PALEVO 1631-0683 Elsevier S1631-0683(06)00185-0 10.1016/j.crpv.2006.12.001 Research article General Palaeontology (Palaeoecology) Bioaccumulations terrestres / Terrestrial bioaccumulations Differential accumulation of miospores in Upper Miocene sediments of the La Cerdaña basin (eastern Pyrenees, Spain) Accumulation différentielle des miospores dans les sédiments du Miocène supérieur du bassin de Cerdagne (Est des Pyrénées, Espagne) Néraudeau Didier Barrón Eduardo e.barron@igme.es a Comas-Rengifo María José b Museo Geominero, Instituto Geológico y Minero de España (IGME), Ríos Rosas 23, 28003 Madrid, Spain Departamento de Paleontología, Facultad de Ciencias Geológicas (UCM), José Antonio Novais 2, 28040 Madrid, Spain 6 1-2 S1631-0683(07)X0034-4 Bioaccumulations et bioconstructions fossiles 157 168 © 2006 Académie des sciences. Published by Elsevier B.V. All rights reserved. 2006 Académie des sciences Full (PDF)

The Upper Miocene basin of La Cerdaña (eastern Pyrenees, Spain) is characterised by siliciclastic sedimentation in alluvial, fluvial, deltaic, and lacustrine depositional settings. The lacustrine deposits in the southwestern part of the basin show intercalations of levels rich in organic matter linked to plant accumulations. The differential accumulation of miospores in this part of the basin is due to the intrinsic features of the miospores themselves and of their producers, and extrinsic factors such as the palaeoecology of the miospore producers and the characteristics of the depositional setting. The features of the miospore assemblages and the factors that produced their accumulation are analysed. The proportions of the different miospores in the studied assemblages showed that their accumulation occurred differently in the various lake sediments; this agrees well with a NE–SW running Neves effect. This latter effect is mainly a result of the type of transport the miospores underwent, which was influenced by the vegetation and climate, the structure of the lake, and other phenomena such as fires. The sediments from swampy areas were characterised by high percentages of local and extra-local miospores of parautochthonous and allochthonous nature, whereas those from deepwater areas showed mainly extra-local and regional miospores. The means by which the miospores were preserved in the different lithologies of the La Cerdaña basin are analysed.

Le bassin de Cerdagne (Est des Pyrénées, Espagne), daté du Miocène supérieur, est caractérisé par une sédimentation silicoclastique où dominent les dépôts alluviaux, fluviatiles et deltaïques. Les dépôts lacustres, riches en matière organique et incluant des restes végétaux, se sont déposés au sud-ouest du bassin. Une accumulation différentielle des miospores est essentiellement due à la coexistence de facteurs intrinsèques (caractéristiques de la miospore et de son producteur) et extrinsèques (paramètres environnementaux). Les différents assemblages de miospores fossiles ainsi que les facteurs qui ont déterminé leur accumulation dans les sédiments lacustres du Miocène supérieur de Cerdagne sont analysés dans ce travail. La concentration pollinique des différents niveaux étudiés révèle une accumulation différentielle dans les sédiments lacustres, en accord avec l’effet Neves, orienté NE–SW. Cet effet est lié au type de transport des miospores, à la végétation et au climat, à la structure du lac et à la fréquence d’événements tels que les incendies. Ainsi, les sédiments provenant de zones marécageuses sont caractérisés par des pourcentages élevés en miospores locales et extra-locales à caractère para-autochtone et allochtone, alors que les sédiments provenant d’eaux profondes révèlent principalement des miospores extra-locales et régionales à caractère allochtone. La préservation des miospores provenant des différentes lithologies rencontrées dans le bassin de Cerdagne a également été analysée.

 Palynology, Taphonomy, Thanatocoenosis, Neves Effect, Upper Miocene, La Cerdaña basin, Spain Palynologie, Taphonomie, Thanatocénose, Effet Neves, Miocène supérieur, Bassin de Cerdagne, Espagne presented Written on invitation of the Editorial Board Introduction

A fossil plant assemblage is an accumulation of plant parts derived from one or more individuals entombed within a volume of sediments laid down under essentially common conditions [64]. The composition and structure of plant fossil assemblages are the result of a combination of biological features (the type, size, shape, number, and composition of their remains), taphonomic factors (transport, degradation, and diagenesis), and ecological patterns (community composition and structure). Thus, a plant assemblage may differ significantly from the plant community from which it was derived [30].

The pollen and spore walls of plants are composed of sporopollenin, a complex polymer of carotenoids and carotenoid esters [21]. This compound is remarkably resistant to degradation (although it eventually decays in aerobic conditions via oxidative processes) [28], [36] and [37]. Miospores (sensu Guennel [35]) can therefore accumulate in abundance and are preserved in an essentially unaltered state (duripartic preservation sensu Schopf [63]) in a variety of materials, from which they can be recovered and identified.

Lake bottoms with anoxic waterlogged sediments are one of the best environments for miospore preservation [40]. In such places, large numbers accumulate and fossilize, and they are easy to extract for study. Their accumulations also provide information about the organization and distribution of the sediments in lacustrine basins, and therefore of the environmental conditions that existed at the time of their formation.

Owing to its richness in preserved phytocoenoses, the palaeobotany of the lacustrine rocks of the La Cerdaña basin has been known since the 19th Century [9], [44], [47] and [58]. These rocks contain huge quantities of well-preserved miospores that represent mixed mesophytic mountainous and subtropical forests [9] and [18]. The first palynological studies were carried out in the 1950s [41] and have continued until the present day [7], [9], [18] and [46].

The main aim of the present work was to describe the different miospore accumulations (thanatocoenoses; sensu Ochev [50]) in the Upper Miocene lacustrine rocks of the La Cerdaña basin. The results complement existing knowledge of the basin's plant taphonomy [8], [9], [10], [44] and [46].

