AMQ 22 Gruger esimpaginazione def prova 1 Available online http://amq.aiqua.it ISSN (print): 2279-7327, ISSN (online): 2279-7335 Alpine and Mediterranean Quaternary, 26 (2), 2013, 101-109 THE POLLEN RECORD OF THE 190 M CORE FERSINA 2, AN ENTIRELY LATE-GLACIAL SEDIMENT SEQUENCE IN THE ADIGE VALLEY AT TRENTO (NE ITALY). Eberhard Grüger 1, Giulio Morteani 2 1) Dep. Palynology & Climate Dynamics, Albrecht von-Haller Institute Plant Science, Georg-August-Univ. of Göttingen, Germany 2 Gmain 1, 84424 Isen, Germany . Corresponding author: E. Grüger ABSTRACT: Pollen analysis of samples taken from the core of the water well Fersina 2 (Adige Valley, Prov. Trento, NE Italy) did not reveal any indication of an interglacial or Holocene age of the uppermost 190 m in the sediment sequence deposited in the over-deepened Adige River Valley. The sediment sequence dates entirely from late-glacial times. Four radiocarbon ages of pieces of wood indicate that about 165 m of the upper part of the profile are of Younger Dryas age. The lower part of the se- quence dates from the Allerød or Bølling/Allerød and a preceding cold phase, probably the Oldest Dryas. Accordingly the depo- sition of the sequence took about 2500 or 3500 years and was completed long before the onset of the Neolithic. Our results are in excellent agreement with findings in other formerly glaciated alpine valleys (e.g. the Traun, Salzach and Enns valleys in the Northern Alps). The final depth of the Fersina 2 well is 190 m. It is very likely that the sediment sequence found below this level in the nearby 423 m deep Fersina 1 well was also deposited after the deglaciation of the Adige Valley at the end of the last glacial period. Key words: Trentino, Adige Valley, Fersina, Younger Dryas, stratigraphy, pollen analysis 1. INTRODUCTION A fairly large number of pollen studies are available on the Late-glacial and Holocene vegetational develop- ment in the Italian Alps and the Apennine mountains (e.g. Vescovi et al., 2010; Drescher-Schneider, 2009; Ravazzi, 2007; Ravazzi et al., 2007; Vescovi et al., 2007; Filippi et al., 2007; Ravazzi et al., 2006; Pini, 2002; Grüger, 1968; Beug, 1964; etc.). These studies can help to date a series of pollen-bearing sediment samples from the 190 m deep water well Fersina 2 drilled in the city area of Trento (north-eastern Italy). The location of the coring site on the bottom of the Adige River Valley at an elevation of only about 200 m, the remarkable length of the core, and the fact that the sediment is lacustrine all combine to promise informa- tion about the development of the vegetation of the Trentino (southern Alps) since the deglaciation of the Adige Valley at the end of the Würmian glacial period. The results of the pollen study are presented below. Regional geology and sedimentology of the core sedi- ments are from Fuganti et al. (1998, 2001). Tab. 1 shows how the late-glacial biozones (based on biological phenomena) correlate with GRIP zones (based on Greenland ice core events). This system is increasingly used in more recent scientific publications on northern Italy. 2. GEOGRAPHICAL SITUATION, POLLEN SOURCE AREA AND SEDIMENTARY ENVIRONMENT Almost all of the modern catchment area of the Adige and its tributaries (more than 10,000 km2) was glaciated during the maximum of the last glacial period (Würmian). When the ice margin of the Adige Valley glacier retreated to the north from its southernmost posi- tion south of Trento, a vast lake formed in the over- deepened Adige Valley (e.g. Venzo, 1957). It must have been many km long, and it was probably as wide as the Fig. 1 - a) Position of the Lake Garda area (black square) in Italy. b) Lake Garda area. Sediment core sites mentioned in the text marked by encircled dots, elevations given in meters above mean sea level. The broken line delimits the southern border of the Würmian moraines south of Lake Garda. c) The alluvial fans of the Trento area (from Venzo 1957, changed), F = profile Fersina 2 (this paper), P3 = site studied by Venzo (1957). The non-shaded part of the inset marks the more or less flat, late-glacial bottom of the narrow Adige Valley. Adige Valley. This lake was still in existence when the Younger Dryas glaciers which did no longer reach the Trento area melted down. Huge amounts of melt water were available at that time to transport an enormous sediment load to the south and thus to fill up the over- deepened Adige Valley. In the Trento area much of the infill material was contributed by the Torrente Fersina. This river, originating in the Dolomites, enters the Adige Valley in the area of the city of Trento and cascades down from a hanging valley about 200 m above the bottom of the modern Adige Valley (Fig. 1) forming an alluvial fan there on which part of the older quarters of the city of Trento and the Fersina coring site are lo- cated. The combined sediment input of Fersina and Adige caused an extraordinarily high sedimentation rate and thus a rapid filling-up of the vast lake basin. The stratigraphy