incipient sedimentation in a proglacial (seasonal?) today. Indeed, receding the tree line moraines. was aboutlandscapes Periglacial 1500 m below the present, with dominantas suggested by Pinus permafrost conditions with a poor plant cover of small scattered grassy Artemisia stands (O1) on suitable microenvironments. Pine forests were downslope, at elevations significantly lower than today. Indeed, the tree line was about 1500 m below the present, as suggested by Pinus pollen percentages (compare with Figure 3), which would be compatible with average temperatures of the order of 6 ◦ C or more below the present. By that time, deciduous forests were probably too far, probably in the basal lowlands, for their pollen to be wind-transported to high- mountain environments and the O2 vegetation had not yet colonized the site. 5.2.2. GI-1 (Bølling/Allerød) The abrupt global warming recorded by 14.7 cal kyr BP (~15 cal kyr BP in the west- ern Pyrenees) did not result in a significant landscape change on Bassa Nera, where O1 Plants 2023, 12, 3644 20 of 29 vegetation was declining and the downslope pine forests were rising, but only in relative abundance. The main vegetation shift took place about 0.5–1 cal kyr after temperature stabilization, in the middle of the interstadial (~14 cal kyr BP, the boundary between pollen zones BN1 and BN2), when pine forests abruptly increased, although PAR values were still lower to support the presence of conifer forests on the coring site. Actually, pine forest percentages indicated a tree line situated approximately 800–1000 m below the present, which suggests a ~500 upward shift with respect to the GS-2/Oldest Dryas stadial. After ~14 cal kyr BP, temperatures started to gradually decrease, which was paralleled by a similar decline of conifer forests and rising O1 and O2 trends (Pollen Zone BN2) compatible with a local increase in plant cover on Bassa Nera. Notably, the two glacier readvances recorded in the adjacent Ruda valley at 13.5 and 13.0 cal kyr BP [9] occurred during this phase and approximately coincided with two significant accelerations of sediment accumulation rates (SAR), which would suggest enhanced glacial erosion (winter) and increased sediment transport of the eroded materials (summer) to the pond. 5.2.3. GS-1 (Younger Dryas) The onset of this stadial was marked by a distinct temperature decline and a major local runoff/moisture increase, as suggested by abrupt increments in SAR and algae influx. Temperatures remained low during the whole stadial but moisture experienced some short-lived (centennial) though significant changes. Indeed, algae decreased again by ~12.6 cal kyr BP and peaked two more times by ~12.3 and ~11.9 cal kyr BP, which would indicate fluctuating water volumes/levels in the pond within a generally wetter climate. SAR showed a different pattern consisting of a continuous decline throughout the GS-1/YD, which would suggest a decreasing importance of winter erosion and/or summer melting, likely due to increased ice accumulation. This, together the rising moisture trend in the Seso record [8] and the occurrence of humid conditions, as deduced from enhanced speleothem growth, in the Moulis cave [48], in the southern and northern lowlands, respectively, strongly suggest cold and wet regional climates, although with eventual short-term local oscillations due to particular site features, during this stadial. Moisture oscillations were paralleled by major vegetation changes. The first local moisture increase (~12.7 cal kyr BP) coincided with an outstanding peak in O1 (Pollen Zone BN3) followed by a sudden decline coeval with a rapid increment of O2 (Pollen Zone BN4), coinciding with an equally rapid and intense moisture decline. This suggests that O1 would be favored by wet climates, whereas O2 would be more linked to dry conditions, at least under maintained cold climates. Subsequent moisture peaks had similar consequences, though less intense, for O1 (Pollen Zone BN5), but O2 never recovered after its BN4/BN5 decline. Therefore, the hydroclimate would be more influential than temperature in deter- mining local vegetation types and cover. Indeed, temperature remained relatively stable during the whole stadial and affected mainly the regional vegetation, especially conifer forests, which were lowered to elevations similar to the GS-2 (OD) ones. The hydroclimate could also have had some influence on deciduous forests, which began to increase during the last moisture peak (Pollen Zone BN5) that occurred by 11.9 cal kyr BP, approximately two centuries before the abrupt warming that initiated the Holocene, when climates were regionally wet, as indicated by the Seso and Moulis speleothem records. The occurrence of plant communities dominated by Artemisia and Poaceae (O1) under wet conditions contrasts with traditional interpretations of these associations as indicative of dry steppes, on the basis of biogeographical inferences [10]. On the one hand, the already discussed lack of sufficient taxonomic resolution for Artemisia pollen records, along with the wide arrangement of environments where the species of this genus may