 Geological setting

The La Cerdaña basin is one of the small basins that formed during the Neogene in connection with a series of NE–SW- and east–west-running fractures [23] and [62] in the Axial Zone of the eastern Pyrenees. At its southeastern edge, it is bound by the La Têt Fault, and by another large, normal east–west-running fault that separates Tertiary from Palaeozoic materials (Fig. 1). The northeastern margin is more irregular, and corresponds basically to an unconformity [59] and [62]. This imparts a very asymmetric shape to the basin. Towards the southeast, the Neogene materials attain a thickness of up to 1000 m [23], [42] and [56]. Two depositional units have traditionally been distinguished within these predominantly siliciclastic materials, each presenting a very different spatial distribution. These correspond to two stages in the evolution of the Neogene basin, in which tectonics and sedimentation played significant roles [6], [23], [60] and [62]. The Lower Unit, which comprises the oldest (Vallesian; Upper Miocene) sediments is responsible for the majority of the present outcrops in the basin, while the Upper Unit, which at its base is Turolian in age [1], is represented only at the southern end of the basin. A clear change in lithology occurs between these two units [61], and the detritic components of their rocks are quite different in character. This is a consequence of an important alteration in the origin of the sediments that continue to fill the basin.

This paper concentrates on the analysis of the miospore associations identified in the Lower Unit materials (400–800 m), a thick succession of alluvial, fluvial, deltaic, and lacustrine sediments [2], [42] and [59].

The Puigcerda sub-basin, situated to the north (Fig. 1), is dominated by alluvial fans and fluvial plains that change distally to deltaic-palustrine facies and finally to lacustrine facies [23]. The southern part of this sub-basin is dominated by the sedimentation of sandstones, lutites rich in organic material, and lignite beds. This part of the sub-basin is home to the localities of Sanavastre (SAN) and Sampsor (SAM), which were studied by Barrón [9], [11] and [12] (Fig. 1 and Fig. 2). The palustrine-deltaic character of the materials in this area provides an interesting succession of miospore associations, which can be readily studied.

The sub-basin of Bellver, situated to the south, is dominated by a very uniform lacustrine sedimentation of highly laminated diatomites and lutites rich in organic materials. At certain points, there is a well-preserved palaeofauna, including insects [14]. The remaining localities studied (Fig. 1 and Fig. 3) lie in this sub-basin, in which all the possible productive levels (from a palynological point of view) were sampled. It is difficult to establish precisely the stratigraphic position of the levels studied, due to the difficulty in finding continuous outcrops; the stratigraphic successions of Roca [61] and [62] and Martín-Closas et al. [46] were therefore taken into account. To facilitate the analysis of the palynological successions, the localities were grouped into two sectors according to the distribution of the facies in the sub-basin. The eastern sector included the outcrops of Coll de Saig (COL), Prats (PRAT), Riu de Santa María (RIU), Baltarga (BAL), Beders (BED), and Torrent de la Bavosa (TOB), while the western sector included Bellver de Cerdanya (BEL), Barranc de Salanca (BAR), and Torrent de Vilella (TOV). The eastern outcrops were related to the fan delta facies, while those of the west corresponded to the more distal sediments of the lake.

 Material and methods

Twenty-two samples from the Sanavastre open cast lignite mine were examined, along with 31 samples from the Sampsor open cast lignite mine (Fig. 2), and 30 from different outcrops containing diatomites (Fig. 3).

Samples were prepared following the standard palynological technique [16] based on acid treatment (HCl, HF, HNO3) at high temperature. Residues were sieved through 500-, 250-, 75-, 50- and 10-μm sieves. Slides were mounted in glycerine jelly. Some 500–1000 palynomorphs (on up to four slides) were identified per sample to determine the species ratios. The miospore concentration was calculated using the formula proposed by Cour [26].

To study the different accumulations of spores and pollen in the Vallesian rocks of the La Cerdaña basin, nine palynological groups of miospores were taken into account according to their bio- and ecological characteristics: (i) pteridophytes (P): trilete and monolete spores; (ii) Taxodiaceae-Cupressaceae (TC); (iii) bisaccate pollen grains (BIS): pollen of conifers such as Abies, Cathaya, Picea, Pinus (Fig. 4.5); (iv) riparian trees (RIP), numerically dominated by Alnus pollen (Fig. 4.7); (v) hydro-hygrophilous (HH): pollen of herbaceous plants from aquatic and shoreline areas such as Potamogeton, Trapa (Fig. 4.13), Sparganium (Fig. 4.11), and Nymphaeaceae; (vi) non-riparian mesophytic trees (M), pollen of deciduous trees such as Fagus, Carpinus, Carya, Quercus (Fig. 4.14), and Tilia (Fig. 4.9), which were not directly associated with the lake shore; (vii) herbs (NAP): pollen of non-riparian and non-aquatic herbs such as Saxifraga (Fig. 4.8), Apiaceae (Fig. 4.6), Caryophyllaceae, Asteraceae; (viii) ‘other miospores’; all those remaining that were identifiable, including pollen grains of taxa such as Engelhardia (Fig. 4.12), Arecaceae and Sapotaceae, which could not be included in other groups because of their ecological preferences); and finally (ix) unidentified pollen grains. Fig. 2 and Fig. 3 show the proportions of these palynological groups for each sedimentary level.

 Results Palynology of the deltaic outcrops

The palynological study of 22 levels of the Sanavastre mine led to the identification of 19 assemblages with more than 300 specimens per sample [9] and [12]. P, BIS, RIP, and HH were the best represented groups (Fig. 2.1). The mean miospore concentration was 579 × 103 grains/g of rock.