of the Fersina cores (Fig. 2) shows that predominantly clay and silt, but also fine sand and some sandy gravel, were deposited in the Trento area. In the study area the Adige Valley is bordered to the West and East by mountain ranges rising to more than 2000 m. The area is consequently characterized by a much diversified altitudinal zonation of the vegeta- tion, including treeless alpine meadows. The pollen produced in these varied vegetational zones was brought to the coring site probably predominantly by the wind. Non- anemogamous species, even components of the bankside vegetation, had almost no chance to be represented in the sediment as the later coring site was probably about 1 km away from the nearest lake shore. Additional pollen was transported to the Trento lake basin by the rivers from more distant regions, con- temporary pollen and possibly also pollen reworked from older sediment well upstream from the coring site. An unknown quantity of the pollen floating on or suspended in the lake water was carried away with the excessive river water. Due to the great sedimentation rate (see below), complete mixing of the annual input of sediment might have been impossible. If so, the pollen contents even of different subsamples of the same annual layer might differ to some degree. From the above, it follows that the sedimentary environment of the Fersina core was so different from that of smaller lake basins and peat bogs - the usual source of information on the former vegetation and of its changes - that a direct comparison of pollen values – especially of less frequent taxa - must not be done. The main features of the vegetational development, how- ever, will remain visible despite loss of pollen and inade- quate size of samples. 3. SAMPLING The well Fersina 2 was drilled (percussion drill) in March/April 1997 in the southern part of Trento near the present-day confluence of the rivers Fersina and Adige (46° 02’ 50” N, 11° 07’ 08.5” E; 190.6 m a.s.l.). The drill- ing was ended at a depth of 190 m when the aquifer was reached. A continuous casing was used during drilling to prevent collapse of the well walls and hence incorporation of younger sediment in the samples. Therefore disturbances of the sedimentary and pollen record can be excluded. The samples were labelled according to the sampling interval of about 1 meter. A Grüger E., Morteani G. 102 Table 1 - Correlation of late-glacial GRIP zones and Swiss biozones according Ammann et al., 2012 and van Raden et al., 2012. Table 2 - Calibration of the conventional radiocarbon dates from Trento and wiggle matching of the Fersina 2 dates. 103 F ig . 3 - P o lle n d ia g ra m . T h e p o lle n v a lu e s a re s h o w n a s p e rc e n ta g e s o f th e p o lle n t o ta l. P o lle n s u m , p o lle n c o n ce n tr a tio n s, a n d a m o u n ts o f st o m a ta a re g iv e n a s n u m b e rs . N o te t h a t th e s ca le s ch a n g e i n o rd e r to s a ve s p a ce . D o ts d e n o te w h e re t h e p e rc e n ta g e v a lu e s a re ≤ 0 .3 o r w h e re o n ly o n e s to m a o f P in u s w a s fo u n d . R a d io ca rb o n d a te d le ve ls : I: 1 1 5 0 0 ± 1 5 6 c a l a B P ; II : 1 1 6 5 3 ± 2 1 4 c a l a B P ; II I: 1 2 1 0 9 ± 2 2 3 c a l a B P ; IV : 1 2 3 7 9 ± 1 6 6 c a l a B P . F ig . 2 - S tr a tig ra p h y o f th e c o re s F e rs in a 1 a n d F e rs in a 2 , d e si g n b y A . F u g a n ti, ch a n g e d b y G . M o rt e a n i a n d C . P re in fa lk , d e p th s o f p o lle n sa m p le s a d d e d b y E . G rü g e r. …. Fersina 2, an entirely late-glacial sediment sequence ….. sediment sample called, e. g., “37 m” represents the sediment found between 36 and 37 m depth. In total, 100 irregularly shaped lumps of sediment (100 to 200 cm3) were available for pollen analysis. All samples consist of white silt, clay, and fine sand. The well Fersina 1, located some 36 m from Fersina 2, was drilled in the same year. This drilling was stopped at 423 m without yet having reached bedrock. No pollen samples are available from this well, but the huge thickness of the sediment clearly shows that the Adige River Valley was over-deepened at least up to Trento. Fig. 2 shows the stratigraphy of the cores Fersina 1 and 2 (Fuganti et al., 2001 and own data). 4. RADIOCARBON DATING / WIGGLE MATCHING Four pieces of wood found in the Fersina 2 sediment were radiocarbon dated (Table 2). The resulting conven- tional and calibrated radiocarbon ages were published by Fuganti et al. (1998). Since an improved calibration data set is now available (Reimer et al., 2009), the ra- diocarbon ages from Trento were re-calibrated using the CalPal software (Weninger & Jöris, 2008). Following a proposal of M. Geyh, the former director of the Han- nover radiocarbon laboratory, the radiocarbon ages of the four pieces of wood were submitted to “wiggle matching”, kindly carried out by B. Weninger (University of Cologne). To accomplish wiggle matching, a calendric dis- tance of 10 ± 50 calyrs was assumed for the two younger pieces