live, hinders specific paleoenvironmental inferences such as those traditionally attempted [55]. On the other hand, it should be noted that, in periglacial environments, wet climates and water stress can coexist, as moisture availability may be reduced by permanent freezing (permafrost) rather than by decreases in the hydrological balance characteristic of arid climates. Indeed, the vegetation of these environments, also known as periglacial deserts, Plants 2023, 12, 3644 21 of 29 occurs on highland tundra-like wet biomes [60,61]. Care must be taken when assigning well-defined vegetational terms to unknown past vegetation types. In the case of the Pyrenees, the use of the term “steppe” for the Glacial and Late Glacial open vegetation may implicitly suggest the existence of specific climatic, ecological and biogeographical conditions that are not necessarily supported by the available paleoclimatic evidence. In the long run, these terms and the associated environmental features may be consolidated by repetition and become established as paradigms in the scientific literature. In this way, unwarranted paleoenvironmental reconstructions may become axiomatic and foster circular interpretations when analyzing the response of vegetation to LGEH climatic shifts. 5.2.4. Early Holocene The initial Holocene warming was characterized by the drastic reduction, and eventu- ally the disappearance, of O1 and O2 vegetation types and the gradual increase of conifer and deciduous forests, which were likely ascending due to the warming. The situation of conifer forests was similar to the GI-2/BA interstadial, when the tree line was 800–1000 m below the present. The well-known high dispersion power of pine pollen prevents us from knowing whether conifer forests dominated by this tree grew on the coring site during the Early Holocene. Indeed, no significant differences were found in pine pollen percentages at elevations with conifer forests and above the tree line (Figure 3). In this case, the presence of Pinus stomata can provide support, as they are deposited mostly locally [46]. These stomata were absent until pollen zone BN7, except for a couple of single occurrences in BN4/BN5 (Figure 6), which suggests that conifer forests were not present in the coring site until the Early Holocene, from ~11 cal yr BP onward. High PAR values also support this interpretation. The case of deciduous forests was very different, as they were absent in the GI-2/BA stadial but very well developed in the Early Holocene. Actually, the pollen signal of these Early Holocene deciduous forests was similar to today’s montane stage (Figure 3), situated 500–1000 m below, under average temperatures 3–6 ◦ C higher than in Bassa Nera. Although such a general elevational shift is hard to accept, it is difficult to escape the idea that decidu- ous forests were closer to Bassa Nera than they are today. A possible explanation is that these forests expanded from nearby small refugia with suitable microclimatic conditions—i.e., microrefugia [62]—as formerly proposed for the northern Pyrenean slope [10,11]. A number of these microrefugia for forest trees within Pleistocene glacial and periglacial environments have been documented in SW France using edaphic charcoal analysis [63]. In the case of evergreen Quercus, this explanation is unlikely, as its climatic habitat lies ~2000 m below the snowline. However, the upward dispersion of its pollen is noteworthy, as demonstrated by the modern-analog study in the Aiguamòg valley, where evergreen oaks are absent but their pollen occurs along the whole transect up to 2600 m elevation [32]. It should be clarified that the occurrence of evergreen Quercus within the Df assemblage is not anomalous, as mixed deciduous and evergreen oak forests are common in the montane stage, which is reflected in modern pollen assemblages [32]. 5.2.5. Fire Incidence Fires began to be important in the Early Holocene, as indicated by the charcoal record. Interestingly, fire incidence (charcoal) and total vegetation cover (PAR) showed a statistically significant correlation (Table 2) for the whole Bassa Nera record, which suggests that fuel availability (in the form of biomass accumulation), along with higher temperatures, was a major driver for fire, as previously noted in other localities from the northern and southern Pyrenean sides [14,20]. It should be noted that the main bulk of charcoal particles found in this study were below the boundary for regional vs. local fire events (150 µm), and, therefore, they record mostly regional fires, likely by burning of lower elevation forests. Charcoal particles larger than 150 µm were comparatively scarce and only increased in zone BN7, coinciding with the appearance of Pinus stomata (Figure 6), which suggests that the burning of in situ pine forests did not occur before those dates. Only forests, especially Plants 2023, 12, 3644 22 of 29 deciduous forests, showed significant correlations with charcoal (Table 2), which suggests that they provided the main fuel for regional forest fires. Table 2. Pearson correlation coefficients between the pollen assemblages defined by cluster