Review of Palaeobotany and Palynology 141 (2006) 53 – 81 www.elsevier.com/locate/revpalbo Wetlands in the Venetian Po Plain (northeastern Italy) during the Last Glacial Maximum: Interplay between vegetation, hydrology and sedimentary environment A. Miola a,⁎, A. Bondesan b , L. Corain c , S. Favaretto b , P. Mozzi b , S. Piovan b , I. Sostizzo b b a Dipartimento di Biologia, Università di Padova, via Ugo Bassi 58/B- 35121 Padova, Italy Dipartimento di Geografia “G. Morandini”, Università di Padova, via del Santo 26- 35123 Padova, Italy c Dipartimento di Scienze Statistiche, Università di Padova, via C. Battisti, 241/243- 35121 Padova, Italy Received 14 December 2004; accepted 20 March 2006 Available online 7 July 2006 Abstract In the low Venetian plain (northeastern Italy) thick sequences of silt and sand layers alternate with common, thin layers of peat and organic silt; the organic layers in the topmost 30 m of the Late Pleistocene alluvial series span between 23,000 and 14,000 yr BP (radiocarbon dating), in an area measuring 100 km by 30 km. They indicate broad areas where wetlands developed. We aim to understand the features and the origin of the wetlands by undertaking sedimentological, pollen, non-pollen palynomorph and plant macrofossil analyses. Thirteen cores were drilled in the central zone of the low Venetian plain near the coast of the Adriatic Sea and 79 samples were analysed. The palaeoenvironmental reconstruction based on previous pollen analysis did not emphasize the areas where peat layers were formed, suggesting a homogenous steppe environment, typical of a cold and dry climate. They were probably waterlogged for most of the year allowing the formation of peat and the development of local plant communities of mainly aquatic species. Macrofossil and pollen analyses suggest that herbaceous plants, such as Cyperaceae and Poaceae (probably Carex fusca and Phragmites australis), and brown mosses (mainly Scorpidium scorpioides) were the most important components of wetland communities. Fossils of obligate aquatic organisms indicate open water environments, these include macrofossils of Nymphaea, Characeae, Bryozoa and Potamogeton, and nonpollen palynomorphs such as algal resting cells, free cells and colonies (Zygnemataceae, Spirogyra, Mougeotia, Closterium idiosporum, Type 225, Type 229, Botryococcus, Pediastrum cf. boryanum, P. cf. simplex, Ceratium hirundinella, Tetraedron cf. minimum and Type 333), oocytes of aquatic invertebrates (Type 353A and 353B) and incompletely known types probably of algal origin (Type 303, Type 74, Type 128A and 128B). A discontinuous occurrence of fungal spores and other microfossils (Type 200, Gaeumannomyces (Type 126), Glomus (Type 207), Type 351, Type 79, and incompletely known types) suggests frequent fluctuations of the water depth with periodic emersions of the bottom of the ponds or fens. The water quality preferred by the identified species, or suggested in literature for the fossil types, is mainly eutrophic to mesotrophic and rich in cations. Peatland formed in wide, low-lying areas between the fluvial ridges which were periodically inundated by the fluctuating groundwater. Peat accumulated in continuous layers only where the fen organic deposition prevailed the alluvial minerogenic sediment. When alluvial deposition buried the organic deposit, the peat level was incorporated into the stratigraphic record. © 2006 Elsevier B.V. All rights reserved. Keywords: Venetian Po Plain; Last Glacial Maximum; pollen; non-pollen palynomorphs; plant macrofossils; palaeohydrology ⁎ Corresponding author. Tel.: +39 0498276267; fax: +39 0498276260. E-mail address: antonella.miola@unipd.it (A. Miola). 0034-6667/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.revpalbo.2006.03.016 54 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 1. Introduction The Last Glacial Maximum (LGM) biome reconstruction of Elenga et al. (2000) shows a remarkably homogeneous pattern in southern Europe and in the lands around the Mediterranean basin. Steppe (i.e. grassland or shrubland) is indicated as the dominant vegetation. Peyron et al. (1998) reconstructed LGM climate of Europe from pollen data using the best analogues approach. The reconstruction shows that the Mediterranean region was relatively wetter than northern Europe. Terrestrial records document paleohydrological fluctuations during the LGM with higher lake levels than present in northern Mediterranean regions (Prentice et al., 1992; Roberts and Wright, 1993; Yu and Harrison ,1995; Chondrogianni et al., 2004; Valero-Garcés et al., 2004). The conflict between paleobotanical and paleohydrological records has been solved by postulating a climatic instability (Prentice et al., 1992). Lacustrine records from central Italy (Chondrogianni et al., 2004) and from northeastern Spain (Valero-Garcés et al., 2004) present a well identified cyclic instability of LGM climate through the results of multidisciplinary studies. Five wet-warm/dry-cold cycles between 24,000 and 15,000 cal yr BP were distinguished and documented by several paleoecological parameters in Lake Albano (central Italy). The alluvial plain sedimentary sequences are much less studied for LGM climate reconstruction because they generally cannot provide “high resolution” records. However, in the north Italy Po Plain they constitute almost the only chance to study the paleoenvironment and paleoclimate of the LGM period. Large portions of the northeastern Po plain formed during the LGM (Castiglioni, 1999; Bondesan et al., 2002; Marchetti et al., 2004). This period featured great sedimentary activity of the Adige, Brenta, Piave and Tagliamento Rivers, which received fluvio-glacial outwash from the eastern Alps glaciers and formed coalescent alluvial megafans (Fontana et al., submitted for publication). In the Venetian low plain, thick sequences of silt and sand layers alternated with frequent thin layers of peat and organic silt. Radiocarbon dates of these organic layers in the topmost 30 m of the Late Pleistocene alluvial series span 23,000 to 14,000 yr BP (Bortolami et al., 1977; Bondesan et al., 2002; Magri and Bondesan, 2004). In the last fifty years many authors reported pollen data from LGM organic layers taken from the Venetian low plain (detailed references in Miola et al., 2003; Mozzi et al., 2003) and they argued that this area was covered by a steppe vegetation, mostly consisting of Poaceae, Artemisia, Juniperus, Ephedra, Chenopodiaceae, Caryophyllaceae, Asteraceae Asteroideae, Apiaceae and rare trees such as Pinus and Betula. This is in accordance with the pollen- based reconstruction for southern Europe of Elenga et al. (2000). Frequent and extensive layers of organic sediments, however, indicate wide areas of wet environments (Fig. 1). Extensive layers of peat occurred also in the western Po Plain, south of Torino (Tropeano and Cerchio, 1984). The authors dated them to Würm 3 by stratigraphic correlation and pollen analysis. This suggests that wet environments developed probably in the entire low Po Plain. No attempts have been made so far to ascertain the origin and features of these environments. Fossil records from local plant communities are not reported, except for Cyperaceae and a few aquatic taxa, and no attempts have been made to identify the other components of local communities. New analyses of macrofossils and NPPs on Po Plain deposits could allow a reinterpretation of pollen analysis. The Po Plain could be an example in southern Europe where LGM steppe vegetation did not develop, making it an interesting case study of LGM climatic instability. Our research aims at understanding the features and the origin of the wet environments through sedimentological analysis, pollen analysis, non-pollen palynomorph (NPP) and plant macrofossil analyses on organic sediments obtained from cores drilled in the eastern Po Plain where the LGM organic layers have been reported. 2. The geomorphological framework The sampling sites are presently located on the North Adriatic coastal plain (Fig. 2). During the LGM, the glacio-eustatic sea level drop of about 120 m shifted the North Adriatic coastline more than 200 km to the south (for a recent review on this topic see Vigliotti, 2004). The North Adriatic shelf was, thus, a fluvial plain and the study area had continental conditions. The Piave valley glacier reached the plain near Vittorio Veneto (Venzo, 1977; Bondesan, 1999), 40 km north of the study area. In the piedmont sector, the Piave fluvio-glacial megafan (Nervesa megafan), as all the other megafans of the Po Plain, was mainly gravel. Within a distance of about 15–20 km from the Alpine foothills there was a drastic change in the grain size of the deposits, related to the decrease of the river transport capacity. Downstream, the Nervesa megafan consisted of low, sandy alluvial ridges separated by extensive silty-clay flood basins. The ridges were 1–3 m above the surrounding plain, up to several hundred metres wide and several kilometres long. Fluvial channels had low sinuosity and channel migration took place by avulsion rather than point-bar lateral shift. Recurrent channel migration through avulsive events during the LGM is A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 55 Fig. 1. Venetian plain sites where peat layers dated from 22,500 to 16,000 yr BP have been found. References are hereafter indicated. 