of wood, and a distance of 90 ± 50 ca- lyrs for the two older pieces. Thus fitted into the Cal- Curve, radiocarbon ages of 11,603 ± 146 cal a BP (final year) and of 12,281 ± 142 cal a BP (final year) result for the two pairs of pieces of wood. 5. POLLEN ANALYSIS 5.1. Pollen preparation and pollen diagram The aliquots for pollen analysis (2.3 to 14 cm3, av- erage 5.2 cm3) were taken in the laboratory from the inner part of the cored material to prevent contamination with recent pollen. The standard mode of sample prepa- ration, with HCl and HF and wet ultrasonic sieving as the final step (width of mashes 8 µm), did not produce a pollen concentration sufficient for a reliable pollen count. Treatment with HCl and NaOH, followed by flotation with sodium tungstate solution (Na2WO4, D = 2.05) as de- scribed by Eisele et al. (1994), but without acetolysis, proved to be more effective. To allow the calculation of pollen concentrations, one Lycopodium spore tablet (containing 10,679 ± 191 spores) was added to each pollen sample at the begin- ning of the treatment (Stockmarr, 1971). The pollen counts range from 204 to 1225 (mean 592) grains per sample. Four samples only contained fewer than 484 pollen grains. The complete results of pollen analysis are presented in Fig. 3 and Table 3. Supplementary data are available at doi:10.1594/PANGAEA.793027. The pollen diagram was drawn using PanPlot (Diepenbroek et al., 2001), with graphical details being improved with microsoft paint. All given pollen values are percentages of the pollen total. 5.2. The pollen record The salient feature of the pollen diagram (Fig. 3) is the dominance of the Pinus sylvestris type in all pollen spectra (88.4 to 34.3 %, mean 71.5 %). Next in fre- quency are the Pinus cembra type (8.8 to 1.3 %, mean 5.1 %) and Betula (1.5 to 18.3 %, mean 5.2%). All other tree pollen types were found less frequently; indeed, their pollen values rarely exceed 1%. Tree species with such low pollen values – among them the thermophilous tree species – can hardly have been of any importance in the area or were even missing when the profile Fersina 2 was deposited. But pine trees occurred in the area, as evidenced by Pinus stomata being present in 14 out of 17 samples. Temporary increases of the NAP (nonarboreal pollen) values combined with a frequent 104 Grüger E., Morteani G. Table 3 - Rare pollen and spore types (numbers). occurrence of pollen grains of shade intolerant species indicate former changes of the vegetation. Details will be discussed in the following. The two lowermost samples (185 and 186 m) are comprised of clay. They show the highest NAP values recorded in the pollen diagram (21.3 and 39%, respec- tively). Up to 12.7 % of the pollen total was produced by the shade intolerant genus Artemisia. Artemisia species never occur in a forest. The same is true for the Cheno- podiaceae (9.8 and 5.3 %). The percentages of the in- determinable pollen types (Indeterminata), certainly not pollen of woody species, are high (11.1 and 6.6 %). The diversity of nonarboreal pollen types (NAP) was never greater than at the time of deposition of these two sam- ples. Among the woody taxa, the pollen frequencies of the pioneer shrub Hippophae (7.4 and 13.3 %), of Juni- perus (0.4 and 0.6 %), and of two Ephedra -pollen types (up to 1 %) are of special interest. These shrubs are sunlight-demanding and cannot thrive in the shade of trees. On the other hand, no other sample of the profile contains more pollen of Picea (2.0 and 2.2 %), Corylus (0.8 and 1.0 % - not much for this shrub!), and Ulmus (0.4 and 0.6 %) than these two samples. Pollen of Quer- cus and Tilia was not found. Pollen concentration is medium in sample 185 m, but very low in sample 186 m (see Fig. 3). A low pollen concentration is caused either by the absence of pollen-producing plants or by a high rate of sedimentation. Never during the deposition of the Fersina 2 sedi- ment was the vegetation around the coring site more open and the climatic conditions less favourable than at the time when the clay below the sandy gravel (i.e. be- low 184 m, Fig. 2) was deposited. The region was not forested, but was probably not treeless. The pine values are lower than further up in the profiles, but – together with the occurrence of one stoma of a pine needle – possibly sufficiently high to assume the presence of some pine trees in the area. The comparatively high values of Picea, Ulmus and Corylus are somewhat con- founding. It is possible that these pollen grains come from the reworking of older sediments. The presence of the mentioned species in the area is unlikely in view of the reconstructed situation. The pollen spectra of the samples 172 and 171 m, taken in a layer of sandy gravel (Fig. 2), indicate warm conditions during the time of their sedimentation. High pine pollen sums (91.3 and 89.3 %), the very low NAP values (1.9 and 3.3 %), and the only occasional occur- rence of pollen grains of shade intolerant taxa indicate that the region was forested when