analysis (Figure 11) and charcoal using influx units (grains/particles cm−2 y−1 ). Significant correlations in bold. Charcoal p (uncorr) PAR 0.6249 9.5 × 10−7 Conifer forest 0.4845 0.0003 Deciduous forest 0.6999 1.1 × 10−8 O1 0.1453 0.3088 O2 0.0162 0.9103 6. Regional Comparisons Pyrenean-wide comparisons among pollen assemblages were carried out for each LGEH stadial/interstadial considering general climatic (wet Atlantic vs. dry Mediter- ranean), biogeographical (Eurosiberian vs. Mediterranean) and elevational (highlands vs. lowland) gradients (Figure 13). Independent paleoclimatic reconstructions are scarce in the surveyed localities, and most paleoenvironmental inferences are based on the same palynological proxies used for vegetation reconstruction, which hinders detailed compar- isons of vegetation responses to past climatic changes. To homogenize nomenclature, we used the classical terminology (Oldest Dryas, Bølling/Allerød, Younger Dryas), as the updated terminology (GS-2, GI-1, GS-1) was unavailable in former studies. Sites with Plants 2023, dating 12, x FORproblems and/or poorly defined LGEH phases (Appendix A) were not considered PEER REVIEW 24 of 3 in this comparison. Figure 13. Sketch map with the main components of pollen assemblages from the LGEH stadials/in Figure 13. Sketch map with the main components of pollen assemblages from the LGEH stadi- terstadials in the different Pyrenean and lowland sites analyzed (see Figure 2 and Appendix A fo als/interstadials in the different Pyrenean details). The Bassa Nera and lowland record sites by is encircled analyzed a white (see circleFigure 2 and for easier Appendix location. As in A Figure 2, the for details). The Bassa Nera record is encircled by a white circle for easier location. As in black line is the border between France (northern Pyrenean slope) and Spain (southernFigure 2, Pyrenean slope), and the black line is the border the white between line (northern France marks the boundary Pyreneanbetween the Eurosiberian slope) and Spain (southernand Mediterranean Pyrenean bioge ographical regions. Localities outside the shadow area are lowland records. slope), and the white line marks the boundary between the Eurosiberian and Mediterranean biogeo- Forest taxa are in black (Cf) and tones of green (Df and Sf) for better appreciation. graphical regions. Localities outside the shadow area are lowland records. Forest taxa are in black (Cf) and tones of green 6.1.(Df andDryas Oldest Sf) for better appreciation. During this stadial, our record (BN) shows an intermediate situation between the Atlantic (western) and Mediterranean (eastern) sides. Indeed, open BN landscapes were codominated by Poaceae and Artemisia, the first being more abundant in the western open formations and the second being more important in nonforested areas from the eastern sector. Forest communities did not show significant differences and were dominated by Plants 2023, 12, 3644 23 of 29 6.1. Oldest Dryas During this stadial, our record (BN) shows an intermediate situation between the Atlantic (western) and Mediterranean (eastern) sides. Indeed, open BN landscapes were codominated by Poaceae and Artemisia, the first being more abundant in the western open formations and the second being more important in nonforested areas from the eastern sec- tor. Forest communities did not show significant differences and were dominated by Pinus on both sides. In the lowlands, the situation was also intermediate but open landscapes predominated over forested formations, except for a northern locality (Br), where Pinus was dominant. Sites from the Mediterranean biome did not show significant differences with Eurosiberian localities and were more consistent with the east–west gradient. The general impression is of relatively open landscapes dominated by Poaceae and Artemisia— with Juniperus in the west and Chenopodiaceae in the east—intermingled with conifer forests everywhere. 6.2. Bølling/Allerød During this interstadial, BN was fully dominated by pine pollen, which strongly contrasts with eastern-western Pyrenean sides and with lowlands, which were covered by mosaic vegetation. In general, the former Poaceae-Artemisia dipole was less pronounced and the forest cover increased, but at that time Betula was as important as Pinus every- where, except in the southern and eastern lowlands, where Betula was not an important element of the vegetation. Also noteworthy is the increase in Juniperus in the northern and southern lowlands from the central and western sectors. In general, during the B/A, major differences were recorded between the northern and southern Pyrenean sides, rather than on the east–west, highland–lowland or Eurosiberian–Mediterranean gradients. 6.3. Younger Dryas During this stadial, open vegetation became more dominant, and the east–west gra- dient accentuated due to the reinforcement of the Poaceae-Artemisia dipole (which was even more pronounced than in the OD) and the newly emerged contrast between western and eastern forests due to the importance of Betula in the former and its absence in the latter. In this case, the BN record was intermediate regarding open formations but more similar to the eastern sector because of the absence of Betula as an important component. In addition, our record differed from all other Pyrenean localities in the presence of the Sax- ifraga-Cichorioideae assemblage. No evident differences between highlands and lowlands or between sites from the Eurosiberian and Mediterranean regions were observed. 