1) Favaretto and Sostizzo, unpublished; 2) Mozzi et al. (2003); 3) Paganelli (1996); 4) Paganelli et al. (1988); Paganelli (1996); 5) Castiglioni et al. (1987); 6) Calderoni et al. (1996); 7) Iliceto et al. (2001); 8), 9) Bortolami et al. (1977); 10) Bertolani Marchetti (1967); 11) Mullenders et al. (1996); 12) Serandrei Barbero et al. (2001); 13), 14) and 15) Bortolami et al. (1977); 16), 17) and 18) this work; 19) and 20) Giovannelli et al. (1986); 21) Marocco (1989); 22) Mozzi et al. (2003); 23) Buurman (1969–1970); 24) this work. documented in the megafans of the Venetian Po Plain, apparently with 10–103 years cyclicity (Bondesan et al., 2002; Mozzi and Bondesan, 2004; Fontana et al., 2004). LGM sedimentation rates in the alluvial plain at the distal reaches of the megafan were high, in the order of 1–3 mm per year as an average (Bondesan et al., 2002; Mozzi and Bondesan, 2004). Peat formation took place in fens located in poorly drained depressions in the alluvial plain; because of the high aggradation rates, the fens were probably active for just a few centuries before being buried by alluvial sediments (Bondesan et al., 2002; Fontana et al., 2004; Mozzi and Bondesan, 2004). 3. Materials Ten medium-deep (10–20 m) cores were drilled in the Ca' Tron Estate (Roncade–Treviso), by the inner border of the lagoon of Venice. Another three 20-mdeep cores were drilled along a 20-km-long N–S transect from San Donà di Piave (Venice) to the coast of the Adriatic Sea. The locations of the coring sites, Ca' Tron, Fiorentina, Palazzetto and Ca' Fornera are shown on Figs. 1 and 2. The analysed samples come from organic layers less than 20–30 cm thick. The stratigraphic logs of Fig. 3 show the organic layers to be mostly embedded within alluvial sequences which are mostly composed by overbank fines, with frequent, scarcely interconnected, usually 1–2-m-thick sandy channel bodies. The organic layers consist of peat and organic silt. In the cross section of Ca' Tron (Fig. 4) some of these organic layers extend laterally for more than 1 km. The cores (∅ = 9 cm) allowed the extraction of limited amounts of material. The difference between “peat” and “organic silt” was based on the sedimentological 56 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Fig. 2. Location of the coring sites in the geomorphological framework of the Central Veneto plain (modified from Bondesan et al., 2002). 1. Nervesa megafan (Late Pleistocene, Holocene); 2. Montebelluna megafan (Late Pleistocene); 3. Bassano megafan (Late Pleistocene); 4. Brenta alluvial plain (Holocene); 5. Monticano, Cervada and Meschio fans (Holocene); 6. Livenza alluvial plain (Holocene); 7. Sile, Dese, Zero alluvial plain (Holocene); 8. Musone alluvial plain (Holocene); 9. Piedmont fans (Holocene); 10. Littoral sandy deposits (Holocene); 11. Moraines (Late Pleistocene); 12. Fluvial erosive scarps; 13. Hills and mountains; 14. Natural (a) and artificial (b) hydrography; 15. Core location. field description of the cores; no geochemical analyses have been carried out in order to quantify the exact percentage of the organic component. The parameters observed were: (1) the degree of darkness, classified with the Munsell Soil Colour Charts system; (2) the degree of stratification; (3) the amount of fragments of ligneous/herbaceous plants; (4) the amount of rootlets; (5) the amount and nature of the minerogenic compo- nents (Aaby and Berglund, 1986). Peat layers have very dark grey to black colours (value b 2, chroma b 2) and are usually massive with fragments larger than 2 mm. Field estimates indicate that the volume percentage of organic matter is more than 50% in volume. Organic silts have dark grey colours (value 4–5, chroma b 2) and are massive or laminated, rarely with fragments. The percentage of minerogenic material is variable, A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Fig. 3. Lithology of the cores. Bars indicate the pollen samples and stars indicate the radiocarbon dated levels. 57 58 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Fig. 4. Geological cross section at Ca' Tron, in the distal Nervesa megafan. 14C dates (yr BP): 1) 17,530 ± 120 (Beta-173736); 2) 3650 ± 40 (Beta173729); 3) 20,300 ± 220 (Beta-173730); 4); 2040 ± 40 (Beta-191293); 5) 2460 ± 70 (Beta-170847); 6) 16,190 ± 50 (Beta-170848); 7) 17,920 ± 130 (Beta-173733); 8) 19,770 ± 140 (Beta-173734); 9) 21,150 ± 190 (Beta-169480); 10) 20,970 ± 140 (Beta-169481) (Modified from Mozzi and Bondesan, 2004). depending on the amount of sediment carried into the fen by floods. Some samples have been taken from the silt layers between organic levels in all the cores, but their pollen content was very low. Twenty-one samples of peat, organic silt or clay and plant material were taken from the cores and submitted to the Beta Analytic Inc. Laboratory (Miami, Florida, USA) for conventional and AMS radiocarbon dating (Table 1). The radiocarbon dates and litho-stratigraphic correlation are discussed in Magri and Bondesan (2004) and Bondesan et al. (2004). The layers analysed in this work are dated to the LGM period, between 22,000 and 16,000 yr BP. Six samples have Holocene ages, therefore their stratigraphy and micro and macrofossil analysis are not discussed in this paper. Nevertheless, these 14C datings are indicated in the stratigraphic logs (Fig. 3) and in the cross section of Ca' Tron (Fig. 4), in order to document the Holocene segments of the cores. 4. Methods 4.1. Macrofossil analysis All organic samples from Fiorentina (27 samples), Palazzetto (8 samples) and Ca' Fornera (6 samples) were analysed for macrofossils (sensu Birks and Birks, 1980). Pollen analyses were done on the same layers. Each whole sample (about 60 cm3) was searched for larger macrofossils with the naked eye. To detect smaller macrofossils about 2 cm3 of sediment from each sample was boiled mildly in a 10% NaOH solution for a few minutes and rinsed with water using A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Table 1 Radiocarbon conventional dates in Fiorentina (FI), Palazzetto (PA), Ca' Fornera (CA) and Ca' Tron (S1–S13) cores (Magri and Bondesan, 2004; Bondesan et al., 2004) Lab. reference Core-sample depth (cm) Beta-170844 Beta-170845 Beta-170846 Beta-168127 Beta-168128 Beta-168129 Beta-173727 Beta-173729 Beta-173730 Beta-191293 Beta-170847 Beta-170848 Beta-173731 Beta-173732 Beta-173733 Beta-173734 Beta-173735 Beta-169479 Beta-169480 Beta-169481 Beta-173736 14 C age (yr BP) Method Materials FI_99–111 FI_1155–1164 FI_1740–1755 3570 ± 120 Standard Peat 18,640 ± 100 Standard Peat 20,930 ± 130 Standard Organic silt PA_581–609 6520 ± 50 Standard Organic clay PA_1323–1333 19,850 ± 120 Standard Peat PA_2049–2059 21,250 ± 150 Standard Organic silt CA_1096–1098 8710 ± 60 AMS Organic silt S1_507–518 3650 ± 40 Standard Organic silt S1_1433–1438 20,300 ± 220 Standard Organic silt S5_164–166 2040 ± 40 AMS Quercus leaves S5_234–238 2460 ± 70 Standard Organic silt S5_267–270 16,190 ± 50 AMS Organic silt S8_718–728 18,820 ± 110 Standard Peat S8_1940–1950 21,470 ± 200 Standard Peat S10_400–415 17,930 ± 120 Standard Peat S10_780–800 19,660 ± 150 Standard Peat S12_418–424 17,920 ± 130 Standard Peat S12_956–971 19,770 ± 140 Standard Peat S12_1294–1300 21,150 ± 190 Standard Peat S12_1610–1622 20,970 ± 140 Standard Peat S13_411–417 17,530 ± 120 Standard Peat a nylon sieve with 200-μm mesh. The residues were suspended in water in a Petri dish and analysed under a stereomicroscope at a magnification of ×10–40. The smallest fragments were observed under a compound light microscope at ×250. The recorded quantities of the fossils are estimations of their abundance in the Petri dish (absent, rare, common, abundant). The identification of roots, leaves, rhizomes and seeds was done using the reference collection of the pollen and macrofossil laboratory at the Department of Biology. The bryophytes have been identified with the keys in Watson (1968) and Nyholm (1979) and using our reference collection. The Potamogeton remains were identified by means of Kuhry (1988), the trichosclereids of Nymphaeaceae (Type 129) by means of Pals et al. (1980), and the Bryozoan statoblasts (Type 390), the mandibles of invertebrates (Type 88), the exoskeletons of mites (Type 36A — Hydrozetes lacustris) and the Characeae oospores (Type 384) using van 59 Geel (1978) and van Geel et al. (1981). The identified taxa are listed in Table 2. 4.2. Microfossil analysis Samples for pollen and NPP analyses were prepared according to Faegri and Iversen (1989), including HCl 10%, hot NaOH 10%, deflocculation, sieving (∅ = 200 μm), cold HF 50% and acetolysis. Pollen concentration (grains/cm3) was estimated by adding Lycopodium tablets to a measured volume of dry sediment. Pollen grains were identified using a magnification of ×400 or ×1000. Pollen nomenclature follows Moore et al. (1991). NPPs were identified using descriptions and photographs of NPP literature hereafter indicated. The identified taxa are listed in Tables 3 and 4. The counting was carried out with a light microscope at a magnification of ×400. Seventynine samples have been analysed (19 from Fiorentina, 13 from Palazzetto, 27 from Ca' Fornera and 20 from the Ca' Tron cores), but only 55 had enough pollen for counting and representation in the pollen diagrams. Herbaceous pollen always exceeds 50% of the total terrestrial plant pollen and the mean count was 190 terrestrial pollen grains. The preservation of pollen grains was generally modest (10–20% of undeterminable grains, mainly broken saccate); poor preservation of the grains was recorded in some silt samples in the Fiorentina, Palazzetto and Ca' Tron S05 cores (30–60% undeterminable grains, mainly broken saccate). The choice of the taxa to include in the pollen sum was determined by the problem to be investigated. In the first part of the analysis the pollen sum (ΣP) has included all pollen, which could have originated from the terrestrial vegetation (trees, shrubs and terrestrial herbs, including Poaceae), with the aim of comparing our percentage data