the two samples were deposited. Pollen of the thermophilous trees Quercus and Ulmus is rare. On the other hand, Tilia cordata-type pollen is in no sample more frequent than in sample 171 m (1.8 %). The pollen spectrum of sample 161 m, taken from sandy gravel, shows with 5.2 % Artemisia, 3.9 % Chenopodiaceae, 8.2 % Indeterminata, and a NAP sum of 21.5 % pointing to a return to climatic conditions simi- lar to those prevailing when the clay of samples 185 m and 186 m was deposited. Evidently the forest had dis- appeared again. Only a few tree specimens may have been left in the area. It is remarkable that in this sample the prequaternary reworked sporomorphs reach their absolute maximum, probably a result of intensive ero- sion of prequaternary pollen-bearing sediments by the rivers Fersina and Adige. Consequently the pollen con- centration is low. With only 396 pollen grains per 1 cm3 of sediment it is the lowest concentration value in our series of samples. The samples 154 m (sandy gravel) (Fig. 2), 139 and 124 m (silty clay with sandy levels) document, with NAP values around 10 % and the occurrence of shade intolerant taxa, climatic conditions that were similar, although probably somewhat less severe, than those indicated by the pollen preserved in the lower part of the profile, below and above the 171 and 172 m samples (see above). Pieces of well-preserved wood prove the presence of Pinus cembra (sample 130 m) and of Pinus mugo/sylvestris (sample 139 m) in the region. The two pieces of wood were radiocarbon dated by wiggle matching to 12,280 ± 142 cal a (Table 2). After the deposition of sample 124 m, the ecological situation changed markedly. Starting with sample 114 m, the NAP values are distinctly lower than before (mostly less than 5 %). Pinus stomata are more fre- quent. Pollen of Hippophae and Artemisia, although present in all but one (Hippophae) and three (Artemisia) of the 17 studied samples, is rare. Ephedra pollen was found in two of the nine pollen samples above the 124 m level only (3 pollen grains). These changed conditions lasted throughout all the time represented by the rest of the profile. Birch was probably present. Pollen of ther- mophilous deciduous trees like Quercus, Ulmus, and Tilia is recorded as before, with variable, but low, val- ues. A few specimens of these trees only can have grown in the vicinity if their pollen grains did not come with the wind from more distant sites or was reworked. In a layer of sand two pieces of Juniperus wood (samples 32 and 33 m), but no pollen grains of this shrub, were found. The radiocarbon age of the two pieces of wood is 11,603 ± 146 cal a BP (wiggle match- ing, Table 2). 6. DISCUSSION 6.1 The age of the sediment of the Fersina 2 well – general remarks The deposition of the lacustrine sediment studied can have started only after the retreat of the Adige gla- cier to the north beyond the Trento area. The sediment met by the Fersina 1 and 2 wells can therefore be of late -glacial up to postglacial age only. The following dating relies on pollen analyses and on four radiocarbon dates. Pollen analytical dating is based on the fact that most European tree species and important groups of herbs are anemogamous. Because such plants let the wind disperse their pollen grains, pollen diagrams of large areas are usually similar to each other. Therefore, a new profile can usually be dated by comparing the composition of its pollen assem- blages and their changes with that of other not-too- distant, reliably dated pollen sequences. The sampling sites next to Trento with pollen dia- 105 …. Fersina 2, an entirely late-glacial sediment sequence ….. grams suitable for a comparison with the Fersina pollen data are mentioned below and are shown in Fig. 1. They are not the only pollen diagrams from this area, but the others are not detailed enough to be useful for our pur- pose, e.g. the preliminary diagram (with only seven pol- len curves) from a peat profile at Isera near Rovereto, about 20 km south of Trento in the Adige Valley (Calderoni et al., 1996) and Kofler’s diagrams from Bon- done and Lago delle Buse (Kofler, 1994). 6.1.1 The Postglacial Three detailed pollen analytical studies from the Trento area are available to answer the question whether the Fersina 2 profile can be of postglacial age. Two of the studied profiles come from filled-up lakes (Grüger, 1968), namely Bondone, at 1550 m, only 8 km southwest of Trento, and Fiavè, at 654 m, about 25 km off Trento, west of Bondone. The third profile is from the extant lake of Lavarone, at 1115 m, about 20 km south- east of Trento (Filippi et al., 2007). The most remarkable event during the early Post- glacial was the expansion of the thermophilous decidu- ous tree genera Quercus, Ulmus, Tilia, Fraxinus, and Acer and of Corylus during the Preboreal (11,500 to 10,200 cal a BP), the first biozone of the Postglacial. Although the center of the distribution of these species lies in the lower altitudinal zones, the sum of their pollen values rises at