6.4. Early Holocene A major regional vegetation shift took place at the onset of the Holocene due to the full dominance of forest formations, except for a southern locality (Es), where stadial-like open landscapes remained. This general forest increase was due to the influx of Betula, Corylus and Quercus rather than to an expansion of conifer forests. In this case, BN experienced the same general replacement of open vegetation by newly arrived forests, which showed similar patterns across the east–west, highland–lowland and Eurosiberian–Mediterranean gradients, except for a few western lowland sites where grasslands were still important. Some minor differences between western and eastern forests were the importance of Corylus in the former and Betula in the latter, with BN in an intermediate position. 7. Final Remarks and Future Prospects In addition to providing the first continuous Late Glacial-Early Holocene palynological record for the Iberian Pyrenees at centennial resolution, this study emphasizes the need for improving paleoclimatic interpretations, especially in reference to the use of pollen- independent proxies to avoid circularity. In particular, inferences about the occurrence of cold/dry climates based on the occurrence of Artemisia-Poaceae steppes during the Younger Dryas are not consistent with independent paleoclimatic evidence suggesting the Plants 2023, 12, 3644 24 of 29 occurrence of wet climates for this stadial. Unfortunately, the lack of pollen-independent paleoclimatic records is insufficient for a sound assessment at regional level. Emphasis should be placed on the development of regional paleoclimatic studies based on suitable physico-chemical and biological proxies before analyzing the response of vegetation to paleoenvironmental changes, rather than inferring paleoclimates from pollen reconstruc- tions, as traditionally attempted. This work has also suggested the potential occurrence of climatic oscillations during the GS-1/YD stadial that are worth to be studied in more detail at higher resolution (decadal or less). In addition, independent paleohydrological reconstructions for Late Glacial phases other than the GS-1/YD are lacking in the Pyrenees and should be encouraged for a more consistent regional paleoclimatic view. Finally, it would be interesting to extend regional paleoclimatic and paleoecological comparisons to the continental ambit for a better appraisal of the influence of climatic, biogeographical and anthropogenic (when present) factors on the composition and tem- poral trends of LGEH montane vegetation across Europe. The use of the European Pollen Database (https://epdweblog.org/, accessed on 20 March 2023) would be especially useful in this sense but should be complemented with more regional compilations similar to those existing for the Iberian Peninsula [64], which have been demonstrated to be much more exhaustive and useful than the available global databases for this type of reconstruction [65]. Author Contributions: Conceptualization, V.R. and T.V.-V.; methodology, V.R., A.B., M.Á.C. and M.B.; formal analysis, V.R., A.B., M.Á.C., M.B. and T.V.-V.; investigation, V.R., A.B., M.Á.C., M.B. and T.V.-V.; writing—original draft preparation, V.R.; supervision, V.R. and T.V.-V.; funding acquisition, V.R. and T.V.-V. All authors have read and agreed to the published version of the manuscript. Funding: Institute of Catalan Studies (IEC), project POLMONT-2012; Autonomous Organization of National Parks (OAPN), projects RECREO (OAPN-387/2011) and LACEN (OAPN-2450S); Man- agement of University and Research Grants Agency (AGAUR), projects GREB-2014 SGR 514 and GREB-2021 SGR 00315. Data Availability Statement: Raw data are available upon request to V.R. or T.V.-V. Acknowledgments: Fieldwork collaborators, Encarni Montoya, Nick Loughlin, Arantza Lara, Núria Cañellas-Boltà, Sandra Garcés-Pastor and Julià López-Vila; modern climatic data, Javier Sigró. Special thanks to Génesis Hernández for the high-quality sample processing, which greatly facilitated palynological work. The authors acknowledge the comments by four anonymous reviewers. Conflicts of Interest: The authors declare no competing interest. Appendix A Table A1. Details of the Late Glacial coring localities with published palynological records, sorted by elevation, including the major pollen taxa used to assemble Figure 13 listed in top-down dominance order. Palynological information has been taken directly from the pollen diagrams displayed in the cited references. Fire (charcoal): + analyzed, - not analyzed; NR, not recorded; ND, possibly present but not defined. Site Latitude N Longitude Elevation Oldest Dryas Bølling/Allerød Younger Dryas Early Holocene Fire References Pinus Pinus Pinus Marboré 42◦ 410 