with similar data in literature. Ferns, helophytes, hydrophytes, undeterminable pollen grains and NPP (algae, fungi, mosses, invertebrates, unknown origin types) percentage data were calculated using the ΣP plus the total count of each group. These percentage data have been used for the construction of pollen diagrams (Figs. 5–8) and for the first Principal Components Analysis (Fig. 10A). Our main task, however, was to understand the origin and features of local environment. To this end, new percentage data were calculated with respect to a different pollen sum (ΣLP) which included all pollen of herbaceous plants confined to the lowland local aquatic and mire vegetation (e.g. obligate aquatic plants and helophytes) and a few grains of herbaceous plants 60 Samples Fiorentina 1888–1889 1886–1888 1883–1885 1880–1882 1877–1879 1873–1875 1871–1873 1868–1870 1866–1868 1863–1865 1756–1758 1753–1755 1750–1752 1748–1750 1745–1747 1743–1745 1741–1743 1738–1740 1458–1469 1241–1243 1234–1236 1200–1202 1155–1164 Lithology Clay–silt # Clay–silt Clay–silt Clay–silt Clay–silt Clay–silt Clay–silt Clay–silt Clay–silt Clay–silt Sand # Clay–silt # Clay–silt Clay–silt Clay–silt Clay–silt Clay–silt Clay–silt # Clay–silt Peat Peat Silt # Peat # Bryophyta remains Herbaceous roots Leaf remains and Carex fusca seeds Rhizome remains (S) Other remains S. s. Papillate C. t. D. e. Leaf remains roots Smooth roots Phragmites Sclereids of Mites Type Cyperaceae Poaceae Potamogeton Small Nymphaea 88 rhizomes rhizomes leaves leaves leaves Oospores of Bryozoa Characeae / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / + + + + + / + / + / + / + + + + / + / + + + + / +S +++S +++ +++ +++S +++ +++S +++S +++ / / /S / / / / / +++ + / / / / / / / / / / / / / +++ ++ + + + / + / / / / / / / / / / / / / / / / / / / / / / / / / / / / / + + + + ++ / / / + + + + ++ ++ ++ ++ ++ ++ / ++ ++ + + +++ +++ +++ +++ +++ +++ +++ ++ +++ +++ ++ + +++ + + + + + +++ + ++ / + / + / / / + / / +++ +++ + ++ ++ ++ ++ ++ ++ ++ / + ++ + + / / / + / / / / / / / / / / / / / / / / / / + / / + + + + + +++ +++ +++ / + / / / + / + / / / / + / / / / / / / / +++ ++ / + ++ ++ ++ ++ +++ ++ / + ++ / / / / / / / / / / + / / / + / + + + / / / / / +++ / / + / / / / + / / / / + + / / / / / / + / / / / / / / + / / / / / / + / / / + + / / / / / / / / / / / / / / / / / + + / / / + / / / / / A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Table 2 Results of macrofossil analysis. Silt # Silt Silt Silt / / / / / / / / / / / / + / + ++ + +++ + ++ / + / + / +++ +++ +++ + / + / / / / / / + + + / / / / + / / / / / / / + / / / / / + / / / / / Clay Clay Clay Sand Sand Silt # Sand Sand # / / / / / / / / / / / / / / / / / / / / / / / / +++ + / + + / + + + + / + + / + / + + / + + / + + /S /S /S /S /S / / / + + + + / + / / / / / / / / + / + + + + / + + + / / / / / / / / / / + / / +++ / / / / + / + + / / / / + / / + / / / / / + / +++ / / / / + + / / / / Peaty silt Peaty silt Peaty silt Peaty silt Peaty silt Peaty silt +++ +++ +++ +++ +++ +++ + + + + + + + / / / / / / / / / / / + ++ +++ + / / ++ ++ ++ +++ + + / + +S + + + / / / / / / / / / / / / / / / + / / / / / / / / / / / / / / / + / + + / / / / + / / / / / / + + / / / / / + / absent, + rare, ++ common, +++ abundant; # sample poor in macrofossils; S.s.: Scorpidium scorpioides, C. t.: Calliergon trifarium, D. e.: Drepanocladus exannulatus. A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 1136–1138 1122–1124 1120–1122 1114–1116 Palazzetto 2042–2049 2040–2042 2034–2036 1834–1849 1819–1834 1316–1318 1155–1157 1152–1154 Ca' Fornera 1989–2001 1978–1984 1967–1978 1956–1967 1879–1880 1877–1878 61 62 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Table 3 Minimum and maximum relative frequency of pollen types recorded in all the cores; percentage of samples where the type is present on the total number of fertile samples (Frequency); percentage of samples where the relative pollen frequency is higher than 50% (% of Fr. N 50%) Groups Taxa Trees and shrubs Pinus undiff. Betula Picea Salix Alnus Larix type Ephedra fragilis type Juniperus type Hippophae rhamnoides Abies + Fagus + thermophilous trees and shrubs Populus Abies Corylus Fagus sylvatica type Ulmus Castanea sativa type, Cornus suecica type, Sambucus nigra type, Quercus robur group, non-operculate Rosaceae Terrestrial herbs Poaceae Artemisia Chenopodiaceae Asteraceae undiff. Apiaceae Galium type Caryophyllaceae undiff. Plantaginaceae Scrophulariaceae Saxifragaceae Thalictrum Helianthemum Anthemis type Ranunculus type Urtica pilulifera Ericaceae Pinguicula Saxifraga stellaris type Other terrestrial herbs Hydrophytes Potamogeton Menyanthes trifoliata Myriophyllum verticillatum Callitriche Lemna Nymphaea alba type Nuphar luteum Myriophyllum spicatum Hydrocharis morsus-ranae Helophytes Cyperaceae undiff Sparganium erectum Typha angustifolia type Minimum Maximum Frequency % of Fr. N 50% 4.7 0.3 0.3 0.3 0.3 0.3 0.3 1.1 0.7 51.6 7.1 8.8 6.1 1.5 3.6 3.9 12.2 1.3 100% 78% 47% 40% 25% 22% 22% 13% 9% 1.8% – – – – – – – – 0.4 0.5 0.4 0.4 0.3 b1 22.1 5.9 1.0 1.8 2.0 b2 29% 15% 13% 7% 7% b4% 16.7 0.6 0.3 0.3 0.3 0.3 0.4 0.3 0.3 0.3 0.3 0.3 0.3 0.5 0.3 0.4 0.5 0.4 90.3 35.2 6.9 34.3 2.0 2.8 5.2 2.6 4.1 3.1 1.8 1.7 1.7 1.8 0.9 1.8 0.8 1.3 100% 100% 75% 64% 55% 51% 40% 38% 35% 27% 18% 16% 7% 7% 7% 5% 5% 5% b2 b4 b4% 0.5 0.3 0.3 0.6 1.0 0.4 0.5 0.2 10.2 52.9 3.6 3.4 7.5 14.5 3.0 0.9 1.2 10.2 84% 40% 20% 16% 16% 9% 7% 4% 2% 1.8% – – – – – – – – 5.0 0.2 0.1 72.6 7.6 12.7 100% 38% 29% 40.0% – – – – – – – – 74.5% – – – – – – – – – – – – – – – – – 63 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Table 3 (continued) Groups Taxa Helophytes Typha latifolia type Caltha type Iris pseudacorus type Minimum Maximum 0.1 0.3 0.3 Frequency 2.6 1.9 0.3 13% 4% 2% % of Fr. N 50% – – – Other terrestrial herbs: Allium type, Astrantia major type, Centaurea scabiosa, Gentiana, Hornungia type, Illecebrum verticillatum, Lamiaceae, Liliaceae, Papaver radicatum type, Primula veris type, Ranunculus arvensis, Rhinanthus type, Saxifraga granulata type, Stachys sylvatica type, Urtica dioica, Viola arvensis type. which could live near the wet environments. Poaceae, which are usually considered among the terrestrial herbs, have been included in the ΣLP in agreement with the results of macrofossil analysis. These data have been used in the second Principal Component Analysis (Fig. 10B). Table 4 Minimum and maximum relative frequency of NPP types recorded in all the cores; percentage of samples where the type is present on the total number of fertile samples (Frequency); percentage of samples where the relative pollen frequency is higher than 50% (% of Fr. N 50%) Groups Taxa Algae Zygnema type Type 314 Spirogyra Type 315 Botryococcus Type 901 Spirogyra cf. scrobiculata Type 342 Mougeotia Type 313 Gloeotrichia Type 146 Pediastrum cf. boryanum Type 900 Type 225 Type 229 Tetraedron cf. minimum Type 371 Zygnemataceae Type 58 Other algae Fungi Type 200 Gaeumannomyces Type 126 Glomus cf. fasciculatum Type 207 Type 351 Type 79 Fungal spores (Pl. III, 24) Type 140 Other fungi Mosses Type 340 Invertebrates Unknowns Min Max Frequency % of Fr. N 50% 0.4 0.4 0.6 0.3 0.3 0.5 0.5 2.8 0.3 0.7 0.4 b5 15.8 7.3 6.3 2.3 2.2 8.3 1.5 5.7 2.6 29.5 4.7 b5 40% 36% 24% 22% 16% 9% 7% 7% 5% 4% 4% b2% – – – – – – – – – – – – 0.5 0.3 0.3 0.1 23.4 0.1 0.3 b20 76.4 41.8 1.3 1.1 51.0 4.5 1.8 b20 58% 56% 13% 9% 9% 7% 5% b4% 1.8% – – – 1.8% – – – 0.5 22.7 24% – Rhabdocoela Type 353B Rhabdocoela Type 353A Eurycercus cf. lamell. Type 72D Type 52 Other invertebrates 0.3 0.3 0.4 0.3 b1 4.4 2.0 3.1 8.4 b1 44% 33% 18% 9% b4% – – – – – Type 303 Type 74 Type 128A Unknown type (Pl.III, 15) Type 128B Type 414 Type 91 Other unknowns 1.1 0.5 0.7 0.3 0.3 0.1 0.4 b2 65.0 15.6 12.3 14.7 13.6 1.0 5.0 b5 91% 73% 58% 55% 33% 11% 9% b9% 5.5% – – – – – – – Other algae: Ceratium hirundinella, Closterium idiosporum Type 60, Pediastrum cf. simplex, Type 225B, Type 230, Type 304, Type 333; other fungi: Type 140, Type 20, Enthorrhiza Type 527, Sordaria Type 55A, Type 173A and B, Type 227, Type 231, Type 324, Type 3A, Type 55C, Type 77A; other invertebrates: Rhabdocoela Type 353D, Type 75; other unknowns: Type 179, Type 187D, Type 180C, Type 187C, Type 219, Type 719B. A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Fig. 5. Pollen and NPP diagram of Fiorentina (Venice) core. 64 65 Fig. 6. Pollen and NPP diagram of Palazzetto (Venice) core. A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Fig. 7. Pollen and NPP diagram of Ca' Fornera (Venice) core. 66 67 Fig. 8. Pollen and NPP diagram of Ca' Tron (Treviso) core. A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 68 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Table 5 Groups of taxa used in the principal components analysis (Fig. 10A and B) Groups for PCA analysis Trees and shrubs Pinus Other trees and shrubs Abies + Fagus + thermophilous trees and shrubs Terrestrial herbs Chamaephytes Asteraceae Chenopodiaceae Caryophyllaceae Ericaceae Poaceae Other terrestrial herbs Helophytes Cyperaceae Helophytes Hydrophytes Hydrophytes Taxa I PCA Pinus mugo/P. sylvestris, Pinus undiff. Betula, Picea, Hippophae rhamnoides Juniperus type, Larix type, Alnus glutinosa/incana, Salix Abies, Fagus sylvatica type, Castanea sativa type, Morus nigra, Populus, Quercus robur group, Ulmus, Cornus suecica type, Corylus, non-operculate Rosaceae, Sambucus nigra type ⁎ ⁎ Artemisia, Helianthemum, Ephedra fragilis type Aster type, Asteraceae Asteroideae, Asteraceae Cichorioideae, Cichorium intibus type, Lactuca sativa type, Scorzonera type, Tragopogon pratensis type Caryophyllaceae undiff., Caryophyllaceae cf Hernaria alpina Calluna vulgaris, Ericaceae undiff., Vaccinium Allium type, Anthemis type, Apiaceae, Astrantia major type, Centaurea scabiosa, Galium type, Gentiana, Hornungia type, Illecebrum verticillatum, Lamiaceae, Liliaceae, Papaver radicatum type, Pinguicula, Plantaginaceae, Primula veris type, Ranunculus arvensis, Ranunculus type, Rhinanthus type, Saxifraga granulata type, Saxifraga stellaris type, Saxifragaceae, Scrophulariaceae, Stachys sylvatica type, Thalictrum, Urtica dioica, Urtica pilulifera, Viola arvensis type ⁎ ⁎ ⁎ ⁎ ⁎ ⁎ ⁎ ⁎ Carex type, Cyperaceae undiff. Caltha type, Filipendula ulmaria, Iris pseudacorus type, Potentilla, Sparganium erectum, Typha angustifolia type, Typha latifolia type, Valeriana officinalis type ⁎ ⁎ Callitriche, Hydrocharis morsus-ranae, Lemna, Menyanthes trifoliata, Myriophyllum spicatum, Myriophyllum verticillatum, Nuphar luteum, Nymphaea alba type, Potamogeton subg. Coleogeton type, Potamogeton subg. Potamogeton type ⁎ 4.3. Presentation of the data Pollen and NPPs of all the analysed samples are listed in Tables 3 and 4, together with the range of their percentage values and the frequency of their occurrence in the spectra, in order to underline the principal components of the spectra from all the cores. In order to provide a synthetic and low-dimensional presentation of the pollen spectra and to detect possible similarities among the samples, principal components analysis (PCA) of the different groups of pollen was carried Plate I. 1. 