Fiavè (at 654 m) from low values to 30%, at Lavarone (1115 m) to 25 %, and to around 10% fur- ther up at Bondone (1550 m). It should be expected that the contemporary sedimentation in the Adige Valley produced, at least in the uppermost part of the profile Fersina 2, values not less than those found at Fiavè and Lavarone; but the Fersina pollen data are mostly much lower. In 11 of the 17 studied samples the pollen values of the thermophilous deciduous tree genera are ≤ 1%, and in a further four samples between 1 and 2%. Only in samples 124 and 139 m are higher percentages reached (4.3 and 3.5 %, respectively, mainly oak pol- len). These two samples can, however, not be of post- glacial age if the results of the radiocarbon dating are accepted. Also no hints can be found in the Fersina 2 pollen diagram of later important vegetational events like the immigration and spreading of Fagus and Abies trees, which were present in the area from about 8000 cal a BP. The pollen curve of Fagus, an important constituent of the montane forests, climbs to more than 40 % of the tree pollen sum at Lavarone and to more than 10 % at Bondone, and Abies reaches more than 15 % at both localities (J. Grüger, unpublished, counts available un- der http://doi.pangaea.de/10.1594/PANGAEA.763901). In the Fersina 2 samples, not a single pollen grain of Fagus, and only one pollen grain of Abies have been found. Evidently no part of the Fersina 2 pollen diagram shows characteristics of the Holocene forest develop- ment. No section of the profile Fersina 2, the uppermost sandy gravel included, can be of postglacial age. The sediment of the entire Fersina 2 profile must have been deposited during late-glacial times, with the only excep- tion of the colluvial material forming the modern surface in the area, a layer being too thin to be shown in the stratigraphical scheme (Fig. 2). 6.1.2 The Late-Glacial The Younger Dryas, the final part of the last glacial period, started around 12,700 cal a BP; it ended when the Holocene vegetation began to develop in the study area about 11,600 to 11,500 cal a BP (Vescovi et al., 2007). The radiocarbon ages of the four pieces of pine and juniper wood found in the Fersina 2 sediment (Table 2) prove that these tree species had been grow- ing in the study area during Younger Dryas times. Being well preserved, these pieces of wood were probably not repeatedly transported, exposed, and temporarily em- bedded, so they most likely date the sediment in which they were discovered. If so, at least the part of the core between 32 and 139 m must be of Younger Dryas age. The pollen spectra of the samples 3, 14, and 24 m do not distinctly differ from those of proven Younger Dryas age below. Apparently a Younger Dryas type of vegetation was still present in the Trento area when the uppermost 30 m of sediment (mainly sandy gravel) were deposited so that this sediment can hardly be of post- glacial age. The radiocarbon age of the upper piece of wood (Table 2) allows, however, a deposition time of a few decades only so that a huge annual sedimentation rate must be assumed. To explain this one must not assume a climatic change, it is sufficient to consider the special conditions of sedimentation on an alluvial fan. Rivers usually shift over their alluvial fans. Where they flow they can accumulate huge amounts of sediment in a short time. The coarseness of the uppermost sedi- ment indicates that at the time of the deposition of this material the River Fersina flowed and dropped its sedi- ment load where the Fersina core was taken. Thus the thickness of the uppermost layer of sandy gravel cannot be used as an argument against a Younger Dryas age of this part of the profile as proposed by the pollen data. Younger Dryas vegetation is documented in the profile downwards to sample 161 m. The next older pollen spectra (171 and 172 m) indicate distinctly warmer, though not interglacial, conditions. These spec- tra can only be of Allerød age. Samples 185 and 186 m, the oldest samples of the Fersina 2 profile, contain a pollen flora which indicates cold conditions. This climatically unfavourable period can either be the colder middle part of the Allerød or one of the preceding cold phases of the late-glacial, namely the Older Dryas or the Oldest Dryas. The Allerød is a tripartite interstadial, that is, it has a climatically less favourable phase sandwiched be- tween warmer phases at its beginning and before its end. The cooler phase, called the Gerzensee oscillation, lasted a few hundred years only (250 years according to Greenland ice core isotope data, GRIP zone GI-1b, Björk et al., 1998, or 400 years according to Swiss lake sediment isotope data, Schwander et al., 2000). Al- though distinct in the oxygen isotope records, only a weak reaction of the vegetation is recorded by the pol- len data. In Switzerland (sites Gerzensee and Leysin; Wick, 2000) and western Germany (site Meerfelder Maar; Stebich, 1999), a decrease in the importance of Betula pollen (percentages as