4400 2◦ 000 3000 E 2612 NR - [66] Corylus Corylus Corylus Pinus Pinus Gourg- Pinus Artemisia 42◦ 3800 02◦ 1300 E 2080 Artemisia Betula - [11] Nègre Artemisia Pinus Betula Quercus Artemisia Pinus Racou 42◦ 330 1500 0◦ 020 2400 E 2000 NR NR Chenopodiaceae Betula - [67] Poaceae Quercus Pinus Pinus Bassettes * 42◦ 390 3000 02◦ 0100 E 1900 NR NR - [11] Poaceae Corylus Plants 2023, 12, 3644 25 of 29 Table A1. Cont. Site Latitude N Longitude Elevation Oldest Dryas Bølling/Allerød Younger Dryas Early Holocene Fire References Pinus Pinus Pinus Artemisia Betula Bassa Nera 42◦ 380 1700 0◦ 550 2700 E 1890 Artemisia Pinus - This paper Poaceae Corylus Poaceae Cichorioideae Quercus Pinus Pinus Pinus Laurenti 42◦ 400 3000 02◦ 010 4600 E 1860 NR Artemisia Artemisia Betula - [11] Poaceae Poaceae Quercus Pinus Pinus Pinus Poaceae Poaceae Portalet 42◦ 4800 00◦ 2300 W 1802 NR Betula + [16,20] Artemisia Artemisia Corylus Juniperus Juniperus Pinus Pinus Pinus Pinus Artemisia Balcère 42◦ 350 2700 02◦ 030 1800 E 1764 Poaceae Betula - [11] Artemisia Poaceae Artemisia Quercus Betula Pinus Pinus Pinus Betula Pouretère 42◦ 480 4000 00◦ 120 1200 W 1720 NR Poaceae Poaceae - [13] Quercus Artemisia Artemisia Corylus Pinus Pinus Sagnes * 42◦ 350 2700 02◦ 030 1800 E 1670 NR NR - [11] Poaceae Betula Pinus Pinus Pinus Pinus Tramacastilla 42◦ 430 2700 00◦ 220 0700 W 1668 Poaceae Poaceae Poaceae Betula - [15] ** Artemisia Artemisia Artemisia Quercus Artemisia Pinus Pinus Pinus Pinus Borde 42◦ 3200 02◦ 0500 E 1660 Poaceae Poaceae Corylus - [11] Poaceae Betula Artemisia Quercus Chenopodiaceae Bious 42◦ 500 00◦ 270 W 1550 ND ND ND ND [68] Artemisia Poaceae Pinus Freychinède 42◦ 4800 01◦ 260 1100 E 1350 Poaceae Artemisia NR Betula - [69] Pinus Betula Quercus Pinus Pinus Pinus Betula Pinus Búbal ** 42◦ 400 5800 00◦ 190 1300 W 1115 Betula Betula Juniperus Betula - [15] Juniperus Juniperus Artemisia Corylus Poaceae Juniperus Bentué 42◦ 200 1100 00◦ 290 2300 W 1000 NR Poaceae NR NR - [18] Pinus Quercus Poaceae Poaceae Poaceae Poaceae Ech 43◦ 0200 00◦ 0500 W 710 Juniperus Pinus Pinus + [14] Pinus Artemisia Betula Betula Betula Steppe elem. Juniperus Pinus Pinus Poaceae Poaceae Estanya 42◦ 0200 00◦ 3200 E 670 Juniperus Juniperus - [4] Pinus Artemisia Poaceae Poaceae Juniperus Pinus Castet 43◦ 020 00◦ 220 W 490 ND ND ND ND - [68] Betula Pinus Pinus Artemisia Barbazan 43◦ 020 0200 00◦ 370 2800 E 480 Corylus - [70] Artemisisa Artemisisa Poaceae Quercus Poaceae Artemisia Quercus Lourdes 43◦ 0200 00◦ 0500 W 430 Betula NR - [12] Poaceae Corylus Juniperus Poaceae Poaceae Poaceae Parco 41◦ 540 4800 00◦ 560 3100 E 420 Chenopodiaceae Asteraceae Chenopodiaceae NR - [71] Asteraceae Pinus Asteraceae Poaceae Poaceae Poaceae Artemisia Betula Pinus Biscaye 43◦ 0200 00◦ 040 1500 E 410 Pinus - [12] Poaceae Juniperus Betula Betula Artemisia Quercus Poaceae Poaceae Quercus Monge 43◦ 0300 00◦ 010 5500 E 394 NR Pinus Pinus Corylus - [12] Betula Betula Pinus Plants 2023, 12, 3644 26 of 29 Table A1. Cont. Site Latitude N Longitude Elevation Oldest Dryas Bølling/Allerød Younger Dryas Early Holocene Fire References Pinus Pinus Poaceae Pinus Estarrès 43◦ 050 3600 00◦ 220 4500 W 356 Artemisia Poaceae - [10] Betula Poaceae Poaceae Betula Pinus Pinus Artemisia Pinus Artemisia Betula Banyoles 42◦ 080 0700 02◦ 450 2300 E 175 Poaceae - [56] Artemisia Poaceae Corylus Betula Quercus Juniperus * Localities with fragmentary Late Glacial palynological records very close to others with complete sequences. ** Dating issues. References 1. Carrión, J.S.; Fernández, S.; González-Sampériz, P.; Gil-Romera, G.; Badal, E.; Carrión-Marco, Y.; López-Merino, L.; López-Sáez, J.A.; Fierro, E.; Burjachs, F. Expected Trends and Surprises in the Lateglacial and Holocene Vegetation History of the Iberian Peninsula and Balearic Islands. Rev. Palaeobot. Palynol. 2020, 162, 458–475. [CrossRef] 2. Rull, V.; Vegas-Vilarrúbia, T.; Corella, J.P.; Valero-Garcés, B. Bronze Age to Medieval vegetation dynamics and landscape anthropization in the southern-central Pyrenees. Palaeogeogr. Palaeoclimatol. Palaeoecol. 2021, 571, 110392. [CrossRef] 3. Rull, V.; Vegas-Vilarrúbia, T.; Corella, J.P.; Trapote, M.C.; Montoya, E.; Valero-Garcés, B. A unique Pyrenean varved record provides a detailed reconstruction of Mediterranean vegetation and land-use dynamics over the last three millennia. Quat. Sci. Rev. 2021, 268, 107128. [CrossRef] 4. González-Sampériz, P.; Aranbarri, J.; Pérez-Sanz, A.; Gil-Romera, G.; Moreno, A.; Leunda, M.; Sevilla-Callejo, M.; Corella, J.P.; Morellón, M.; Oliva, B.; et al. Environmental and Climate Change in the Southern Central Pyrenees Since the Last Glacial Maximum: A View from the Lake Records. Catena 2017, 149, 668–688. [CrossRef] 5. Rull, V.; Vegas-Vilarrúbia, T. A spatiotemporal gradient in the anthropization of Pyerenan landscapes. Preliminary report. Quat. Sci. Rev. 2021, 258, 106909. [CrossRef] 6. Rasmussen, S.O.; Bigler, M.; Blockley, S.P.; Blunier, T.; Buchardt, S.L.; Clausen, H.B.; Cvijanovic, I.; Dahl-Jensen, D.; Johnsen, S.J.; Fischer, H.; et al. A stratigraphic framework for abrupt climatic changes during the Last Glacial period based on three synchronized