2. 3. 4. 5,6. 7. 8. II PCA Fossil root (1000×). Fossil root (200×). Root fragment of Carex fusca All. (200×). Root fragment of Carex fusca All. (400×). Fossil unidentified small rhizome (100×). Fossil rhizome of Phragmites sp. (1×). Fossil rhizome of Phragmites sp. (2×). out on the set of the pollen spectra from all the cores. Since the large number of taxa could produce unacceptable statistical results, we decided to aggregate the most frequent taxa of the same taxonomical group and/or the taxa which now live under similar environmental conditions (Table 5). Percentage data have been used. Pollen spectra with a sum of less than 200 have been excluded. Pollen data of the principal cores (Fiorentina, Palazzetto, Ca' Fornera, Ca' Tron S05) are also presented in pollen diagrams (Figs. 5–8). A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 69 70 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 5. Results and analyses 5.1. Macrofossils of the local communities The results are presented in Table 2. The most common macrofossils are herbaceous roots, fragments of rhizomes, monocotyledonous leaves and moss remains. Two different types of roots are common in the analysed samples. The first type has characteristic papillae on the epidermis (Plate I, 1–2). Isolated papillae are also very common microfossils in the pollen samples where the roots are abundant. In order to identify the producers of these roots we examined many species of helophytes (Carex elata All., C. fusca All., C. pendula Hudson, C. rostrata Stokes, C. riparia Curtis, C. davalliana Sm., C. distans L., C. lepidocarpa Tausch, C. diandra Schrank, C. dioica L., C. paniculata L., Eriophorum angustifolium Honckeny, Trichophorum caespitosum (L.) Hartm., T. alpinum (L.) Pers., Sparganium erectum L., Typha latifolia L., Phragmites australis (Cav.) Trin.) and hydrophytes (Menyanthes trifoliata L.), which are common today in fens and small ponds of northeastern Italy (A.R.P.A.V., 2001). Among these species, only Carex fusca has roots with papillae with the same morphology as the fossil ones (Plate I, 3–4). Seeds of Carex fusca have been found in some samples of the Fiorentina, Palazzetto and Ca' Fornera cores, where papillae were also abundant. Therefore it is likely that the herbaceous roots of the first type have been produced by C. fusca plants, even in the samples containing only root remains with papillae. Carex fusca is a typical spring-water species and lives in rich fens (Oberdorfer, 1949). The occurrence of Geumannomyces hyphopodia (see Sec. 5.2) confirms the presence of local stands of sedges (van Geel et al., 1989). The second type of root remains is present in smooth fragments (less than 1 mm thick and of variable length) with radical apex. Their morphological characteristics are very common in herbaceous roots of monocotyledon plants (e.g. Phragmites australis, Carex sp. div. other than C. fusca). Two different types of rhizome fragments are present. The most common are small fragments (0.05–0.2 cm thick and less than 0.5 cm long) characterized by abundant parenchyma and few sclerenchymatic cells (Plate I, 5–6). They present longitudinal dark bands and many rootlets arranged in dense whorls. We suppose that they are rhizomes of aquatic plants such as Nymphaea sp. and/or Potamogeton sp. The second type of rhizome fragments matches the morphology of Phragmites australis rhizomes (Plate I, 7–8). In the samples Fi_1866– 1868 and Fi_1863–1865 they are 1–1.5 cm thick and 2–10 cm long. In other samples from the Fiorentina core (1738–1758 cm below the ground surface), there are smaller and more fragmentary remains with the same features as the previous ones. The leaf remains feature rectangular cells with undulate walls. The veins are all parallel. They are very similar to Carex sp. leaves (Plate II, 9). It is likely that the leaves are of C. fusca, because of the presence of its papillate root remains and/or seeds in the same samples. Some leaf fragments in the Palazzetto and in the Fiorentina cores have the typical silical trichomes of the Phragmites australis leaves and its circular attachments on the epidermis (Plate II, 10–11). A few leaf fragments of Potamogeton have been recognized in the Palazzetto and Fiorentina cores. Moss remains are very common in the samples taken from the Ca' Fornera core. They have been identified as leaflets, and stems with leaflets, of Scorpidium scorpioides Hedw. (Plate II, 12–14). Scarce leaflets of Calliergon trifarium Web. et Mohr. and Drepanocladus exannulatus B., S. and G. are also present. In the Palazzetto and Fiorentina cores moss remains are also present, but are less abundant and too small for identification. Also sporadically present are sclereids of Nymphaea, oospores of Characeae, mites, mandibles of invertebrates and Bryozoa statoblasts (cf. Plumatella repens L.). In four silt or sand samples of sediments with a very poor organic content from the Fiorentina core (1155–64 cm, 1753– 56 cm, 1756–58 cm) and from the Palazzetto core (1316– 1318 cm), sclereids of Nymphaea and/or oospores of Characeae are very abundant. In these samples root and rhizome fragments are rare. Fossils of herbaceous monocotyledon plants, such as Carex fusca and Phragmites australis, are a common feature of the analysed samples. Therefore we suppose that in the low Venetian plain Cyperaceae and Poaceae were the most important components of wet communities, with sedges on soils periodically inundated and reeds on soils always inundated. The brown mosses were probably common in the moss layer of these communities; perhaps they also formed almost pure communities that were submerged in shallow waters. Scorpidium scorpioides, Drepanocladus exannulatus and Calliergon trifarium usually live in mesotrophic or eutrophic sedge or reed swamp. S. scorpioides, a calciphile, is an indicator species of rich fens (Watson, 1968). Fossils of Nymphaea, oospores of Characeae, statoblasts of Bryozoa and Potamogeton indicate open water environments. The three sites are well differentiated with regards to macrofossils. The deepest Fiorentina samples have abundant Cyperaceae remains (1889–1868 cm) and are followed by samples with both Poaceae and Cyperaceae remains (1868–1136 cm). In the uppermost layers A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Plate II. 9. 10. 11. 12. 13. 14. Fossil leaf fragment of Carex (1000×). Leaf fragment of Phragmites australis (900×). Fossil leaf fragment (900×). Fossil leaves of Scorpidium scorpioides (100×). Fossil leaves of Scorpidium scorpioides, apex (200×). Fossil leaves of Scorpidium scorpioides, central portion (200×). 71 72 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Cyperaceae are dominant again. Small fragments of mosses are usually present, but never abundant. In the samples of Palazzetto, macrofossils are generally scarce. In a silt layer (1318–1316 cm) there are abundant remains of obligate aquatic organisms (Nymphaea and Characeae). Finally in the Ca' Fornera core moss remains are dominant. These observations suggest that the local plant communities changed in space and time. No records have been detected that suggest the development of raised bogs (e.g. Sphagnum and Ericaceae) or colonization by trees. 