Réf. David 1997 - A

Référence bibliographique complète

DAVID, F. 1997. Holocene tree limit history in the northern French Alps stomata and pollen evidence. Review of Palaeobotany and Palynology, Vol. 97, 3–4, 227–237. [PDF]

Abstract: Certain aspects of the Holocene tree limit history in the northern French Alps are inferred from both pollen and stomata analysis conducted in a subalpine marsh (2080 m a.s.l.). The proposed chronology is based on 14C AMS dating of terrestrial plant macrofossils. In the area, Betula invaded this altitudinal level at the beginning of the Holocene prior to 9100 yr B.P., whereas the first occurrence of Pinus stomata is recorded much later at around 7000 yr B.P. Afterwards five periods of stomata accumulation are recorded (6480-6130, 5600-4990, 4730-4470, 3200-2870 and 2680-2300 yr B.P.). The low pollen influx in the first four periods and the occurrences of Artemisia at the beginning of these periods indicate that stomata accumulation resulted from a reduction of the tree cover. The climatic and/or anthropogenic origin of these stomata accumulations is discussed. A good correlation appeared with five of the nine phases of Holocene climatic deterioration described in the Swiss and Austrian Alps. In this study, the combination of pollen influx and stomata determination proved to be a useful tool for understanding the tree limit history. Moreover, discrepancies in the occurrence of Pinus in an adjacent area imply it is difficult to find universal indicators of tree line fluctuations even in a small area. This warrants methodological discussion.

Résumé : L'étude du contenu en pollen et stomates d'un petit site de l'étage subalpin asylvatique de la zone intermédiaire des Alpes françaises du nord permet de proposer un schéma de l'évolution holocène de la limite supérieure des arbres dans cette aire et une chronologie basée sur 8 datations AMS de macrorestes végétaux terrestres. Le bouleau colonise ce niveau altitudinal dès le début de l'holocène (avant 9100 yr B.P.) et précède les pins dont les premiers stomates sont enregistrés autour de 7000 yr B.P. Cinq niveaux d'accumulation de stomates sont ensuite enregistrés jusqu'aux déboisements de la période romaine (6480-6130, 5600-4990, 4730-4470, 3200-2870, 2680-2300 yr B.P.). La diminution des flux polliniques dans 4 de ces niveaux et la réapparition d'Artemisia au début de ces périodes soulignent la relation entre ouverture du milieu et accumulation de stomates. L'origine climatique ou anthropique de ces phases est discutée et une correlation apparaît avec 5 des neuf phases holocènes de péjoration climatique décrites dans les Alpes suisses et autrichiennes. La combinaison des comptages de pollen et de stomates est un bon outil pour décrire les fluctuations de la limite des arbres si l'on considère les flux polliniques et non les seules fréquences relatives. Cependant les différences chronologiques entre régions dans l'apparition des pins soulignent la difficulté à définir des indicateurs universels des fluctuations de la limite des arbres.


Holocene tree limit - Stomata - Pollen influx - AMS dates


Organismes / Contact


(1) - Paramètre(s) atmosphérique(s) modifié(s)

(2) - Elément(s) du milieu impacté(s)

(3) - Type(s) d'aléa impacté(s)

(3) - Sous-type(s) d'aléa






Pays / Zone

Massif / Secteur

Site(s) d'étude



Période(s) d'observation








(1) - Modifications des paramètres atmosphériques










Informations complémentaires (données utilisées, méthode, scénarios, etc.)



(2) - Effets du changement climatique sur le milieu naturel


Results: Vegetation dynamics and the upper tree limit

The six pollen zones correspond to six major stages of the Holocene vegetation history at the same altitudinal level in the area (Ponel et al., 1992; David, 1995). Nevertheless, some comment can be given. In pollen zone M 1, high percentages of Pinus cannot be interpreted as Pinus stand. The presence of steppic herbs indicates that high Pinus percentages probably represent Pinus over-representation in poorly pollen productive alpine grassland (Markgraf, 1980). Comparison of the cumulative flux of the different taxa illustrates very well the vegetation history (ecological succession) and supports this interpretation. Dominance of each taxon is clearly in this order: Artemisia, Juniperus, Betula, Pinus, Alnus and then again Juniperus.