The Sampsor mine (Fig. 4.1) had 15 levels that yielded statistically representative assemblages [9] and [11]. The groups identified were the same as those in the Sanavastre mine, although HH was always in proportions below 12% (Fig. 2.2). The mean miospore concentration in this mine was 387 × 103 grains/g of rock. The proportions (and therefore the ratios) of the different palynological groups in the rocks of these mines were not related to any particular lithology: P, BIS, and RIP were equally present in lignites and lutites (Fig. 2). Similarly, the proportion of a particular group sometimes changed between neighbouring levels of the same lithology. For instance, in the Sanavastre mine, the percentage of spores decreased from 27.6 to 1.85% in different samples (SAN11–SAN13) of lignite (Fig. 2.1). This was also seen in grey lutite samples, in which BIS ranged from 42.63% in SAN2 to just 5.97% in SAN4.

In this mine, the absolute miospore concentrations did seem to be related to the lithology. The greatest miospore concentration was found in lutites and the lowest in sands. SAN7 was the richest sample, with 1864 × 103 grains/g of rock, while the poorest sample was SAN14, with 112 × 103 grains/g of rock [9] and [12]. In the Sampsor mine, both the highest and lowest pollen concentrations were seen in lignites [9] and [11].

The Sanavastre mine and the middle section of the Sampsor mine showed similarities in terms of the proportions of palynomorph groups in each level; the mid section of the latter mine showed a stratigraphic organization similar to that seen in the Sanavastre mine [9], [11] and [12]. Although some of the other levels of the Sampsor mine showed individual resemblance to some of those of the Sanavastre mine, their miospore proportions were modified by an increase in the quantity of sand.

 Palynology of lacustrine outcrops

The rocks of the western lacustrine sector are formed from diatomites and mudstones. Sixteen samples showed a statistically representative number of palynomorphs (Fig. 3.1); all showed a scarcity of spores. The basal samples (BEL2 and BEL3) showed high percentages of RIP (42.8 and 41.5%, respectively) and M (22.1 and 30.5%, respectively). The percentage values for RIP and M underwent a marked reduction towards the top of the section, while those of BIS increased in number and became predominant in the remaining samples (except for TOV5). In all these levels, the percentages of M were higher than those of RIP. Low percentages of BIS (14.67%) and RIP (13.54%) and high percentages of M (30.48%) and undetermined pollen grains (23.17%) were seen in TOV5. The mean miospore concentration in this section was 618 × 103 grains/g of rock; the highest value was recorded for the basal sample BEL3 (992 × 103 grains/g of rock) and the lowest for TOV5 (302 × 103 grains/g of rock).

The comparative analysis of the assemblages from the lignite mines and of those from the lacustrine outcrops clearly reflects the Neves effect [24], [25], [49] and [66]. This is shown in the basin by the significant difference in the percentages of P, BIS, and RIP taxa in the sediments deposited in the alluvial delta zones. The three groups show noticeable percentages (Fig. 2) when compared with sediments from the deep and open lacustrine zones, which mainly contained BIS (Fig. 3.1).

Six outcrops of diatomites and mudstones were studied in the eastern sector of the palaeolake (Fig. 3.2). Generally, these samples were dominated by BIS, with percentages of 52.8–87.9%. RIU showed the smallest numbers of BIS (3.65%), but the highest of M (31.5%), RIP (26.46%), and NAP (8.62%). A large number of undetermined palynomorphs (27.43%) has also been evidenced. M was also well represented in BED (20.64%). Spores were very scarce except in PRAT (4.63%). In general, the pollen assemblages of the eastern sector (Fig. 3.2) showed a strong resemblance to those of the western sector (Fig. 3.1). BIS was the largest palynological group, except in RIU, where M and RIP predominated. The highest mean miospore concentrations were seen in the eastern section (1560 × 103 grains/g of rock); the highest were recorded for levels BALL, BED, and TOB (more than 1800 × 103 grains/g of rock), the lowest for the Coll de Saig outcrop (315 × 103 grains/g of rock in COL2 and 393 × 103 grains/g of rock in COL6).

 Discussion and conclusions Recruitment of miospores

The miospores incorporated into the thanatocoenoses of the La Cerdaña palaeolake clearly represent a mixture of components recruited from incoming water and the air. Surface water appears to have been the main vector of miospore transport to the lake [20], [45], [51] and [52]. The incoming streams would have received large quantities of palynomorphs from the catchment area by runoff. In addition, this basin had high rainfall (up to 1000 mm/year), and it was probably located at a height of 1000–1300 m above sea level [9] (assuming no isostatic elevation occurred). Thus, as occurs in present-day miospore accumulations [20] and [29], the samples derived from runoff were biased in their miospore content due to differences in their dispersal methods, and the effects of weathering and redeposition. When miospores fall into water, their sizes and the competence of the transporting currents may allow a sorting effect to occur [38]. The most abundant spores belonged to the genus Osmunda (Fig. 4.3). Their primary dispersal mechanism is by wind [33] and [34], but they are generally heavier than pollen grains and they accumulate close to the ferns that produce them (leptokurtic distribution [31] and [32]). In addition, they possess no suitable structures for flotation such as cinguli or coronae. The great amount of spores in the mines (Fig. 2) may simply be the result of the proximity of fern habitats to the sampling site, as seen in Lake Tulane (Florida) [67]. The small numbers of P in the deep lake sediments (Fig. 3) indicate the low transport efficiency of this group's spores. The majority of pollen types with pores and/or colpi take up water rapidly and sink [39]. Bisaccate pollen grains, however, float for longer periods since air remains trapped in the sacci. Pine pollen can retain its buoyancy for as long as four years [39], whereas other bisaccate pollen grains such as those of Abies and Picea (which are heavier than Pinus) sink very quickly [57]. Large numbers of broken bisaccate pollen grains, which must have spent long periods afloat were detected in the deep lacustrine outcrops. The type-3 palynofacies described by other authors [46] contained abundant, fragmented bisaccate pollen grains. These were probably brought to the palaeolake by water.