5.2. Pollen and non-pollen palynomorphs We present now the results of pollen data in order to correlate the cores and to compare them with pollen data from the literature. In Tables 3 and 4 pollen and NPP types of all the sequences are listed and the most frequent types are shown. The data of the Fiorentina, Palazzetto, Ca' Fornera and Ca' Tron S05 cores are also represented by pollen diagrams (Figs. 5–8). A common feature is the dominance of herbs: pollen of Poaceae is always present and it is the dominant pollen type. The average diameter of Poaceae pollen grains has been measured in pollen samples with high percentage of Poaceae pollen and abundant macrofossils of Phragmites australis from the Fiorentina core (50 grains for each sample). More than 50% of the grains has an average diameter of 32 μm and 24% of 36 μm. The others have an average diameter of 39 μm. The annulus diameter is always smaller than 10 μm. It seems likely that almost all smaller grains can be attributed to P. australis (average diameter 33 μm in glycerol or 22–24 μm in silicon oil, anulus smaller than 10 μm in glycerol or smaller than 8 μm in silicon oil, according to Andersen (1978) and Erdtman et al. (1961)). The following herbaceous pollen types have been recorded in almost all the samples, at least with a few grains: Artemisia, Chenopodiaceae and Asteraceae. Common but less frequent are Caryophyllaceae, Apiaceae, Galium type, Helianthemum, Plantaginaceae undiff., Scrophulariaceae and Thalictrum. Sporadically present are Anthemis type, Ranunculus type, Urtica pilulifera, Fig. 9. Total pollen concentration (trees, shrubs, terrestrial and aquatic herbs) and total NPP concentration (mosses, algae, fungi, invertebrates, incompletely known types) as grains/cm3 of dry sediment in Fiorentina, Palazzetto, Ca' Fornera and Ca' Tron S05 cores. A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Pinguicula, Saxifraga stellaris type, Allium type, Astrantia major type, Centaurea scabiosa, Gentiana, Hornungia type, Illecebrum verticillatum, Lamiaceae, Liliaceae, Papaver radicatum type, Primula veris type, Ranunculus arvensis, Rhinanthus type, Saxifraga granulata type, Stachys sylvatica type, Urtica dioica and Viola arvensis type. Pinus and Betula are the most common trees. Sporadically present are Picea, Larix and the shrubs Salix, Juniperus and Ephedra. Pinus is always present but its percentage values are generally low (average 17%). It slightly increases in the upper levels of the Fiorentina (Fig. 5) and Ca' Tron S05 (Fig. 8) cores. Abies, Fagus and thermophilous broadleaf trees and shrubs are rare. The total pollen concentration (Fig. 9) is generally low (average 42,000 grains/cm3, range 200–140,000 grains/cm3). In order to represent all the pollen spectra in a reduced data dimension, and to detect possible similarities among the spectra, the PCA has been applied to the principal terrestrial pollen groups (Table 5) and the pollen spectra have been positioned on the first two principal component axes (Fig. 10A). The pollen spectra are mainly concentrated around the origin of the axes, suggesting more similarities Fig. 10. PCA for pollen groups as indicated in Table 5. The proportion of overall variance accounted for the first two principal components is equal to 51.3% (A) and to 47.7% (B). 73 than differences among the spectra and probably a good biostratigraphic correlation among them. These pollen results are characteristic of a cold and dry open steppe environment, in accordance with pollen analyses of LGM sediments in the northeastern Italy plain carried out by Bertolani Marchetti (1966–67), Buurman (1969–70), Bortolami et al. (1977), Pellegrini et al. (1984), Giovannelli et al. (1986), Calderoni et al. (1996), Mullenders et al. (1996), Paganelli (1996), and Mozzi et al. (2003). The pollen spectrum at the bottom of Ca' Fornera core (CA_2027) is anomalous. It could have originated during a previous more temperate and humid period, supported by the relative high presence of Picea and Abies pollen and by the presence of some grains produced by broadleaf trees and shrubs which are very scarce in the other spectra, but common in pollen spectra of the period that immediately precedes the beginning of LGM period (Bortolami et al., 1977). Some doubts arise about the LGM palaeoenvironmental reconstructions made in the northeastern Italy plain from pollen analysts considering only upland pollen grains. If we consider the frequent occurrence of LGM peat layers and the results of macrofossil and pollen analyses about the local plant communities, we can conclude that wetlands were common over a large area, about 100 km by 30 km, from the eastern side of Berici Hills (Vicenza) and Lessini Mountains (Verona) to the Friulian plain (Fig. 1). They were probably waterlogged for most of the year judging from the formation of peat deposits. Therefore the local plant communities had to consist mainly of aquatic and telmatic species. The very low values of total terrestrial plant pollen concentration (average 20,000 grains/cm3) is probably due to limited cover of terrestrial plant communities, probably living in dry areas on the fluvial ridges, or to long-distance transport. Pollen analyses of surface samples in modern arctic–alpine open environments by Pardoe (2001) showed that pollen of wind-pollinated trees such as Betula, Pinus and Picea does not correlate with the cover of the trees. For example, even where Betula trees are absent, the pollen has a mean frequency of 32% of all the terrestrial plant pollen. Pinus normally produces more pollen than Betula, therefore we expect that its pollen too does not correlate with the presence of pine trees in open environments. It is therefore likely that pollen of wind pollinated trees and shrubs in the peat layers was due to long-distance transport instead of local production. Additionally, the wind pollinated herbs such as Cyperaceae and Poaceae do not show a strong relationship between the frequency of plants and pollen in modern arctic–alpine open environments (Pardoe, 2001). However in our research the frequent occurrence of 74 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Cyperaceae and Poaceae macrofossils (Phragmites australis) confirms their presence in the past local communities as indicated by the very high pollen percentage values of both taxa (higher than 50% of the ΣP + Total Helophytes sum in 40% of the samples for Cyperaceae and higher than 50% of the ΣP in 74.5% of the samples for Poaceae). We consider them therefore as the principal components of the wet plant communities. The other aquatic plants in the pollen spectra are represented by helophytes and hydrophytes, which required different hydrological conditions: the first (Sparganium erectum, Typha angustifolia, Typha latifolia, Caltha palustris, Iris) are rhizomatous herbs growing in wet places at the margins of rivers and lakes, with rhizomes and roots under the water level and the epigeous organs above the water level (Pignatti, 1982); the latter (Menyanthes trifoliata, Nymphaea alba, Nuphar, Myriophyllum verticillatum, M. spicatum, Callitriche, Potamogeton, Hydrocharis morsus-ranae) are obligate aquatic taxa which live partially or entirely submerged in fresh and shallow water. As with the two principal components of the pollen association (Poaceae and Cyperaceae), the aquatic taxa indicate wet environments characterized by different water levels. In order to investigate the local plant communities we recalculated the pollen data as percentage of all the herbaceous plant pollen. We included also the very few other terrestrial herbs present, such as Ericaceae, Caryophyllaceae, Ranunculaceae, Saxifraga and Asteraceae, because, according to Pardoe (2001), the occurrence of their pollen in open arctic–alpine environments indicates the local presence of the plants. By applying the PCA to the herbaceous pollen groups (Table 5), it is possible to arrange the samples from all the cores in a graphical representation with the first two principal components, as it is shown in Fig. 10B. The position of the samples is much dispersed in the plane: on the right hand side is a group of Fiorentina samples dominated by Poaceae (probably Phragmites australis, as shown by macrofossils and pollen analysis), on the left a group of Ca' Fornera samples with similar frequencies of Plate III. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. Unidentified microfossil with S-shaped furrow. Type 128A. Type 128B. Fossil Type 303. Actual Type 303. Type 74 (low focus). Type 74 (high focus). Type 340. Type 200. Unidentified fungal spores. 75 Poaceae and Cyperaceae, and a relative high percentage of hydrophytes, and on the top are the samples characterized by high percentages of Cyperaceae. None of these groups is well differentiated from the others and many samples intermingle. Also the samples of each core are dispersed in the plane. PCA of the herbaceous local pollen data suggests that different plant communities developed at each core site during the LGM period: reedswamps alternated with sedgefens and open water environments. NPPs represent a component of the local community that has never been investigated in the previous works in the research area. They are very abundant in the analysed samples (Table 4), sometimes their concentration in the samples is higher than the pollen concentration (average 28,000 microfossils/cm3, range 100–135,000 microfossils/cm3). The percentage data of each group of NPPs are presented in the pollen diagrams (Figs. 5–8). The most frequent types are palynomorphs of unknown origin: Type 303 (van Geel et al., 1981), Type 74 (van Geel, 1978), Type 128A and Type 128B (van Geel et al., 1983, 1989) and a new type illustrated in Plate III (15), similar to Type 128A and B (Plate III, 16–17) except for its smooth wall (see Appendix). They all generally occur when algal types and/or obligate aquatic plants are present. Type 303 (Plate III, 18–19) correlates well with the approximate abundance of S. scorpioides, and it probably was produced by aquatic organisms which live in the same environments as the brown moss (see Appendix). Type 74 (Plate III, 20–21) is not as abundant as Type 303, but it is commonly present when the latter is abundant. Type 128A is more common than Type 128B. Both types are possible algal palynomorphs (Bakker and van Smeerdijk, 1982; van Geel et al., 1989): the first has been mainly recorded in fossil assemblages characteristic of eu- to mesotrophic conditions, the latter in meso-oligotrophic conditions. The new type illustrated in Plate III (15) occurs mainly with Types 303 and 128A. These types of unknown origin have been recorded in sediments where many different types of algal remains have been 76 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Table 6 Preferred trophic conditions