At the beginning of the Holocene, the Plan des Mains area was a dwarfshrub tundra. The oldest Betula remains were dated at 9100 yr B.P. Betula spreading in alpine grassland coincided with changes in grass communities. Steppic herbs (e.g. Artemisia) and Selaginella selaginoides receded to the benefit of tall herbs (e.g. Apiaceae). The first occurrence of Pinus is shown by the presence of stomata at around 7000 yr B.P. In the lower belt, Abies, present since 8000 yr B.P., has its maximum expansion. No Abies macroremains were found at 2080m. Nevertheless, although the high Abies percentage (30%) is not of the greatest importance, we can assume the proximity of trees as demonstrated for the same altitudinal level in the northern French Alps, Taillefer Plateau (Ponel et al., 1992) and Beaufortin (David, 1993a). The limit between Abies and Pinus stands fluctuated in altitude depending on local parameters (e.g. topography, exposure) during the climatic optimum.

In pollen zone M5, the higher Pinus percentage (80%) and the periodic occurrence of stomata until 2400 yr B.P. suggest permanent Pinus stands until deforestation by man. Pollen flux shows more significant variations except for the period 3800 3500yr B.P. when the Pinus percentage decreased significantly. After 2400 yr B.P. Pinus reached its lowest percentages in less than 200 years (Fig. 2, M6). A dwarfshrub tundra reappeared with a higher representation of Juniperus, Cyperaceae, Cichoroideae, Rosaceae, Selaginella selaginoides and Plantago lanceolata. The absence of stomata confirms that the area was above the upper tree limit as at present time, even during the period 1250-1100 yr B.P., when Pinus values reached 60% and Cyperaceae showed low values. After the regression of the Cyperaceae cover, the increase of all taxa influx indicates a better incorporation of pollen in the sediment.

Interpretation and discussion

1. Origin of stomata accumulation

Comparison of Pinus pollen and stomata phases show relatively low pollen influx corresponding to the five first stomata accumulation periods. Six phases can be distinguished: around 7000, 6480-6130, 5600-4990, 4730-4470, 3200-2870 and 2680 2300 yr B.P.

Artemisia occurrences briefly preceded the stomata periods: 6660, 5680, 4790 and 3200 yr B.P. Artemisia is not only an alpine pollen type but it may also be a long distance pollen (El-Moslimany, 1987). Though no distinction was made between the different Artemisia species, this occurrence led us to interpret these stomata phases as a result of a landscape opening. Small peaks of Juniperus and/or Salix during stomata periods support this interpretation.

During the Holocene, the opening of the vegetation cover is mainly attributed to climatic and/or human impact. In the area different studies have shown a human impact between 5000 and 2500 yr B.P. (David, 1993a,b; Wegmtiller, 1975, 1977). In a nearby valley, a recent pedoanthracological study dated a significant fire phase at around 3800 yr B.P. (Carcaillet and Thinon, 1996). We have to consider the human impact hypothesis particularly for the two last stomata periods. Nevertheless, if we compare the chronology of stomata accumulation, glacier advances, and lake level rise, it appears that the periodicity of the three phenomena coincides.

In detail, the first long period of stomata accumulation (6480-6130 yr B.P.) coincides with the bipartite phase of glacier advance observed in the eastern Alps (Frosnitz) and the western Alps (Misox) between 6500 and 6000 yr B.P. (Zoller, 1977; Bortenschlager, 1978, 1982). However, the real beginning of this glacier advance remains doubtful. In the western Alps, a 300-year period of glacier advance was recorded (7100-6800 yr B.P.) which corresponds with the first brief occurrence of stomata at around 7000 yr B.P. The Cerin lake level rise shows a same bimodal picture between 6800 and 6100yr B.P. (Magny, 1995).

The next long period of stomata accumulation (5590-4490 yr B.P.) can be linked with the shorter period of glacier advance Rotmoos I (Piora I) (5300-4400 yr B.P.). The rise of Grand Maclu lake level took place within a similar time interval (500 years).

The following stomata period (4730-4470 yr B.P.) can be linked with Rotmoos II and Piora II which appeared to be longer (5300-4400 yr B.P.). This also partly fits with the first part of the complex and longer Chalain Phase.

The two last stomata periods (3200-2870 yr B.P.) and (2680-2,300 yr B.P.) cover the bipartite G6schenen 1 phase in the western Alps but appear out of phase with glacier advances in the eastern Alps and lake level rises in the western Alps (Bourget). Moreover, no stomata were recorded during the L6bben period (3500-3200 yr B.P. in the eastern Alps) corresponding to Pluvis lake level rise. This glacier advance appeared very briefly, within less than hundred years, in the western Alps.