Greenland ice-core records: Refining and extending the INTIMATE event stratigraphy. Quat. Sci. Rev. 2014, 106, 14–28. [CrossRef] 7. Bernal-Wormull, J.; Moreno, A.; Pérez-Mejías, C.; Bartolomé, M.; Aranburu, A.; Arriolabengoa, M.; Iriarte, E.; Cacho, I.; Spötl, C.; Edwards, R.; et al. Immediate temperature response in northern Iberia to last deglacial changes in the North Atlantic. Geology 2021, 49, 999–1003. [CrossRef] 8. Bartolomé, M.; Moreno, A.; Sancho, C.; Stoll, H.M.; Cacho, I.; Spötl, C.; Belmonte, Á.; Edwards, R.L.; Cheng, H.; Hellstrom, J.C. Hydrological changes in southern Europe responding to increasing North Atlantic overturning during Greenland Stadial 1. Proc. Natl Acad. Sci. USA 2015, 112, 6568–6572. [CrossRef] 9. Fernandes, M.; Oliva, M.; Vieira, G.; Palacios, D.; Fernández-Hernández, J.M.; Garcia-Oteyza, J.; Schimmelpfennig, I.; Team, A.; Antoniades, D. Glacial Oscillations during the Bølling-Allerød Interstadial-Younger Dryas Transition in the Ruda Valley, Central Pyrenees. J. Quat. Sci. 2022, 37, 42–58. [CrossRef] 10. Jalut, G.; Montserrat Martí, J.; Fontugne, M.; Delibrias, G.; Vilaplana, J.M.; Julia, R. Glacial to interglacial vegetation changes in the northern and southern Pyrénées: Deglaciation, vegetation cover and chronology. Quat. Sci. Rev. 1992, 11, 449–480. [CrossRef] 11. Reille, M.; Lowe, J.J. A Re-Evaluation of the Vegetation History of the Eastern Pyrenees (France) from the End of the Last Glacial to the Present. Quat. Sci. Rev. 1993, 12, 47–77. [CrossRef] 12. Reille, M.; Andrieu, V. The late Pleistocene and Holocene in the Lourdes Basin, Western Pyrénées, France: New pollen analytical and chronological data. Veg. Hist. Archaeobot. 1995, 4, 1–21. [CrossRef] 13. Aubert, S.; Belet, J.M.; Bouchette, A.; Otto, T.; Dedoubat, J.J.; Fontugne, M.; Jalut, G. Dynamique tardiglaciaire et holocène de la végétation à l’étage montagnard dans les Pyrénées centrales. Comptes Rendus Biol. 2004, 327, 381–388. [CrossRef] [PubMed] 14. Rius, D.; Galop, D.; Doyen, E.; Millet, L.; Vannière, B. Biomass burning response to high-amplitude climate and vegetation changes in Southwestern France from the Last Glacial to the early Holocene. Veget. Hist. Archaeobot. 2014, 23, 729–742. [CrossRef] 15. Montserrat Martí, J.M. Evolución Glaciar y Postglaciar del Clima y la Vegetación en la Vertiente sur del Pirineo: Estudio Palinológico; Monografías del Instituto Pirenaico de Ecología nº 6; Consejo Superior de Investigaciones Científicas: Zaragoza, Spain, 1992; pp. 1–147. 16. González-Sampériz, P.; Valero-Garcés, B.; Moreno, A.; Jalut, G.; García-Ruiz, J.M.; Martí-Bono, C.; Delgado-Huertas, A.; Navas, A.; Otto, T.; Dedoubat, J.J. Climate variability in the Spanis Pyrenees during the last 30,000 yr revealed by the El Portalet sequence. Quat. Res. 2006, 66, 38–52. [CrossRef] 17. García-Ruiz, J.M.; Palacios, D.; González-Sampériz, P.; de Andrés, N.; Moreno, A.; Valero-Garcés, B.; Gómez-Villar, A. Mountain glacier evolution in the Iberian Peninsula during the Younger Dryas. Quat. Sci. Rev. 2016, 138, 16–30. [CrossRef] Plants 2023, 12, 3644 27 of 29 18. González-Sampériz, P.; Valero-Garcés, B.; Carrión, J.S.; Peña-Monné, J.L.; García-Ruiz, J.M.; Martí-Bono, C. Glacial and lateglacial vegetation in northern Spain: New data and a review. Quat. Int. 2005, 140–141, 4–20. [CrossRef] 19. González-Sampériz, P.; Leroy, S.A.; Carrión, J.S.; Fernández, S.; García-Antón, M.; Gil-García, M.J.; Uzquiano, P.; Valero-Garcés, B.; Figueiral, I. Steppes, savannahs, forests and phytodiversity reservoirs during the Pleistocene in the Iberian Peninsula. Rev. Palaeobot. Palynol. 2010, 162, 427–457. [CrossRef] 20. Gil-Romera, G.; González-Sampériz, P.; Lasheras-Álvarez, L.; Sevilla-Callejo, M.; Moreno, A.; Valero-Garcés, B.; López-Merino, L.; Carrión, J.S.; Sanz, A.P.; Aranbarri, J.; et al. Biomass-modulated fire dynamics during the Last Glacial-Interglacial transition at the Central Pyrenees (Spain). Palaeogeogr. Palaeoclimatol. Palaeoecol. 2014, 402, 113–124. [CrossRef] 21. Theuerkauf, M.; Couwenberg, J. ROPES reveals past land cover and PPs from single pollen records. Front. Earth Sci. 2018, 6, 14. [CrossRef] 22. Cook, E.J.; van Geel, B.; van der Kaars, S.; van Arkel, J. A review of the use of non-pollen palynomorphs in palaeoecology with examples from Australia. Palynology 2011, 35, 155–178. [CrossRef] 23. Shumilovskikh, L.S.; van Geel, B. Non-pollen palynomorphs. In Hanbook for the Analysis of Micro-Particles in Archaeological Samples; Henry, A.G., Ed.; Springer Nature: Cham, Switzerland, 2020; pp. 65–94. 24. Garcés-Pastor, S.; Cañellas-Boltà, N.; Clavaguera, A.; Calero, M.A.; Vegas-Vilarrúbia, T. Vegetation shifts, human impact and peat bog development in Bassa Nera pond (Central Pyrenees) during the last millennium. Holocene 2016, 27, 553–565. [CrossRef] 25. Garcés-Pastor, S.; Cañellas-Boltà, N.; Pèlachs, A.; Soriano, J.-M.; Pérez-Obiol, R.; Pérez-Haase, A.; Calero, M.-A.; Andreu, O.; Escolà, N.; Vegas-Vilarrúbia, T. Environmental history and vegetation dynamics in response to climate variations and human pressure during the Holocene in Bassa Nera, Central Pyrenees. Palaeogeogr. Palaeoclimatol. Palaeoecol. 