and water levels of some identified taxa Groups Taxa Preferred trophic condition Water level Ref. Hydrophytes Callitriche Hydrocharis morsus-ranae Lemna Menyanthes trifoliata Myriophyllum spicatum Myriophyllum verticillatum Nuphar luteum Nymphaea alba Potamogeton sp. Carex fusca Filipendula ulmaria Phragmites australis Eu- to oligotrophic Mesotrophic Eutrophic Partially submerged Partially submerged Partially submerged Partially submerged Submerged Submerged Partially submerged Partially submerged Submerged Damp soil, periodically inundated Damp soil, swamps Marshes, swamps, along streams, lakes, ponds, ditches idem idem idem Damp soils Open water Open water Shallow open water Open water Open water Temporary small and shallow ponds Open water Small lakes and ponds Open water and ephemeral pools Shallow open water Stagnant shallow water Shallow open water Shallow open water Open water Open water Open water 1 1 1 1 1 1 1 1 1 1, 7 1 1 Helophytes Algae Mosses Unknowns Sparganium erectum Typha angustifolia Typha latifolia Valeriana officinalis Botryococcus Type 901 Closterium idiosporum Type 60 Mougeotia Type 313 Pediastrum cf. boryanum Type 900 Pediastrum cf. simplex Spirogyra cf. scrobiculata Type 342 Spirogyra Type 315 Tetraedron cf. minimum Type 371 Type 225 (algal spore?) Zygnema type Type 314 Zygnemataceae Type 58 Scorpidium scorpioides Calliergon trifarium Type128 A Type 128 B Type 303 Eu- to mesotrophic, calcareous water Eutrophic, base-rich water Oligotrophic Oligotrophic Eu- to oligotrophic Eutrophic Eu- to oligotrophic Eu- to mesotrophic Mesotrophic Eu- to mesotrophic Eu- to mesotrophic Eutrophic Eu- to mesotrophic Eu- to mesotrophic Eu- to mesotrophic Mesotrophic Eutrophic, calciphile Eutrophic Eu- to mesotrophic Meso- to oligotrophic Eu- to mesotrophic 7 7 7 1 2 3 3 2 2 3 3 4 6 3 3 8 9 5, 6 6 3 1. Pignatti (1982); 2. Kuhry (1997); 3. van Geel et al. (1981); 4. Bakker and van Smeerdijk (1982); 5. Pals et al. (1980); 6. van Geel et al. (1989); 7. Polunin and Walters (1985); 8. Watson (1968); 9. Nyholm (1979). identified: resting cells of Zygnemataceae, Spirogyra, Mougeotia, Closterium idiosporum, Type 225 and Type 229 (van Geel et al., 1989), Type 417B (Kuhry, 1997), and free-floating cells or colonies of Botryococcus, Pediastrum cf. boryanum, P. cf. simplex, Ceratium hirundinella, Tetraedron cf. minimum and Type 333 (van Geel et al., 1981). The occurrence of algal remains, both freefloating and resting cells, suggests that open water environments frequently occurred and that they periodically dried out. Most of the identified algae have been recorded in fossil assemblages characteristic of eu-mesotrophic conditions (Table 6). The occurrence of Type 340 moss spores (Plate III, 22) corresponds to the presence of brown moss remains in the Ca' Fornera and Palazzetto cores (Figs. 5 and 6); van Geel et al. (1989) report the same correspondences in the Usselo section (The Netherlands). On the contrary, Type 340 is absent in the Fiorentina core where moss remains are scarce and Cyperaceae or Poaceae remains dominant. Even if we did not find a strong correlation between spores and macrofossils, we suppose that Type 340 was produced by the most abundant identified brown moss, Scorpidium scorpioides. In fact the spores of S. scorpioides (Boros and Járai-Komlódi, 1975; Reille, 1998) are very similar to Type 340 and it is reported that sporogonia are seldom developed (Boros and Járai-Komlódi, 1975). Many types of fungal remains have been identified. The most abundant and frequent are Type 200, Glomus (T. 207; van Geel et al., 1989), Gaeumannomyces (T. 126; Pals et al., 1980), Type 351 (van Geel et al., 1981), Type 79 (van Geel, 1978) and incompletely known types. Their presence is discontinuous in each sequence, suggesting frequent fluctuations of the water depth and periodic emersions of the bottom of the ponds or fens. The emersion of the wetland can activate the fungal decomposition of organic materials on the surface. A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Nevertheless, we cannot exclude an erosive origin of the fungal spores, especially in the sediments where undeterminable pollen frequency and mineral content are high (e.g. FI _1873). Some eggs of Neorhabdocoela (Microdallyellia armigera Type 1-B, Gyratrix hermaphroditus Ecotype I, Strongylostoma radiatum Type 1-A) have been identified following Haas (1996). They are produced by two eurythermic species (M. armigera and G. hermaphroditus) and by a warm stenothermal summer species (S. radiatum) with preference for lake environments with low water levels. They have all been recorded in eu-mesotrophic environments (Table 6). 6. A model of peat formation Peat layers in the overbank fines indicate wetlands in the LGM alluvial plain. These peat layers, which, in different sites in the Veneto–Friuli plain, have been documented to extend for 104–106 m2 (Iliceto et al., 2001; Bondesan et al., 2002; Fontana, 2004; Bondesan et al., 2004) formed because of both: i) waterlogged soils and/or ponds; ii) low minerogenic vs. organic sedimentation. Crevasses in the natural levees, or spillovers during bankfull stages, could flood large sectors of the plain, forming ponds. Nevertheless, Piave fluvioglacial waters had high sediment loads, and during floods large amounts of sediments were brought to the plain. These events, thus, led to the rapid aggradation of the floodplain, but did not fill the condition (ii) for the formation of peat levels, that is the deposition of prevalently organic material. On the other hand, it has to be recalled that, in present times, the underground water table in the low Veneto– Friuli plain is generally at depths of less than 1.5–2 m. More specifically, in the low Nervesa megafan at elevations of 10 to 5 m a.s.l., the depths are between 1 and 1.5 m (Ragazzi et al., 2004). These depths are artificially controlled by means of hydraulic artefacts: without human intervention, the water table would be closer to the surface. This water table is discontinuous, and exists because the silty-clay deposits stop the infiltration of rain water from the surface and seepage water from rivers and irrigation canals. If we transpose this hydrogeological situation to the LGM climatic conditions, the presence of the pensile fluvial channels running on top of the fluvial ridges could allow a constant recharge of the groundwater table in the surrounding areas. This may easily happen by means of water seeping through the sandy channel deposits. Furthermore, the lower evapo-transpiration rates due to the cold climate imply the possibility of a groundwater 77 table higher than the one which exists in the temperate Holocene climate. Altogether, it is reasonable to hypothesise a near-surface water table during the LGM in the study area (Fig. 11). The local outcrop of the water table in the low lying, inter-ridge depressions, could allow the formations of fens. As a result of the high water table, the waterlogged soils would not be efficient in absorbing rain and snowmelt water, which would accumulate in depressions, forming shallow ponds. The nutritive quality of the water preferred by the identified species, or suggested in literature for the recognized fossil types, is typical of eutrophic to mesotrophic waters, rich in cations (Table 6). This suggests that feeding waters were rich in cations. High values of conductivity have been measured in water samples from a lake located in the research area and fed by the ground-water table (Salmaso et al., 1995). Such evidence suggests that wet conditions were most probably related to the rise of the ion-rich water table rather than to the formation of ponds fed by rain and/or snow-melt water. This can be regarded as evidence that fens developed when precipitation and evaporation were rather low, with limited surface runoff and that the recharge of the groundwater was due mainly to seepage from the fluvio-glacial Piave river. Dry climate conditions in the lowlands of the Adriatic Basin are expected in the accepted palaeoclimatic reconstructions for the LGM (for a recent review see Ravazzi et al., 2004). Nevertheless all the fossil records indicate fluctuations of the wetland water level: i) micro and macrofossils of helophytes, hydrophytes and mosses, which grow in places periodically or always inundated; ii) free-floating algae, which live in open water, and the resting cells of some of them (mainly Zygnemataceae), which develop in critical hydrological or climatic conditions; iii) fungal spores, which generally develop in dry conditions; iv) remains of invertebrates which prefer shallow water environments. These variations were due to changes in the water table depth, related to fluctuations in the groundwater recharge from the fluvial system. These may happen for two reasons: i) variability of the Piave River discharges; ii) proximity/remoteness of the active river channels. Concerning the first point, in a fluvioglacial alluvial system a seasonal variability may be expected, with peak discharges in spring and summer and minimum in fall and winter. On longer time scales, variability could also be due to climatic instability during the LGM, as distinct cycles of climatic oscillations with centennial to millennial duration are recorded in the LGM lacustrine sediments from central Italy (Chondrogianni et al., 2004). Unfortunately, available data only allow us to detect variability, without providing 78 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Fig. 11. Schematic evolution of LGM peat-producing fens in the low Venetian plain (distal reaches of the Nervesa megafan). Explanations in the text. information on the duration of the cycles. Concerning the second hypothesis, published