well as concentrations and influx) is the most indicative reaction. The more 106 Grüger E., Morteani G. thermophilous vegetation growing at that time in Tus- cany, 340 km south of Trento (Lago dell’Accesa, Dre- scher-Schneider et al., 2006), however, responded dis- tinctly to the changed conditions as is clearly indicated by lowered Quercus and higher NAP (mainly Poaceae) values, and by a series of minor changes of a few fur- ther pollen curves (Accesa II oscillation). The pollen diagrams from the Trento area, Bondone, Fiavè (Grüger, 1968) and Lavarone (Filippi et al., 2007) do not show any indication of a climatic deterioration during the Allerød, probably as no plant species requiring higher temperatures were present at that height and were at best rare at lower altitudes. The pollen diagram of Sal- tarino Sotto only, a low-lying site (194 m) 75 km south- west of Trento near Lake Garda (Grüger, 1968), shows changes in the Allerød time samples (320 cm and 325 cm) which most likely are the expression of a cooler climate. Betula values diminished by about 9 %, while NAP (about 10 %, mainly Poaceae and Artemisia) and juniper values (>1 % instead of 0.4 % before and 0.8 % after the oscillation) increased. The rise of the QM (Quercetum mixtum) curve – here mainly oak – stopped, but the rise continued when the climatic conditions im- proved, so that QM values of 30% were reached at the end of the Allerød. The QM values dropped to a much lower level during the following Younger Dryas stadial. Considering the weak response of the vegetation around Saltarino Sotto to the Gerzensee oscillation, it appears unlikely that the change from forest (samples 171 and 172 m) to an open type of vegetation (185 and 186 m) at Trento was caused by this climatic deteriora- tion. Consequently, part of the lower sandy gravel (Fig. 2) must represent the entire Allerød, the Bølling possibly included, interstadials during which the annual sedimen- tation rate was apparently much reduced compared to that of Younger Dryas time. The two lowermost samples of core Fersina 2 (185 and 186 m) must date from either the Older or the Old- est Dryas. It is not possible to decide which of these two stadial phases is the one truly represented. Fig. 2 shows that the lowermost clay of the profile Fersina 2 can be correlated with the uppermost part of a thick sequence of rather homogeneous fine grained lacustrine sediment in the 423 m long profile Fersina 1. As no drastic changes in the composition of the sedi- ments are apparent, it is likely that the lower part of the profile Fersina 1 is also of late-glacial age. Older sedi- ment is not likely to have escaped erosion by the water running off under the large, melting Adige glacier, espe- cially in the middle of the valley where the coring site is located. Pre-Würmian sediment can be preserved only in rare sheltered positions. 7. CONCLUSIONS In contrast to that supposed so far, the Fersina 2 profile does not reach back in time to the last intergla- cial. Instead, it was deposited totally during the young- est part of the Late-glacial, i.e., during Upper Palaeo- lithic times. The sedimentation of Fersina 2 (190 m of clay, sand and gravel) took only about 2500 to 3500 years, and it was finished before the Holocene (= Post- glacial) development of species-rich forests began. This happened in the Trento area around 11,500 cal a BP, in archaeological terms, during Mesolithic times, several thousand years before the Neolithic culture reached the Trento area. The Ancient Neolithic units at the archaeo- logical sites Riparo Gaban and Ala le Corone in Trento date from 6749-6993 and 6551-6892 cal a BP, respec- tively (Angelucci & Bassetti, 2009). That the entire Fersina 2 sediment sequence was deposited during the Late-glacial is in agreement with the situation in the Eastern Alps, where the filling up of glacially over-deepened valleys was completed before the Postglacial started (van Husen, 1979). The results of the Fersina 2 study necessitate re- considering the dating of the 193 m long profile P3, taken about 5 km north of Fersina 2 in the industrial zone of Trento at an elevation of 190 m (Venzo, 1957, 1979). Venzo stated that “The stratigraphical and sedi- mentological data testify to the existence of an ancient lacustrine basin immediately following the last glaciation in the Adige River Valley (I dati stratigrafici e sedimen- tologici della serie testimoniano l’esistenza nella zona di Trento di un antico bacino lacustre di grande esten- sione, immediatamente successivo al ritiro dell’ultima glaciazione)“. Venzo called the basin also a “bacino lacustre post-würmiano”, “a post-glacial lacustrine ba- sin”, definitely referring to the proper Postglacial (=Holocene). Venzo totally disregarded the preceding more than 15,000-year-long late-glacial period, during which the Adige Valley of the Trento area was ice free, a necessary condition of the deposition of the Fersina 1 and 2 sediment. As Venzo’s site P3 (Venzo 1957) must have been located in the