The anemophilous pollen of trees that grew near the lakeshore (RIP taxa, e.g., Alnus), and of non-riparian trees with high pollen production (BIS and M taxa, e.g., Pinus, Quercus, and Betula) were over-represented. RIP may have fallen with a greater preponderance into the lake waters by gravity, whereas BIS, and M were transported by wind above the canopy, and also by water in the case of BIS. Both types correspond to plants that lived at a short distance from the lake; the intensity of palynological deposition decreases the further away a plant is from the deposition point, resulting in a leptokurtic distribution of miospores [19], [31] and [32]. High production and buoyancy may account for the high levels of BIS (mainly Pinus; Fig. 4.5) in almost all the samples studied.

The composition of miospore assemblages was also affected by the Ferguson–Spicer effect [31] and [32]. According to Tauber [65] and Holmes [38], miospores deposited on soils can be later refloated by surface runoff and carried to lakes. Shore vegetation can bring a large contribution to the refloated miospore count [29]. Miospores from riparian plants in particular can occur in disproportionately large numbers in lacustrine sediments [57]. The conspicuous percentages of RIP (mainly represented by Alnus) in the samples from the Sanavastre and Sampsor mines may be explained by this (Fig. 2). The same process may have led to the high percentages of HH in sample SAN5 (Fig. 2.1). The low percentages of TC in the mines may indicate the low importance of this group in riparian communities (Fig. 2). Members of the Taxodiaceae and Cupressaceae may have been integrated into coniferous and mesophytic forests, as the presence of Sequoia-type pollen (Fig. 4.4) and the megaremains of Cryptomeria and Cupressaceae suggest [13].

As in extant mesophilous communities [4], the trees of the La Cerdaña forests showed different pollen productivities. Taxa such as Tilia (Fig. 4.9) and Acer, whose common presence in the basin is indicated by megaremains [9], are under-represented in pollen diagrams, since they are entomophilous. These pollen grains may have arrived in the lake by (i) surface runoff, (ii) the direct fall of inflorescences and flowers [9] and [55], or (iii) by carriage on insects such as bees and bibionids [5] and [48]; the pollen could have reached the lake when these insects become trapped on the surface of the water.

Other factors can affect the recruitment of miospores in a lacustrine basin, such as the morphology of the lake basin, the dominant winds of the region, the area's topography, the ecological and spatial structure of the plant communities [15] and [43], and phenomena such as fires. Palaeobotanical and sedimentological studies have inferred the occurrence of fires in the basin [9] and [44] from taphofacies containing burned plant remains [44]. Accumulations of charcoal are certainly seen in the deep lacustrine sediments. The percentages of M and undetermined miospores in samples such as TOV5 and RIU also suggest that fires occurred (Fig. 3); conifers are rapidly affected by fire, and they remain unable to produce pollen for longer periods than mesophilous plants do.

Another phenomenon that can distort the composition of thanatocoenoses is resedimentation [27] and [40]. Annual variations in water density produce seasonal circulation patterns that resuspend pollen from the surface sediments, even from deep areas of lakes. Studies of similar deposits in recent basins suggest that only 20% of the pollen in the lake water represents new input, while some 80% is redeposited [27]. However, in the present study, it was impossible to differentiate between newly and redeposited miospores.

 Biostratinomy

The miospore assemblages of the alluvial sediments of both mines were mainly represented by local and extra-local elements (sensu Jacobson and Bradshaw [40] and Prentice [57]), and mostly transported by water or wind above the canopy. A portion of the wind-transported pollen, i.e. that produced by genera such as Pinus (Fig. 4.5), Cathaya, Sciadopytis, Quercus (Fig. 4.14), and Arecaceae, was probably of regional origin. In contrast, the assemblages from the deep lacustrine facies showed higher extra-local to regional rather than local components, indicated by the proportions of BIS and M. Thus, the presence of taxa such as Pinus and Quercus reflects a more regional input than the presence of Fagus, Acer, Tilia (Fig. 4.9), and many herbs [3], [19], [31] and [40], which may have come from the area that Pfefferkorn [53] refers to as the ‘uplands’.

In the present assemblages, extra-regional pollen deposited far from the area of its production was recorded, including a very small number of pollen grains of the palm Nypa (Fig. 4.10). This palm typically grew in the mangrove swamps that developed along Tethysian coastlines during the Palaeogene [54]. The presence of its pollen grains in the La Cerdaña basin corroborates the existence of Indo-Pacific-type Upper Miocene mangroves in the Palaeomediterranean.

Taking into account the different miospore transport processes possible, the thanatocoenoses of both the alluvial and the deep lake environments of the La Cerdaña basin might be termed allocoenoses (sensu Ochev [50]), as they result from the accumulation of allochthonous remains at the sites of their production. However, they are in fact mixocoenoses, since they show high proportions of algal remains such as Botryococcus colonies and zygnematalean spores, which were surely buried at the site of their production. The miospore assemblages from the alluvial fan sediments would have been produced by both parautochthonous and allochthonous elements, whereas those of the deep lake sediments would have mainly come from allochthonous elements. The term ‘parautochthonous’ refers to remains that have been transported only over a small distance toward the deposition site [17]. This is the case of P, RIP and HH, TC, NAP, as well as that of the Sapotaceae, which may also have been produced near the lake. The presence of a large number of parautochthonous taxa in the sediments of the studied mines confirms the evidence provided by megaremains in other studies [9] and [44]. Remains moved from their site of production, and therefore away from their original habitat, are considered as ‘allochthonous’ [17]. The miospores in BIS and M and some taxa of TC, NAP and other pollen, such as that of Arecaceae, may be allochthonous in nature.