data (Bondesan et al., 2002; Mozzi and Bondesan, 2004; Fontana et al., 2004, submitted for publication) indicate a high channel mobility of the LGM alluvial systems in the Venetian Po Plain. Fig. 11 shows how the LGM peat levels in the lower Nervesa megafan may have formed and have been incorporated in the stratigraphic record. Because of near surface water table, fens can develop in the depressions between abandoned fluvial ridges (1) and (2). The absence of fluvial sedimentary activity in such locations allows the development of aquatic plant communities and the sedimentation of mainly organic material, with the formation of an extensive peat layer. At the foot of the active ridge (3) mineral sedimentation is dominant, and the organic material, eventually related to the existence of ephemeral fens, is dispersed in the flood basin silty-clay sediments. Meanwhile the evolution of the hydrosere is stopped by the burial of the plants by sediments. At the distal end of the flood basin, towards the foot of ridge (2), the sedimentation A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 rate of minerogenic material is increasingly lower than the rate of peat production, and a mainly organic layer can form. Continuous vertical aggradation of the alluvial ridge and of the connected floodplain leads to the burial of peat levels and of the abandoned ridges. The time of formation and burial of the peat level can be estimated in the order of 10–102 years. In such unstable hydrological and sedimentological conditions the plant communities did not develop into terrestrial or raised bog communities. 7. Conclusions During the LGM in the Venetian Po plain, wide and low-lying areas between the fluvial ridges were periodically inundated by the outcropping of groundwater. Rich fens developed in these water-logged areas and the local wetland plant communities were characterized by reeds, sedges and mosses. The identification of NPPs and macrofossils has been crucial to the reconstruction of local plant communities, mainly because in open sites very poor correlations exist between pollen and plants as Pardoe (2001) demonstrated in modern arctic–alpine environments. The fluctuation of the water level indicated by the fossil assemblages, was due to variations of the ground water table depth: a low water table led to dry surface conditions. The depression of the water table probably depended on the effectiveness of the aquifer recharge by seeping water from the fluvio-glacial Piave River, in relation to: i) low autumn and winter river discharges and/or ii) millennial to centennial climatic cycles and/or iii) migration of the active channel belt several kilometres away from the core location. Peat accumulation in continuous layers could take place only where the fen organic deposition prevailed over the alluvial minerogenic sedimentation. This was possible in sheltered areas, away from the active river channels. The rapid vertical aggradation and high lateral mobility of the active channel belts permitted brief fen development, probably in the order of decades to centuries. When the alluvial deposition outranged the organic deposition, the peat level was buried and incorporated into the stratigraphic record. The evidence of very short-lived fens derive from the analysis of the alluvial chronostratigraphy, and is consistent with the limited development of plant communities observed in the different case studies. The limited time did not allow to complete the theoretical hydroseral development up to the formation of raised bog, with the colonization of the wetlands by trees and Sphagnum. 79 Acknowledgements Financial support by Fondazione Cassamarca (Treviso) and by Italian MIUR (ex-60% Bondesan, “Geomorphological evolution of flood plains, with particular interest in Veneto–Friuli plain”) are gratefully acknowledged. We thank John H. McAndrews and S. J. P. Bohncke for helpful comments and corrections on the manuscript. The geomorphology and sedimentology were undertaken by A. Bondesan, P. Mozzi and S. Piovan; palynology by S. Favaretto, A. Miola and S. Piovan; paleobotany by I. Sostizzo; statistical analysis by L. Corain. Appendix A. Descriptions of some microfossils of incompletely known origin Type shown on Plate III, 15: globose microfossil (12) 14–18(26) μm in diameter, wall 1.5–2 thick, smooth and with a S-shaped furrow. Its morphology may suggest a taxonomical relationship with Type 128A and Type 128B, particularly for its S-shaped furrow. In our samples it occurs with Type 128A, which is an eu- to mesotrophic open water environments indicator (van Geel et al., 1989). Type shown on Plate III, 24: fungal spores, multiseptate, dark brown, up to 300 μm long, 5–7 μm wide, with no broader part, often broken off at one or both ends. Type 74 (Plate III, 20 and 21): undefined microfossil, globose, hyaline, 20–25 μm in diameter, reticulate, with 2 μm high ridges. Meshes of the reticulum about 5 μm wide. We observed cells very similar to Type 74 in surface sediment samples collected from a prealpine small pond (Val Piana–Belluno, northeastern Italy). Many Clamydophyceae (Chlorophyta) produce zygotes very similar to both fossil Type 74 and our observed objects (Ettl, 1983). In order to identify the organisms, producers of zygotes, future studies will involve the analysis of water samples during blooming season of the algal community. Type 303 (Plate III, 18): unidentified microfossil, globose, hyaline, (12)15–20(22) μm in diameter. Wall smooth, yellow to brownish yellow, 1–1.5 μm thick, often with a furrow. Type 303 is frequently recorded in fossil associations of open water environments in euto mesotrophic conditions (van Geel et al., 1981). In our samples Type 303 and remains of Scorpidium scorpioides have been frequently recorded together. We tentatively analysed a sample of surface sediment from a small pond with variable water level (Val Piana – Belluno, northeastern Italy), where a rare community of S. scorpioides is still present. In the acetolysed samples many globose, smooth, yellow cells with a 80 A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 variable 12–25 μm diameter and often with a furrow (Plate III, 19) have been recorded. They are very similar to Type 303. Basic fucsin colours both the microfossil and the observed cell walls, therefore we can exclude a fungal or animal origin for both of them. Probably Type 303 is produced by aquatic organisms that form dense populations in the same environments of “brown mosses” as S. scorpioides. A very common aquatic taxon in small North-Italian ponds is Chlamydomonas sp. (Stoch, 2004), some species of which produce resting zygotes with the same morphology of Type 303 (Ettl, 1983, see e.g. pp. 360, 364, 373). So our future researches will be focused on this group of algae. References A.R.P.A.V., 2001. Studio di 15 biotopi in area dolomitica. Duck, Belluno. Aaby, B., Berglund, B.E., 1986. Characterization of peat and lake deposits. In: Berglund, B.E. (Ed.), Handbook of Holocene Palaeoecology and Palaeohydrology. John Wiley & Sons, London, pp. 231–246. Andersen, S.T., 1978. Identification of wild grass and cereal pollen. Danm. Geol. Unders. Årbog 69–92. Bakker, M., van Smeerdijk, D.G., 1982. A palaeoecological study of a Late Holocene section from “Het Ilperveld” western Netherlands. Rev. Palaeobot. Palynol. 36, 95–163. Bertolani Marchetti, D., 1966. Vicende climatiche e floristiche dell'ultimo glaciale e del postglaciale in sedimenti della laguna veneta. Mem. Biogeogr. Adriatica 7, 193–225. Birks, H.J.B., Birks, H.H., 1980. Quaternary Palaeoecology. Edward Arnold, London. Bondesan, A., 1999. Una nuova data 14C nell'anfiteatro morenico di Vittorio Veneto e la costruzione del suo sandur (pianura del F. Meschio). Atti VIII Convegno Glaciologico Italiano, Bormio, p. 2. Bondesan, A., Mozzi, P., Calderoni, G., 2002. L'assetto geomorfologico della pianura veneta centro-orientale, stato delle conoscenze e nuovi dati. In: Varotto, M., Zunica, M. (Eds.), Scritti in ricordo di Giovanna Brunetta. University of Padua, Dept. of Geography, Padua, pp. 19–38. Bondesan, A., Finzi, E., Fontana, A., Francese, R., Magri, S., Mozzi, P., Primon, S., Zamboni, C., 2004. La Via Annia a Ca' Tron: nuovi contributi della geomorfologia, della geofisica e del telerilevamento. In: Busana, M.S., Ghedini, F. (Eds.), La via Annia e le sue infrastrutture, Atti delle Giornate di Studio (Ca' Tron, Roncade, 6 – 7 novembre 2003). Grafiche Antiga, Cornuda, Treviso, pp. 109–146. Boros, A., Járai-Komlódi, M., 1975. An Atlas of Recent European Moss Spores. Akadémiai Kiadó, Budapest, Hungary. 466 pp. Bortolami, G.C., Fontes, J., Markgraf, V., Saliege, J.F., 1977. Land, sea and climate in the northern Adriatic region during Late Pleistocene and Holocene. Palaeogeogr. Palaeoclimatol. Palaeoecol. 21, 139–156. Buurman, P., 1969. Pollen analysis of Holocene and Pleistocene sediments in the neighbourhood of Portogruaro (Venezia)—Italy. Mem. Biogeogr. Adriatica 8, 41–45. Calderoni, G., Castiglioni, G.B., Foddai, D., Gallio, S., Hinterwipflinger, V., Lombardo, M., Miola, A., Zangheri, P., 1996. Palaeoenvironmental features of a perieuganean (Padua, N-Italy) depression during the Late Quaternary, first results. Il Quaternario 9, 667–670. Castiglioni, G.B., 1999. Geomorphology of the Po Plain. Geogr. Fis. Din. Quat. (Suppl. III), 7–20. Castiglioni, G.B., Girardi, A., Rodolfi, G., 1987. Le tracce degli antichi percorsi del Brenta per Montà ed Arcella nei pressi di Padova, studio geomorfologico. Mem. Sci. Geol. 39, 129–149. Chondrogianni, C., Ariztegui, D., Rolph, T., Juggins, S., Shemesh, A., Rietti-Shati, M., Niessen, F., Guilizzoni, P., Lami, A., McKenzie, J.A., Oldfield, F., 2004. Millennial to interannual climate variability in the Mediterranean during the Last Glacial Maximum. Quat. Int. 122, 31–41. Elenga, H., Peyron, O., Bonnefille, R., Jolly, D., Cheddadi, R., Guiot, J., Andrieu, V., Bottema, S., Buchet, G., de Beaulieu, J.L., Hamilton, A.C., Maley, J., Marchant, R., Perez-Obiol, R., Reille, M., Riollet, G., Scott, L., Straka, H., Taylor, D., Van Campo, E., Vincens, A., Laarif, F., Jonson, H., 2000. Pollen-based biome reconstruction for southern Europe and Africa 18,000 yr BP. J. Biogeogr. 