2017, 479, 48–60. [CrossRef] 26. López-Vila, J.; Montoya, E.; Cañellas-Boltà, N.; Rull, V. Modern non-pollen palynomorphs sedimentation along an elevational gradient in the south-central Pyrenees (southwestern Europe) as a tool for Holocene paleoecological reconstruction. Holocene 2014, 24, 327–345. [CrossRef] 27. López-Vila, J.; Montoya, E.; Rull, V. Non-pollen palynomorphs analysis of the last two millennia in the Bassa Nera peat bog record (South-Central Pyrenees, SW Europe). In Proceedings of the VI Workshop Non-Pollen Palynomoprhs, Tallin, Estonia, 16–20 June 2014. 28. Casas, J.M.; Álvaro, J.J.; Clausen, S.; Padel, M.; Puddu, C.; Sanz-López, J.; Sánchez-García, T.; Navidad, M.; Castiñeiras, P.; Liesa, M. Palaeozoic basement of the Pyrenees. In The Geology of Iberia: A Geodynamic Approach; Quesada, C., Oliveira, J.T., Eds.; Springer Nature: Cham, Switzerland, 2019; pp. 229–259. 29. Delgado, J.; Soler, A. Ilvaite stability in skarns from the northern contact of the Maladeta batholith, Central Pyrenees (Spain). Eur. J. Mineral. 2010, 22, 363–380. 30. López-Moreno, J.I.; Alonso-González, E.; Monserrat, O.; DEL Río, L.M.; Otero, J.; Lapazaran, J.; Luzi, G.; Dematteis, N.; Serreta, A.; Rico, I.; et al. Ground-based remote-sensing techniques for diagnosis of the current state and recent evolution of the Monte Perdido Glacier, Spanish Pyrenees. J. Glaciol. 2019, 65, 85–100. [CrossRef] 31. Rivas-Martínez, S.; Penas, A.; del Río, S.; Díaz, T.E.; Rivas-Sáenz, S. Bioclimatology of the Iberian Peninsula and the Balearic Islands. In The Vegetation of the Iberian Peninsula; Loidi, J., Ed.; Springer Nature: Cham, Switzerland, 2017; pp. 29–80. 32. Cañellas-Boltà, N.; Rull, V.; Vigo, J.; Mercadé, A. Modern pollen-vegetation relationships along an altitudinal transect in the central Pyerenees (southwestern Europe). Holocene 2009, 19, 1185–1200. [CrossRef] 33. Pérez-Haase, A.; Ninot, J.M. Hydrological heterogeneity rather than water chemistry explains the high plant diversity and uniqueness of a Pyrenean mixed mire. Folia Geobot. 2017, 52, 143–160. [CrossRef] 34. Gassiot Ballbè, E.; Mazzucco, N.; Clemente Conte, I.; Rodríguez Antón, D.; Obea Gómez, L.; Quesada Carrasco, M.; Díaz Bonilla, S. The beginning of high mountain occupations in the Pyrenees. Human settlements and mobility from 18,000 cal BC to 2000 cal BC. In High Mountain Conservation in a Changing World; Catalan, J., Ninot, J.M., Aniz, M., Eds.; Springer Nature: Cham, Switzerland, 2017; pp. 75–105. 35. Vandergoes, M.J. AMS dating of pollen concentrates—A methodlological study of Late Quaternary sediments from South Westland, New Zealand. Radiocarbon 2003, 45, 479–491. [CrossRef] 36. Reimer, P.J.; Austin, W.E.N.; Bard, E.; Bayliss, A.; Blackwell, P.G.; Ramsey, C.B.; Butzin, M.; Cheng, H.; Edwards, R.L.; Friedrich, M.; et al. The IntCal20 Northern Hemisphere radiocarbon age calibration curve (0–55 cal kBP). Radiocarbon 2020, 62, 725–757. [CrossRef] 37. Blaauw, M.; Christen, J.A. Flexible paleoclimate age-depth models using an autoregressive gamma process. Bayesian Anal. 2011, 6, 457–474. [CrossRef] 38. Bennett, K.D.; Willis, K.D. Pollen. In Tracking Environmental Change Using Lake Sediments. Volume 3: Terrestrial, Algal and Siliceous Indicators; Smol, J.P., Birks, H.J.B., Last, W.M., Eds.; Kluwer: New York, NY, USA, 2001; pp. 5–32. 39. Stockmarr, J. Tablets with spores used in absolute pollen analysis. Pollen Spores 1971, 13, 615–621. 40. Moore, P.D.; Webb, J.A.; Collinson, M.E. Pollen Analysis; Blackwell: Oxford, UK, 1991. 41. Reille, M. Pollen et Spores d’Europe et d’Afrique du Nord; URA-CNRS: Marseille, France, 1992. 42. Baker, A.G.; Bhagwat, S.A.; Willis, K.J. Do dung fungal spores make a good proxy for past distribution of large herbivores? Quat. Sci. Rev. 2013, 62, 21–31. [CrossRef] Plants 2023, 12, 3644 28 of 29 43. Cugny, C.; Mazier, F.; Galop, D. Modern and fossil non-pollen palynomorphs from the Basque mountains (western Pyrenees, France): The use of coprophilous fungi to reconstruct pastoral activity. Veget. Hist. Archaeobot. 2010, 19, 391–408. [CrossRef] 44. Gauthier, E.; Jouffroy-Bapicot, I. Detecting human impacts: Non-pollen palynomorphs as proxies for human impact on the environment. In Applications of Non-Pollen Palynomorphs: From Palaeoenvironmental Reconstruction to Biostratigraphy; Marret, F., O’Keefe, J., Osterloff, P., Pound, M., Shumilovkikh, L., Eds.; Geological Society of London: London, UK, 2021; pp. 233–244. 45. Lee, C.M.; van Geel, B.; Gosling, W.D. On the use of spores of coprohilous fungi preserved in sediments to indicate past herbivore presence. Quaternary 2022, 5, 30. [CrossRef] 46. García Álvarez, S.; Morla Juaristi, C.; Solana Gutiérrez, J.; García-Amorena, I. Taxonomic differences between Pinus sylvestris and P. uncinata revealed in the stomata and cutilce characters for use in the study of fossil material. Rev. Palaeobot. Palynol. 2009, 155, 61–68. [CrossRef] 47. Rull, V. A note on pollen counting in palaeoecology. Pollen Spores 1987, 29, 471–480. 