 Sedimentological features

The lack of a relationship between the proportion of different palynological groups and the lithology of the levels has no satisfactory explanation, although the small size of the delta may have conditioned this. Miospores produced closest to the shore would be deposited in similar amounts in both lignites and lutites. The palaeobotanical data clearly indicate that the delta area was covered by swamp vegetation – riparian alder forests, with a fern undergrowth of Osmundaceae. There are no conclusive studies confirming the presence of Taxodium in the basin. The lignite seams are probably autochthonous, but mostly formed from parautochthonous plant remains [46].

Miospore preservation in the deltaic deposits of the La Cerdaña basin is however related to the type of rock. The lignites showed pollen grains with semi-destroyed exines, whereas spores were well preserved. This differential preservation is due to the high resistance of spores to diagenesis in organic matter-rich deposits (see [40]). The destruction of pollen exines is related to their lower resistance to the humification, which occurs during coalification. The same has been recorded in Oligocene lignites, where the spores of pteridophytes are dominant [22]. In the Sampsor mine, the lignite levels showed conspicuous quantities of undetermined pollen grains (Fig. 2.2) and low miospore concentrations. In contrast, the lignite levels of the Sanavastre mine had no large numbers of undetermined pollen grains (Fig. 2.1) and high miospore concentrations [9]. No satisfactory explanation can be given for this, though the lignites of the studied mines might be of different sedimentological (the siliciclastic rocks of the Sampsor mine are coarser than those of the Sanavastre mine are) and diagenetic history. The lutites showed the best-preserved palynological assemblages. The different preservation of the miospores in the lutites and lignites corresponded with their description in palynofacies 1 and 2 (respectively), as published by other authors [46].

The sands preserved the miospores less well than either the lutites or lignites (see [9], [11] and [12]). The presence of palynomorphs in sandy levels is not common [36] and is related to the coarseness of the deposits; this was reflected in the sandy levels of the Sampsor mine (SAM4, SAM8, SAM20, SAM22, SAM24, SAM26, SAM28, SAM30), and level SAN22 of the Sanavastre mine. In contrast, palynomorphs were always present in the fine-grained sands of levels SAN3, SAN9, and SAN14 (Fig. 2.1), and SAM12, and SAM27 (Fig. 2.2).

The diatomitic sediments of the deep lacustrine facies showed the best-preserved miospore assemblages of the entire basin, a consequence of their rapid burial in an anoxic environment. All levels were characterised by the occurrence of Botryococcus colonies. In these rocks, plant megaremains were abundant and included Equisetum, Populus, and Lauraceae, which were not recorded via their miospores (see [9] and [13]). The lack of miospores of these taxa is related to their low sporopollenin content [66]. These sediments also had the highest miospore concentrations of the basin, which can be explained by an increase in the pollen influx associated with the rate of sedimentation and the action of winds in an open environment. The mean miospore concentration was significantly greater in the eastern deep lacustrine sector, which experienced a higher rate of sedimentation than the western sector.

The variation in the proportions of the palynological groups in the miospore assemblages does not allow the differentiation of palynofacies 3 and 4 described by Martín-Closas et al. [46]. These levels are characterised by high percentages of BIS (Fig. 3). The miospore associations present depend on (1) the action of fires (samples TOV5 and RIU), (2) the leaching of organic matter, and (3) changes in the vegetation due to climatic factors. The action of fires leads to reduced miospore concentrations. Therefore, levels TOV5 and RIU, though they showed remarkable values in terms of grains/g of rock [9], showed lower miospore concentrations than did other levels from the deep lacustrine facies (except for the Coll de Saig outcrop). The leaching of organic matter might also explain the low miospore concentrations of samples COL2 and COL6 from the Coll de Saig outcrop (Fig. 4.2). The sediments here showed a yellowish colour and had a lower proportion of organic matter than the rest of the outcrops studied. This might be due to a greater exposure of the outcrop to meteoric modifications [8]. Only two out of the seven samples collected from the Coll de Saig outcrop showed a palynological content, characterised by very high percentages of BIS (>80%) (Fig. 3.2).

In the western sector, the contrasting proportions of the palynological groups shown by levels BEL2 and BEL3, along with their high absolute miospore concentrations, may indicate environmental conditions that favoured M and RIP over BIS (Fig. 3.1).

 Acknowledgements

We would like to thank Dr. D. Néraudeau for inviting us to publish in this issue, as well as for his interesting opinions. We acknowledge the expert assistance of Dr C. Martín-Closas; his contribution greatly improved this paper. We also thank Dr C. Diéguez, Dr C. Arias and D. Peyrot for their constructive comments, and Dr A. Moguilevsky and A. Burton for their language review.