27, 621–634. Erdtman, G., Berglund, B., Praglowski, J., 1961. Introduction to a Scandinavian Pollen Flora Vol. I. Almqvist & Wiksell, Stockholm. Ettl, H., 1983. Chlorophyta. I. Phytomonadina. In: Pascher, A. (Ed.), Suβwasserflora von Mitteleuropa. Gustav Fischer Verlag, Stuttgart. Faegri, K., Iversen, J., 1989. Textbook of Pollen Analysis, IV Ed. John Wiley & Sons, London. Fontana, A., 2004. Tra Tagliamento e Livenza. In: Bondesan, A., Meneghel, M. (Eds.), Note illustrative della carta geomorfologica della provincia di Venezia. Esedra, Venezia, pp. 195–217. Fontana, A., Mozzi, P., Bondesan, A., 2004. L'evoluzione geomorfologica della pianura veneto–friulana. In: Bondesan, A., Meneghel, M. (Eds.), Note Illustrative della carta Geomorfologica della Provincia di Venezia. Esedra, Venezia, pp. 113–136. Fontana A., Mozzi P., Bondesan A., submitted for publication. Alluvial megafans in the Venetian–Friulian Plain (North-eastern Italy): evidence of sedimentary and erosive phases during Late Pleistocene and Holocene. Quaternary International. Giovannelli, M.M., Rizzi Longo, D., Stolfa, D., Zucchi Stolfa, M.L., 1986. Considerazioni paleoecologiche sui sondaggi S19–Lignano e S20–Bevazzana (Delta del fiume Tagliamento). Gortania 7, 87–112. Haas, J.N., 1996. Neorhabdocoela oocytes – palaeoecological indicators found in pollen preparations from Holocene fresh water lake sediments. Rev. Palaeobot. Palynol. 91, 371–382. Iliceto, V., Meloni, F., Mozzi, P., Rizzetto, F., 2001. Il sottosuolo della Cappella degli Scrovegni a Padova. Geol. Tec. Ambient. 9, 3–17. Kuhry, P., 1988. A palaeobotanical and palynological study of Holocene peat from the El Bosque mire. Located in a volcanic area of the Cordillera central of Colombia. Rev. Palaeobot. Palynol. 55, 19–72. Kuhry, P., 1997. The palaeoecology of a treed bog in western boreal Canada: a study based on microfossils, macrofossils and physicochemical properties. Rev. Palaeobot. Palynol. 96, 183–224. Magri, S., Bondesan, A., 2004. Le radiodatazioni al 14C nella carta geomorfologica. In: Bondesan, A., Meneghel, M. (Eds.), Note Illustrative della carta Geomorfologica della Provincia di Venezia. Esedra, Venezia, pp. 98–100. Marchetti, M., Mozzi, P., Ravazzi, C., 2004. Large alluvial fans active during the LGM in the Alpine foothills. In: Antonioli, F., Vai, G.B. (Eds.), CLIMEX Litho-Palaeoenvironmental Maps of Italy During the Last Two Climatic Extremes: Explanatory Notes. Museo Geologico G. Cappellini, Bologna, pp. 32–33. Marocco, R., 1989. Evoluzione quaternaria della laguna di Marano (Friuli–Venezia Giulia). Il Quaternario 2, 125–137. Miola, A., Albanese, D., Valentini, G., Corain, L., 2003. Pollen data for a biostratigraphy of LGM in the Venetian Po Plain. Il Quaternario 16, 21–26. A. Miola et al. / Review of Palaeobotany and Palynology 141 (2006) 53–81 Moore, P.D., Webb, J.A., Collinson, M.E., 1991. Pollen Analysis, II Ed. Blackwell, Oxford, Great Britain. Mozzi, P., Bondesan, A., 2004. La stratigrafia del sottosuolo della tenuta di Ca' Tron. In: Busana, M.S., Ghedini, F. (Eds.), La via Annia e le sue infrastrutture, Atti delle Giornate di Studio (Ca' Tron, Roncade, 6 – 7 novembre 2003). Grafiche Antiga, Cornuda, Treviso. Mozzi, P., Bini, C., Zilocchi, L., Becattini, R., Mariotti Lippi, M., 2003. Stratigraphy, paleopedology and palynology of Late Pleistocene and Holocene deposits in the landward sector of the lagoon of Venice (Italy), in relation to the caranto level. Il Quaternario 16 (1Bis), 193–210. Mullenders, W., Favero, V., Ciremans, M., Dirickx, M., 1996. Analyses polliniques de sondages a Venise (VE-I, VE-I bis, VE-II). In: Gullentops, F. (Ed.), Pleistocene Palynostratigraphy, Aardk. Mededel., vol. 7, pp. 87–117. Nyholm, E., 1979. Illustrated Moss Flora of Fennoscandia. Gleerup and Swedish Natural Science Research Council. Reprinted in Sweden by Gotab Kungälv. Oberdorfer, E., 1949. Pflanzensoziologische Exkursions Flora. Eugen Ulmer, Stuttgart. Paganelli, A., 1996. Evolution of vegetation and climate in the Veneto–Po plain during the Late-Glacial and the Early Holocene using pollen–stratigraphic data. Il Quaternario 9, 581–590. Paganelli, A., Miola, A., Todaro, A., 1988. Ricerche palinologiche nel Basso Veneto. 1. Biostratigrafia palinologica e dati radiometrici di alcuni depositi fluvio-lacustri. Athesia 2, 105–122. Pals, J.P., van Geel, B., Delfos, A., 1980. Paleoecological studies in the Klokkenweel bog near Hoogkarspel (Noord Holland). Rev. Palaeobot. Palynol. 30, 371–418. Pardoe, H.S., 2001. The representation of taxa in surface pollen spectra on alpine and sub-alpine glacier forelands in southern Norway. Rev. Palaeobot. Palynol. 117, 63–78. Pellegrini, G.B., Paganelli, A., Penso, D., 1984. Aspetti geomorfologici e palinologici dei depositi fluviali nei dintorni di Carturo sul Brenta (Padova). Geogr. Fis. Din. Quat. 7, 36–39. Peyron, O., Guiot, J., Cheddadi, R., Tarasov, P., Reille, M., de Beaulieu, J.L., Bottema, S., Andrieu, V., 1998. Climatic reconstruction in Europe for 18,000 yr B.P. from pollen data. Quat. Res. 49, 183–196. Pignatti, S., 1982. Flora d'Italia. Edagricole, Bologna, Italy. Polunin, O., Walters, M., 1985. A Guide to the Vegetation of Britain and Europe. Oxford University Press. Prentice, I.C., Guiot, J., Harrison, S.P., 1992. Mediterranean vegetation, lake levels and palaeoclimate at the Last Glacial Maximum. Nature 360, 658–660. Ragazzi, F., Vinci, I., Garlato, A., Giandon, P., Mozzi, P., Fiorentin, R., Calzolari, C., Ungaro, F., 2004. Carta dei suoli del bacino scolante in laguna di Venezia. ARPAV, Osservatorio Regionale Suolo, pp. 158–189. Ravazzi, C., Orombelli, G., Tanzi, G., CLIMEX group, 2004. An outline of the flora and vegetation of the Adriatic basin (northern Italy and eastern side of the Apennine) during the Last Glacial Maximum. In: Antonioli, F., Vai, G.B. (Eds.), CLIMEX LithoPalaeoenvironmental Maps of Italy During the Last Two Climatic Extremes: Explanatory Notes. Museo Geologico G. Cappellini, Bologna, pp. 15–19. 81 Reille, M., 1998. Pollen et Spores d'Europe et dAfrique du Nord. Supplement 2. Laboratoire de Botanique Historique et Palynologie, Marseille, France. Roberts, N., Wright, H.E., 1993. Vegetation, lake-level and climatic history of the near east and southwest Asia. In: Wright Jr., Kutzbach, J.E., Webb III, Ruddiman, W.F., Street-Perot, F.A., Bartlein, P.J. (Eds.), Global Climates Since the Last Glacial Maximum. University of Minnesota Press, Minneapolis, pp. 94–220. Salmaso, N., Manfrin, M., Cordella, P., 1995. Struttura e dinamica della comunità fitoplanctonica in un piccolo lago di falda (Rubàno, Padova). Atti VI Congresso Nazionale Società Italiana di Ecologia Edizioni Zara, Venice, Italy, pp. 703–706. Serandrei Barbero, R., Lezziero, A., Albani, A., Zoppi, U., 2001. Depositi tardo-pleistocenici ed olocenici nel sottosuolo veneziano: paleoambienti e cronologia. Il Quaternario 14, 9–22. Stoch, F., 2004. Forme microscopiche di vita fra gli sfagni, Le torbiere montane. In: Minelli, A., Ruffo, S., Stoch, F. (Eds.), Quaderni Habitat. Museo Friulano di Storia Naturale Udine, pp. 58–59. Tropeano, D., Cerchio, E., 1984. L'orizzonte torboso würmiano nel sottosuolo della pianura piemontese meridionale. Osservazioni preliminari. Boll. Assoc. Min. Subalp. 21 (3), 199–221. Valero-Garcés, B.L., González-Sampériz, P., Navas, A., Machín, J., Delgado-Huertas, A., Pena-Monné, J.L., Sancho-Marcén, C., Stevenson, T., Davis, B., 2004. Paleohydrological fluctuations and steppe vegetation during the Last Glacial Maximum in the central Ebro valley (NE Spain). Quat. Int. 122, 43–55. van Geel, B., 1978. A palaeoecological study of Holocene peat bog sections in Germany and the Netherlands, based on the analysis of pollen, spores and macro- and microscopic remains of fungi, algae, cormophytes and animals. Rev. Palaeobot. Palynol. 25, 1–120. van Geel, B., Bohncke, S.J.P., Dee, H., 1980–1981. A palaeoecological study of upper Late Glacial and Holocene sequence from “De Borchert”, The Netherlands. Rev. Palaeobot. Palynol. 31, 367–448. van Geel, B., Hallewas, D.P., Pals, J.P., 1983. A Late Holocene deposit under the Westfriese Zeedijk near Enkhuizen (prov. of N-Holland, The Netherlands): palaeoecological and archaeological aspects. Rev. Palaeobot. Palynol. 38, 269–335. van Geel, B., Coope, G.R., van der Hammen, T., 1989. Palaeoecology and stratigraphy of the Lateglacial type section at Usselo (The Netherlands). Rev. Palaeobot. Palynol. 60, 25–129. Venzo, S., 1977. Depositi quaternari e del Neogene superiore nella bassa valle del Piave da Quero al Montello e del Paleopiave nella valle del Soligo (Treviso). Mem. Ist. Geol. Mineral. Univ. Padova 30, 64. Vigliotti, L., 2004. Late Glacial reconstruction of Adriatic Sea fluvial pattern. In: Antonioli, F., Vai, G.B. (Eds.), CLIMEX LithoPalaeoenvironmental Maps of Italy During the Last Two Climatic Extremes: Explanatory Notes. Museo Geologico G. Cappellini, Bologna, pp. 39–40. Watson, E.V., 1968. British Mosses and Liverworts, II Ed. Cambridge University Press, Cambridge. Yu, G., Harrison, S.P., 1995. Lake status record from Europe: data base documentation. NOAA Paleoclimatology Publication Series Report, vol. 3. World Data Center-A for Paleoclimatology, Boulder, CO, USA. 451 pp.