48. Perrin, C.; Honiat, C.; Prestimonaco, L. Deciphering paleohydrology from the petrography and microstratigraphy of stalagmites (MIS 5–Holocene, Moulis Cave, Ariège, S. France). Sediment. Geol. 2022, 440, 106256. [CrossRef] 49. Whitlock, C.; Larsen, C. Charcoal as a fire proxy. In Tracking Environmental Change Using Lake Sediments. Volume 3: Terrestrial, Algal and Siliceous Indicators; Smol, J.P., Birks, H.J.B., Last, W.L., Eds.; Kluwer: New York, NY, USA, 2001; pp. 75–97. 50. Bennett, K.D. Determination of the number of zones in a biostratigraphical sequence. New Phytol. 1996, 132, 155–170. [CrossRef] 51. Sokal, R.R.; Michener, C.D. A statistical method for evaluating systematic relationships. Univ. Kansas Sci. Bull. 1958, 38, 1409–1438. 52. Bennett, K.D. Psimpoll—A quickbasic program that generates postcript page description of pollen diagrams. INQUA Newsl. 1992, 8, 11–12. 53. Kovack, W. MVSP—A Multivariate Statistical Package for Windows, version 3.1; Kovach Computing: Pentraeth, UK, 1999. 54. Hammer, Ø.; Harper, D.A.T.; Ryan, P.D. Past: Paleontological statistics sofware package for education and data analysis. Palaeontol. Electr. 2001, 4, 4. 55. Subally, D.; Quézel, P. Glacial or interglacial, a plant indicator with dual responses. Rev. Palaeobot. Palynol. 2002, 120, 123–130. [CrossRef] 56. Mazier, F.; Galop, D.; Brun, C.; Buttler, A. Modern pollen assemblages from grazed vegetation in western Pyrenees, France: A numerical tool for more precise reconstruction of past cultural landscapes. Holocene 2006, 16, 91–103. [CrossRef] 57. Pérez-Obiol, R.; Julià, R. Climatic change on the Iberian Peninsula recorded in a 30,000-yr pollen record from Lake Banyoles. Quat. Res. 1994, 41, 91–98. [CrossRef] 58. van der Knaap, W.O.; van Leeuwen, J.F.N. Late Glacial and Early Holocene vegetation succession, altitudinal vegetation succession, and climatic change in the Serra da Estrela, Portugal. Rev. Palaeobot. Palynol. 1997, 97, 239–285. [CrossRef] 59. Vigo, J. L’Alta Muntanya Catalana: Flora i Vegetació; Institut d’Estudis Catalans: Barcelona, Spain, 2009. 60. Bertran, P.; Sitzia, L.; Banks, W.E.; Bateman, M.D.; Demars, P.-Y.; Hernandez, M.; Lenoir, M.; Mercier, N.; Prodeo, F. The Landes de Gascogne (southwest France): Periglacial desert and cultural frontier during the Paleolithic. J. Archaeol. Sci. 2013, 40, 2274–2285. [CrossRef] 61. Ponel, P.; Parchoux, F.; Andrieu-Ponel, V.; Juhasz, V.; de Beaulieu, J.-L. A late-glacial—Holocene fossil insect succession from Vallée des Merveilles, French Alps, and its paleoecological implications. Arct. Antarct. Alpine Res. 2001, 33, 481–484. [CrossRef] 62. Rull, V. Microrefugia. J. Biogeogr. 2009, 36, 481–484. [CrossRef] 63. de Lafontaine, G.; Amasifuen, C.A.; Ducousso, A.; Petit, R.J. Cryptic no more: Soil macrofossils uncover Pleistocene forest microrefugia within a periglacial desert. New Phytol. 2014, 204, 715–729. [CrossRef] 64. Carrión, J.S.; Ochando, J.; Munuera, M.; Casas-Gallego, M.; González-Sampériz, P.; López-Sáez, J.A.; Postigo-Mijarra, J.M.; Barrón, E.; Badal, E.; Carrión-Marco, Y.; et al. Paleoflora y Paleovegetación Ibérica II: Pleistoceno; Ministerio de Ciencia e Inno- vación/Fundación Séneca: Murcia, Spain, 2022. 65. Rull, V.; Burjachs, F.; Carrión, J.S.; Ejarque, A.; Fernández, S.; López-Sáez, J.A.; Luelmo-Lautenschlaeger, R.; Ochando, J.; Pérez- Díaz, S.; Revelles, J.; et al. Historical biogeography of Cannabis in the Iberian Peninsula: A probabilistic aproach based on palynological evidence. Persp. Plant Ecol. Evol. Syst. 2003, 58, 125704. [CrossRef] 66. Leunda, M.; González-Sampériz, P.; Gil-Romera, G.; Aranbarri, J.; Moreno, A.; Oliva-Urcia, B.; Sevilla-Callejo, M.; Valero-Garcés, B. The Late-Glacial and Holocene Marboré Lake sequence (2612 m a.s.l., Central Pyrenees, Spain): Testing high-altitude sites sensitivity to millennial scale vegetation and climate variability. Glob. Planet. Chang. 2017, 157, 214–231. [CrossRef] 67. Guiter, F.; Andrieu-Ponel, V.; Digerfeldt, G.; Reille, M.; de Beaulieu, J.L.; Ponel, P. Vegetation history and lake-level changes from the Younger Dryas to the present in Eastern Pyrenees (France): Pollen, plant macrofossils and lithostratigraphy from Lake Racou (2000 m a.s.l.). Veget. Hist. Archaeobot. 2005, 14, 99–118. [CrossRef] 68. Jalut, G.; Andrieu, V.; Delibrias, G.; Fortugne, M.; Pagès, P. Palaeoenvironment of the valley of Ossau (western french Pyrénées) during the last 27000 years. Pollen Spores 1988, 30, 357–394. 69. Jalut, G.; Delibrias, G.; Dagnac, J.; Mardonès, M.; Bouhours, M. A palaeoecological approach to the last 21,000 years in the Pyrénées: The peat bog of Freychinède (alt. 1350 m, Ariège, South France). Palaeogeogr. Palaeoclimatol. Palaeoecol. 1982, 40, 321–359. [CrossRef] Plants 2023, 12, 3644 29 of 29 70. Jalut, G.; Turu, V. Le dernier cycle glaciaire-interglaciaire dans les Pyrenees: Englacement, climat, vegetation. In Pyrenees d’Hier et d’Aujourd’hui; Canérot, J., Coli, J.-P., Platel, J.-P., Bilotte, M., Eds.; Altantica: Biarritz, France, 2008; pp. 145–161. 71. Bergadà, M.; Burjachs, F.; Fullola, J.M. Évolution paléoenvironnemental du 14.500 au 10.000 BP dans le Prépyrénées catalans: La Grotte du Parco (Alòs de Balaguer, Lleida, Espagne). L’Anthropologie 1999, 103, 249–264. Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.