 Agustí J. Oms O. Furió M. Pérez-Vila M.-J. Roca E. The Messinian terrestrial record in the Pyrenees: The case of Can Vilella (Cerdanya Basin) Palaeogeogr. Palaeoclimatol. Palaeoecol. 238 2006 179–189 Anadón P. Cabrera L. Julià R. Roca E. Rosell L. Lacustrine oil-shale basins in Tertiary grabens from NE Spain (western European rift system) Palaeogeogr. Palaeoclimatol. Palaeoecol. 70 1989 7–28 Andersen S.T. The differential pollen productivity of trees and its significance for the interpretation of a pollen diagram from a forested region Birks H.J.B. West R.G. Quaternary Plant Ecology 1973 Blackwell Scientific Publications London 109–115 Andersen S.T. Influence of climatic variation on pollen season severity in wind-pollinated trees and herbs Grana 19 1980 47–52 Arillo A. Bremond C.S. Barrón E. La entomofauna fósil del Mioceno Superior de la Depresión Ceretana. Dípteros Alemany A. Historia Natural 91 1992 Universitat de les Illes Balears Palma de Mallorca, vol. 2 553–561 Astre G. Le bassin néogène de Bellver Bull. Soc. Hist. Nat. Toulouse 56 1927 231–258 Baltuille J.M. Becker-Platen J.D. Benda L. Ivanovic Calzaga Y. A contribution to the subdivision of the Neogene in Spain using palynology Newslett. Stratigr. 27 1992 41–57 Barrón E. Taphonomic studies of the plant remains from the Ceretana Basin (Lérida, Spain) Kaupia 2 1993 127–132 E. Barrón, Estudio tafonómico y análisis paleoecológico de la macro y microflora miocena de la cuenca de la Cerdaña, Tesis Doctoral, Universidad Complutense de Madrid, 1996a (773 p.). Barrón E. Sesgos en la distribución de palinomorfos en el registro fósil. Ejemplos del lago mioceno de la Cerdaña (España) Comunicación de la II reunión de Tafonomía y fosilización Zaragoza, Spain 1996 47–54 Barrón E. Estudio palinológico de la mina de lignito vallesiense de Sampsor (La Cerdaña, Lérida, España) Rev. Esp. Paleontol. 12 1997 91–101 Barrón E. Estudio palinológico de la mina de lignito vallesiense de Sanavastre (La Cerdaña, Lérida, España) Rev. Esp. Micropaleontol. 29 1997 139–157 Barrón E. Diéguez C. Cuticular study of leaf compressions from the Late Miocene (Vallesian) diatomites of Cerdaña Basin (Eastern Pyrenees, Spain) N. Jahrb. Geol. Palaeontol. Abh. 237 2005 61–85 Barrón E. Arillo A. Diéguez C. El Mioceno Superior de la Cuenca de la Cerdaña (Pirineos orientales) Temas Geol.-Min. 26 1999 63–69 Barthélemy L. Réflexions sur la répartition du pollen. Conséquences pour l’Archéologie Renault-Miskovsky J. Bui-thi M. Girard M. Palynologie archéologique 1985 Éditions du Centre national de la recherche scientifique Paris 53–86 Batten D.J. Small palynomorphs Jones T.P. Rowe N.P. Fossil plants and spores: modern techniques 1999 Geological Society London 15–19 Behrensmeyer A.K. Hook R.W. Paleoenvironmental contexts and taphonomic modes in the terrestrial fossil record Behrensmeyer A.K. Damuth J.D. DiMichele W.A. Potts R. Sues H.-D. Wing S.L. Terrestrial Ecosystems Through Time. Evolutionary paleoecology of terrestrial plants and animals 1992 University of Chicago Press Chicago 15–38 Bessedik M. Reconstitution des environnements miocènes des régions nord-ouest mediterranéennes à partir de la palynologie 1985 université des sciences et techniques du Languedoc, Académie de Montpellier (162 p.) Bonny A.P. Recruitment of some pollen to the seston and sediment of some Lake District lakes J. Ecol. 64 1976 859–887 Bonny A.P. The effect of pollen recruitment processes on pollen distribution over the sediment surface of a small lake in Cumbria J. Ecol. 66 1978 385–416 Brooks J. Shaw G. Identity of sporopollenin with older kerogen and new evidence for the possible biological source of chemicals in sedimentary rocks Nature 220 1968 678–679 Bruch A.A. Mosbrugger V. Palaeoclimate versus vegetation reconstruction – palynological investigations on the Oligocene sequence of the Sava Basin, Slovenia Rev. Palaeobot. Palynol. 122 2002 117–141 Cabrera L. Roca E. Santanach P. Basin formation at the end of a strike-slip fault: the Cerdanya Basin (eastern Pyrenees) J. Geol. Soc. Lond. 145 1988 261–268 Chaloner W.G. The Paleoecology of fossil spores Drake E.T. Evolution and Environment 1968 Yale University Press New Haven and London 125–138 Chaloner W.G. Muir M. Spores and Floras Marendson L. Westord J. Coal and coal-bearing strata 1968 Oliver & Boyd Edinburgh, UK 127–146 Cour P. Nouvelles techniques de détection des flux et des retombées polliniques : étude de la sédimentation des pollens et des spores à la surface du sol Pollen Spores 16 1974 103–141 Davis M.B. Pollen grains in lake sediments: Redeposition caused by seasonal water circulation Science 162 1968 796–799 Faegri K. The preservation of sporopollenin membranes under natural conditions Brooks J. Grant P.R. Muir M. van Gijzel P. Shaw G. Sporopollenin 1971 Academic Press New York, and London 256–270 Faegri K. Iversen J. Textbook of pollen analysis 4th edition 1989 John Wiley & sons Chichester, UK (328 p.) Ferguson D.K. The origin of leaf-assemblages – New light on an old problem Rev. Palaeobot. Palynol. 