area of the same late-glacial lake as the Fersina 2 site, the P3 profile must also be of late-glacial age. This is even in agreement with Venzo’s note cited above on the formation of the lake basin. The same applies to the 30 m deep well Via Giuseppe Verdi in the center of Trento, 3 km north of the Fersina 1 and 2 wells (Fuganti et al., 1998). It may be surprising that in the core of this well, 11 m below the modern surface, a piece of wood with a radiocarbon age of only 4710 ± 80 cal a BP (Table 2) was found in late- glacial sediment. This discrepancy can be explained if it is assumed that the meandering rivers Adige and Fersina cut channels into the late-glacial sediment which were later filled with material from the surround- ings containing wood of (in this case) subboreal age. The number of pollen types found in the Fersina 2 samples is low when compared with the Younger Dryas pollen samples of the other mentioned profiles. This difference results from the fact that their Younger Dryas samples contain the pollen rain of many years (e.g. about 14 and about 20 years, respectively, in the pro- files Saltarino Sotto and Bondone) whereas the Fersina 2 samples contain, due to the huge annual sedimenta- tion rate, less than one year’s pollen precipitation only. As pollen analytical dating is mostly based on common pollen types, the Fersina 2 pollen data are sufficient to date the core, they but are not detailed enough to dis- cuss problems of the North Italian vegetation history. ACKNOWLEDGEMENTS We thank Andrea Fuganti (Trento) very much for 107 …. Fersina 2, an entirely late-glacial sediment sequence ….. the permission to study the samples of the Fersina 2 well, and for discussions about the morphology of the Trento area and the influence of the River Fersina on the sedimentation processes in the Adige Valley. We owe the wiggle matching to Bernhard Weninger and Mebus A. Geyh. Ruth Drescher-Schneider, Angelika Kleinmann, Werner Kofler, Josef Merkt, and Roberta Pini helped by sending reprints, by contributing informa- tion, and by discussing special problems. The layout of Fig. 2 is the work of Christine Preinfalk. Gisela and Wolfgang Tambour touched up the original drafts of the figures and converted them into the necessary elec- tronic files. Dirk van Husen patiently taught the first au- thor the principles of sedimentation in alpine valleys. Dean Hansen improved the English. Many thanks to all of them! REFERENCES Ammann B., van Leeuwen J.F.N., van der Knaap W.O., Lischke H., Heiri O, Tinner W. (2012) - Vegetation responses to rapid warming and to minor climatic fluctuations during the Late-Glacial Interstadial (GI -1) at Gerzensee. Palaeogeography, Palaeoclima- tology, Palaeoecology. http://doi.10.1016/ j.palaeo.2012.07.010 Angelucci D.E., Bassetti M. (2009) - Humans and their landscape from the alpine last glacial maximum to the Middle Holocene in Trentino: geoarcheological considerations. Preistoria alpina, 44, 1-20. Beug H.-J. (1964) - Untersuchungen zur spät- und post- glazialen Vegetationsgeschichte im Gardaseege- biet unter besonderer Berücksichtigung der medi- terranen Arten. Flora, 154, 401-444. Björck S., Walker M.J.C., Cwynar L.C., Johnsen S., Knudsen K.-L., Lowe J.J., Wohlfarth B. (1998) - An event stratigraphy for the Last Termination in the North Atlantic region based on the Greenland ice-core record: a proposal by the INTIMATE group. Journal of Quaternary Science, 13, 283– 292. Calderoni G., Finotti F., Iliceto V., Leonardi D., Paganelli A. (1996) - Topography-based identification of a palaeopeat-bog at Isera, near Rovereto (Trento, Italy) and first stratigraphic, radiocarbon and paly- nological results. Il Quaternario – Italian Journal of Quaternary Sciences, 9, 671-678. Diepenbroek M., Grobe H., Sieger R. (2001) - PanPlot. http: //www.pangaea.de/Software/PanPlot. Drescher-Schneider R. (2009) - La storia forestale delle Alpi Sud-Orientali e del margine pedemontano durante gli ultimi 25 mila anni / The forest history in the Southeastern Alps and the foothills during the last 25,000 years. In: Peresani M., Ravazzi C. (eds.): Le foreste dei cacciatori paleolitici. Supple- mento al Bollettino della Società Naturalisti Silvia Zenari, 27-51 and 52-64, Pordenone, 2009. Drescher-Schneider R., de Beaulieu J.-L., Magny M., Walter-Simonnet A.