46 1985 117–188 Ferguson D.K. Plant taphonomic studies with special reference to Messel Kaupia 2 1993 117–126 Ferguson D.K. Plant part processing and community reconstruction Eclog. Geol. Helv. 88 1995 627–641 Ferguson D.K. Knobloch E. A fresh look at the rich assemblage from the Pliocene sink-hole of Willershausen, Germany Rev. Palaeobot. Palynol. 101 1998 271–286 Gastaldo R.A. Riegel W. Puttmann W. Linnemann U.G. Zetter R. A multidiscipinary approach to reconstruct the Late Oligocene vegetation in central Europe Rev. Palaeobot. Palynol. 101 1998 71–94 Guennel G.K. Fossil spores of the Allegheinian coals in Indiana Indiana Geol. Surv. Rep. Prog. 4 1952 1–40 Havinga A.J. Palynology and pollen preservation Rev. Palaeobot. Palynol. 2 1967 81–98 Havinga A.J. A 20-year experimental investigation into the differential corrosion susceptibility of pollen and spores in various soil types Pollen Spores 26 1984 541–557 Holmes P.L. Differential transport of spores and pollen: a laboratory study Rev. Palaeobot. Palynol. 64 1990 289–296 Hopkins J.S. Differential floration and deposition of coniferous and deciduous tree pollen Ecology 31 1950 633–641 Jacobson G.L. Bradshaw R.H.W. The selection of sites from paleovegetation studies Quat. Res. 16 1981 80–96 Jelgersma S. Investigaciones palinológicas de lignitos terciarios procedentes de Cerdaña y del Valle de Arán (Pirineos españoles) Cursillos Conf. Inst. Lucas Mallada 4 1957 159–162 R. Julià, Síntesis geológica de la Cerdanya, Girona, in : CSIC (Ed.), El borde mediterráneo español : origen del erógeno bético y geodinámica de las depresiones neógenas, Consejo Superior de Investigaciones Científicas, Granada, Spain, 1984, pp. 95–98. Lehman J.T. Reconstructing the rate of accumulation of lake sediment: the effect of sediment focusing Quat. Res. 5 1975 541–550 Martín-Closas C. Plant Taphonomy of La Cerdanya Basin (Vallesian, Eastern Pyrenees) Geobios, M.S. 18 1995 287–298 Martín-Closas C. Gomez B. Taphonomie des plantes et interprétations paléoécologiques. Une synthèse Geobios 37 2004 65–88 Martín-Closas C. Permanyer A. Vila M.-J. Palynofacies distribution in a lacustrine basin Geobios 38 2005 197–210 Menéndez Amor J. La depresión ceretana española y sus vegetales fósiles. Característica fitopaleontológica del Neógeno de la Cerdaña española Mem. R. Acad. Cien. Exact. Fís. Nat. Madrid, Ser. Cien. Nat. 18 1955 1–232 Nel A. Martínez-Delclòs X. Arillo A. Peñalver E. A review of the Eurasian fossil species of the bee Apis Palaeontology 42 1999 243–285 Neves R. Upper Carboniferous plant spore assemblages from Gastrioceras subcrenatum horizon, North Staffordshire Geol. Mag. 95 1958 1–19 Ochev V.G. Types of thanatocoenoses and burial patterns of terrestrial organisms Paleontol. J. 27 1993 127–136 Peck R.M. Pollen budget studies in a small Yorkshire catchment Birks H.J.B. West R.G. Quaternary Plant Ecology 1973 Blackwell Scientific Publications London 43–60 W. Pennington (Mrs. T.G. Tutin), The origin of pollen in lake sediments: an enclosed lake compared with one receiving inflow streams, New Phytol. 83 (1979) 189–213. Pfefferkorn H.W. A note on the term ‘upland flora’ Rev. Palaeobot. Palynol. 30 1980 157–158 Plaziat J.-C. Cavagnetto C. Koeniguer J.-C. Baltzer F. History and biogeography of the mangrove ecosystem, based on the critical reassessment of the paleontological record Wetland Ecol. Manage. 9 3 2001 161–179 Postigo J.M. Barrón E. Morla C. Inflorescences of the subclase Commelinidae (Liliatae, Magnoliophyta) in the Neogene of the Iberian Peninsula N. Jahrb. Geol. Palaeontol. Monatsh. 2003 229–243 Pous J. Julià R. Sole-Sugranes L. Cerdanya basin geometry and its implication on the Neogene evolution of the Eastern Pyrenees Tectonophysics 129 1986 355–365 Prentice C. Records of vegetation in time and space: The principles of pollen analysis Huntley B. Webb T. III Vegetation History 1988 Kluwer Academia Dordrecht, The Netherlands 17–24 Rérolle M.L. Études sur les végétaux fossiles de Cerdagne Rev. Sci. Nat. Montpellier, 3e ser. 4 1884–1885 167–191 252–298, 368–386 E. Roca, Estudi geologic de la Fossa de la Cerdanya, Tesis de Licenciatura, Departament de Geomorfologia i Tectònica, Facultat de Geologia, Universitat de Barcelona, Barcelona, 1986, (109 p., unpublished). Roca E. Santanach P. Génesis y evolución de la fosa de la Cerdanya (Pirineos orientales) Geogaceta 1 1986 37–38 Roca E. The Neogene Cerdanya and Seu d’Urgell intramontane basins (eastern Pyrenees) Friend P.F. Dabrio C. Tertiary basins on Spain. The stratigraphic record of crustal kinematics 1996 Cambridge University Press Cambridge 114–119 Roca E. Estructura, estratigrafía y evolución tectonosedimentaria de las cuencas neógenas de la Cerdanya y Seu d’Urgell (Pirineos orientales) Vera J.A. Geología de España 2002 SEG-IGME Madrid 573–576 Schopf J.M. Modes of fossil preservation Rev. Palaeobot. Palynol. 20 1975 27–53 Spicer R.A. The formation and interpretation of plant fossil assemblages Adv. Bot. Res. 16 1989 95–191 Tauber H. Investigations on the mode of pollen transfer in forested areas Rev. Palaeobot. Palynol. 3 1967 277–286 Traverse A. Paleopalynology 1988 Unwin Hyman Boston (600 p.) Watts W.A. Hansen B.C.S. Pre-Holocene and Holocene pollen records of the vegetation history from the Florida peninsula and their climatic implications Palaeogeogr. Palaeoclimatol. Palaeoecol. 109 1994 163–176

Geological context of La Cerdaña basin modified from Roca [61]. The location of the studied outcrops is shown. (1) Opencast Sanavastre lignite mine. (2) Opencast Sampsor lignite mine. (3) Bellver de Cerdanya outcrop. (4) Barranc de Salanca outcrop. (5) Torrent de Vilella outcrop. (6) Coll de Saig outcrop. (7) Prats outcrop. (8) Riu de Santa María outcrop. (9) Baltarga outcrop. (10) Beders outcrop. (11) Torrent de la Bavosa outcrop.

Fig. 1. Contexte géologique du bassin de Cerdagne, modifié d’après Roca [61]. La localisation des affleurements est signalée. (1) Mine de lignite à ciel ouvert de Sanavastre. (2) Mine de lignite à ciel ouvert de Sampsor. (3) Gisement de Bellver de Cerdanya. (4) Gisement de Barranc de Salanca. (5) Gisement de Torrent de Vilella. (6) Gisement de Coll de Saig. (7) Gisement de Prats. (8) Gisement de Riu de Santa María. (9) Gisement de Baltarga. (10) Gisement de Beders. (11) Gisement de Torrent de la Bavosa.

Proportions of the different palynological groups in the La Cerdaña basin at (1) the Sanavastre mine (SAN), (2) the Sampsor mine (SAM).

Fig. 2. Pourcentages des différents groupes palynologiques définis pour le bassin de Cerdagne : (1) mine à ciel ouvert de lignite de Sanavastre (SAN), (2) mine à ciel ouvert de lignite de Sampsor (SAM).

Proportions of the different palynological groups in the La Cerdaña basin. (1) Western lacustrine sector. BEL: Bellver de Cerdanya outcrop, BAR: Barranc de Salanca outcrop, TOV: Torrent de Vilella outcrop. (2) Eastern lacustrine sector. COL: Coll de Saig outcrop, PRAT: Prats outcrop, RIU: Riu de Santa María outcrop, BAL: Baltarga outcrop, BED: Beders outcrop, TOB: Torrent de la Bavosa outcrop.

Fig. 3. Pourcentages des différents groupes palynologiques définis pour le bassin de Cerdagne. (1) Secteur lacustre ouest. BEL : Affleurement de Bellver de Cerdanya ; BAR : affleurement de Barranc de Salanca ; TOV : Affleurement de Torrent de Vilella. (2) Secteur lacustre est. COL : affleurement de Coll de Saig ; PRAT : affleurement de Prats ; RIU : affleurement de Riu de Santa María ; BAL : affleurement de Baltarga ; BED : affleurement de Beders ; TOB : affleurement de Torrent de la Bavosa.

(1) Photograph of the Sampsor mine taken in 1995. (2) Frontal view of the diatomiferous outcrop of Coll de Saig. (3) Tetrad of Osmunda, the most abundant trilete spore in the sediments of the La Cerdaña basin (included in palynological group P), Sanavastre lignite mine, level SAN7. (4) Taxodiaceae (Sequoia pollen type; palynological group TC), Barranc de Salanca outcrop, level BAR4. (5) Pinus, the most common pollen of the BIS palynological group, Torrent de Vilella outcrop, level TOV2. (6) Pollen produced by herbs of the family Apiaceae (NAP palynological group), Barranc de Salanca outcrop, level BAR1. (7) Alnus, the most abundant pollen of the RIP palynological group, Torrent de Vilella outcrop, level TOV6. (8) Saxifraga, entomophilous pollen related to the NAP palynological group, Torrent de Vilella outcrop, level TOV6. (9) Tilia, entomophilous pollen related to the M palynological group, Barranc de Salanca outcrop, level BAR1. (10) Extra-regional pollen of Nypa defined as belonging to ‘other miospores’, Sampsor mine, level SAM3. (11) Sparganium pollen produced by hygrophilous herbaceous plants (palynological group HH), Sanavastre mine, level SAN5. (12) Engelhardia pollen produced by subtropical/tropical trees and consequently considered to belong to ‘other miospores’, Sampsor mine, level SAM10. (13) Trapa pollen produced by aquatic (hydrophilous) plants (palynological group HH), Sanavastre mine, level SAN2. (14) Quercus, anemophilous pollen of the M palynological group, Beders outcrop, level BED. Bar = 10 μm.

Fig. 4. (1) Aspect général de la mine de lignite à ciel ouvert de Sampsor, en 1995. (2) Vue frontale du gisement à diatomites de Coll de Saig. (3) Tétrade d’Osmunda, la spore trilète la plus abondante dans les sédiments du bassin de Cerdagne, incluse dans le groupe palynologique P, mine à ciel ouvert de lignite de Sanavastre, échantillon SAN7. (4) Taxodiaceae (type Sequoia) du groupe palynologique TC, gisement de Barranc de Salanca, échantillon BAR4. (5) Pinus, grain de pollen le plus fréquent du groupe palynologique BIS, gisement de Torrent de Vilella, échantillon TOV2. (6) Grain de pollen produit par des herbes appartenant à la famille des Apiaceae (groupe palynologique NAP), gisement de Barranc de Salanca, échantillon BAR1. (7) Alnus, le grain de pollen le plus abondant du groupe palynologique RIP dans le bassin de Cerdagne, gisement de Torrent de Vilella, échantillon TOV6. (8) Saxifraga, grain de pollen de plante entomophile rattaché au groupe palynologique NAP, gisement de Torrent de Vilella, échantillon TOV6. (9) Tilia, grain de pollen de plante entomophile rattaché au groupe palynologique M, gisement de Barranc de Salanca, échantillon BAR1. (10) Grain de pollen extrarégional de Nypa inclus dans le groupe dénommé « autres miospores », mine de lignite à ciel ouvert de Sampsor, échantillon SAM3. (11) Sparganium, grain de pollen produit par des plantes herbacées hygrophiles (groupe palynologique HH), mine à ciel ouvert de lignite de Sanavastre, échantillon SAN5. (12) Engelhardia, grain de pollen produit par des arbres tropicaux/subtropicaux inclus dans le groupe dénommé « autres miospores », mine de lignite à ciel ouvert de Sampsor, échantillon SAM10. (13) Trapa, grain de pollen produit par des plantes aquatiques (hydrophiles) (groupe palynologique HH), mine de lignite à ciel ouvert de Sanavastre, échantillon SAN2. (14) Quercus, grain de pollen anémophile du groupe palynologique M, gisement de Beders, échantillon BED. Barre d’échelle = 10 μm.