-V., Bossuet G., Millet L., Brugiapaglia E., Drescher A. (2006) - Vegetation history, climate and human impact over the last 15,000 years at Lago dell’Accesa (Tuscany, Cen- tral Italy). Vegetation History and Archaeobotany, 16, 279-299. Eisele G., Haas K. Liner S. (1994) - Methode zur Aufbe- reitung fossilen Pollens aus minerogenen Sedi- menten. (Supplement to: Frenzel B. - Über Proble- me der holozänen Vegetationsgeschichte Ostti- bets). Göttinger Geographische Abhandlungen, 95, 165-166. Filippi M.L., Heiri O., Arpenti E., Angeli N., Bortolotti M., Lotter A.F., van der Borg K. (2007) - Evoluzione paleoambientale dal Tardoglaciale a oggi ricostrui- ta attraverso lo studio dei sedimenti del Lago di Lavarone (Altopiano di Folgaria e Lavarone, Tren- tino). Studi Trentini di Scienze Naturali, Acta Ge- ologica, 82 (2005), 279-298. Fuganti A., Bazzoli G., Morteani G. (1998) - The Qua- ternary evolution of the Adige Valley near the city of Trento (Northern Italy) as deduced from wells and radiocarbon dating. Preliminary results. Studi Trentini di Scienze Naturali, Acta Geologica, 73 (1996), 93-97. Fuganti A., Bazzoli G., Morteani G. (2001) - La genesi della Valle dell’Adige. Studi Trentini di Scienze Naturali, Acta Geologica, 77 (2000), 205-219. Grüger J. (1968) - Untersuchungen zur spätglazialen und frühpostglazialen Vegetationsentwicklung der Südalpen im Umkreis des Gardasees. Botanische Jahrbücher, 88, 163-199. Husen D. van (1979) - Verbreitung, Ursachen und Fül- lung glazial übertiefter Talabschnitte an Beispielen in den Ostalpen. Eiszeitalter und Gegenwart, 29, 9 -22. Kofler W. (1994) - Die Vegetationsentwicklung im Spät- paläolithikum und Mesolithikum im Raume Trient (Lo sviluppo della vegetazione nel Tardo Paleoliti- co e Mesolitico nel Trentino). Preistoria Alpina – Museo Tridentino di Scienze Naturali, 28 (1992), 83-103. Pini R (2002) - A high-resolution Late-Glacial – Holo- cene pollen diagram from Pian di Gembro (Central Alps, Northern Italy). Vegetation History and Ar- chaeobotany, 11, 251-262. Ravazzi C. (2007) - Il Tardoglaciale: suddivisione strati- grafica, evoluzione sedimentaria e vegetazionale nelle Alpi in Pianura Padana. Studi Trentini di Scienze Naturali, Acta Geologica, 82 (2005), 17- 29. Ravazzi C., Donegana M., Vescovi E., Arpenti E., Cac- cianiga M., Kaltenrieder P., Londeix L., Marabini S., Pini R., Vai G.B, Wick L. (2006) - A new Late- glacial site with Picea abies in the northern Appen- nine foothills: an exception to the model of glacial refugia of trees. Vegetation History and Archaeo- botany, 15, 357-371. Ravazzi C., Peresani M., Pini R., Vescovi E. (2007) - Il tardoglaciale nelle Alpi Italiane e in Pianura Pada- na. Evoluzione stratigrafica, storia della vegetazio- ne e del popolamento antropico. Il Quaternario - Italian Journal of Quaternary Sciences, 20, 163- 184. Reimer P.J., Baillie M.G.L., Bard E., Bayliss A., Beck J.W., Blackwell P.G., Bronk Ramsey C., Buck C.E., Burr G.S., Edwards R.L., Friedrich M., Grootes P.M., Guilderson T.P., Hajdas I., Heaton 108 Grüger E., Morteani G. T.J., Hogg A.G., Hughen K.A., Kaiser K.F., Kromer B., McCormac F.G., Manning S.W., Reimer R.W., Richards A.A., Southon J.R., Talamo S.,Turney C.S.M., van der Plicht J., Weyhenmeyer C.E. (2009) - IntCal09 and Marine09 radiocarbon age calibration curves, 0– 50,000 years cal BP. Radiocarbon, 51, 1111- 1150. Schwander J., Eicher U., Ammann B. (2000) - Oxygen isotopes of lake marl at Gerzensee and Leysin (Switzerland), covering the Younger Dryas and two minor oscillations, and their correlation to the GRIP ice core. Palaegeography, Palaeoclimatolo- gy, Palaeoecology, 159, 203-214. Stebich M. (1999) - Palynologische Untersuchungen zur Vegetationsgeschichte des Weichsel-Spätglazial und Frühholozän an jährlich geschichteten Sedi- menten des Meerfelder Maares (Eifel). Dissertatio- nes Botanicae, 320, 127 pp. Stockmarr J. (1971) - Tablets with spores used in abso- lute pollen analysis. Pollen et Spores, 13, 615- 621. Van Raden U.J., Colombaroli D., Gillia A., Schwander J., Bernasconi S.M., van Leeuwen J., Leuenber- ger M., Eicher U. (2012) - High-resolution late- glacial chronology for the Gerzensee lake record (Switzerland): δ18O correlation between a Ger- zensee-stack and NGRIP. Palaeogeography, Pa- laeoclimatology, Palaeoecology. http:// dx.doi.org/10.1016/j.palaeo.2012.05.017 Venzo G.A. (1957) - Ricerche sulla serie lacustre e fluviale attraversata de pozzi trivellati nella zona industriale di Trento. Giornale di Geologia, 26, 173 -188. Venzo G.A. (1979) - Glaziale Übertiefung und postglazi- ale Talverschüttung im Etschtal im Raum von Trient (Italien). Eiszeitalter und Gegenwart, 29, 115-121. Vescovi E., Kaltenrieder P., Tinner W. (2010) - Late- glacial and Holocene vegetation history of Pavullo nel Frignano (Northern Appennines, Italy). Review of Palaeobotany and Palynology, 160, 32-45. Vescovi E., Ravazzi C., Arpenti E., Finsinger W., Pini R., Valsecchi V., Wick L., Ammann B., Tinner W. (2007) - Interactions between climate and vegeta- tion during the Lateglacial period as recorded by lake and mire sediment archives in Northern Italy and Southern Switzerland. Quaternary Science Reviews, 26, 1650-1669. Weninger B., Jöris, O. (2008) - A 14C age calibration curve for the last 60 ka: the Greenland-Hulu U/Th timescale and its impact on understanding the Middle to Upper Paleolithic transition in Western Eurasia. Journal of Human Evolution, 55, 772-781. Wick L. (2000) - Vegetational response to climatic changes recorded in Swiss Late-glacial lake sedi- ments. Palaeogeography, Palaeoclimatology, Palaeoecology, 159, 231-250. 109 …. Fersina 2, an entirely late-glacial sediment sequence ….. Ms. received: April 05, 2013 Final text received: Jun 25, 2013