The climatic origin of glacier movements are well established (Le Roy Ladurie, 1967; Bortenschlager, 1978; Holzhauser, 1984; R6thlisberger, 1986; Pfister, 1988). However, if we assume the same climatic cause for stomata accumulation, this does raise further questions.

The presence of stomata in the sediment results from leaf fall and decomposition before or after transport. Irregular stomata accumulation could result from either considerable leaf fall of overhanging trees due to a rough mortality or litter transfer into the catchment. Irregularity of this stomata accumulation excludes a continual leaf fall in this site and the good preservation of pollen excludes an in situ decomposition of organic matter. This invokes litter transport into the catchment but does not exclude destruction of tree cover that protected the soils from the weather, thus allowing a better litter influx. Occurrences of Artemisia pollen prior to stomata occurrence and the accumulation of undifferentiated organic rich material support this idea and involve a time lag between climatic change and stomata incorporation. Moreover, the increase of pollen flux during Artemisia occurrences around 5600, 4900 and 3200 yr B.P. might correspond to the first indication of litter leaching, which facilitates water transport of organic matter, like pollen, into the catchment. The duration of the leaching period before a strong supply of litter has been estimated at between 150 and 200 calendar years. One must remember that such a calculation involves errors that could only be corrected by a great number of AMS datings. The calculations reported here must be considered as a rough estimate to initiate further studies.

The last strong stomata supply around 2680 yr B.P. corresponding to an increase of Pinus influx, without premonitory sign (Artemisia) indicates a brutal destruction of the tree cover that allowed a rapid litter transport. Such a feature can be easily attributed to human impact as shown by Plantago among other anthropogenic indicators. The correspondence with the end of the Göschenen 1 phase could suggest a combination between climate deterioration and human impact. Appearance of Artemisia after 5000 yr B.P. can also indicate Neolithic long distance pollen transport. Nevertheless, if we trust the different chronologies and assume the absence of a general and simultaneous periodical human activity throughout the Alps for 5000 years, this study seems to indicate general climatic trends.

2. Selection of the appropriate indicator for tree line fluctuations

In this study, relative pollen frequencies and especially the AP/NAP ratio did not show significant fluctuations, that could have been interpreted as tree recession. Only high NAP percentages in the early and late Holocene indubitably give evidence of a treeless vegetation cover. The composition and the structure of the vegetation cover around the site must be considered. The use of AP/NAP ratio seems adequate in tree limit ecotone. Austrian and Swiss authors have drawn their conclusions based on AP/NAP fluctuations from sites situated at the limit between the alpine and subalpine vegetation belts. At lower altitude, tree stands must have been less sensitive to climatic deterioration owing to the altitudinal variation of climatic parameters and to their own inertia due to clump effect. Thus we can state that during the Holocene "lower sites provide very few records o/ NAP phase compared to the Late-Glacial" (Burga, 1993). During the middle Holocene the Plan des Mains area was clearly below the upper tree limit in the subalpine vegetation belt, which excludes the use of AP/NAP ratio (David, 1995).

Moreover, trees spread earlier in the central Alps and this allows study of the upper tree limit at around 2000 m as early as the Preboreal time. In the northern French Alps, during the early Holocene Betula constituted the upper tree limit between 1600 and 1800 m (David, 1995). At that time, the Plan des Mains area (2080 m) was clearly above the upper tree limit, which also excludes the use of AP/NAP variations for describing the upper tree limit fluctuations. Some of the strong decreases of the Betula influx at around 9100, 8800-8700 and 8300 yr B.P. could be attributed to the Venediger (Schams) oscillation corresponding to the Joux lake level rise (8700-8000 yr B.P.).  Studies of Betula influx at lower sites seem more advisable to describe early Holocene tree line fluctuations as long as no competition between tree taxa existed (David and Barbero, 1995).

At the study site, the selection of a single indicator of tree limit fluctuation appeared impossible. This study strengthens the notion of 'key massif' composed of 'key sites' located at different altitudes in order to follow the climatic impact on vegetation throughout a complete climatic cycle in a mountainous area (David, 1993b). Each site gives a best record for a peculiar climatic event, the lowest sites for the oldest events and the highest sites for the youngest events. At a regional scale, the description of tree line fluctuation with a single indicator proves to be illusive and thus one must combine several methods.



The main purpose of this study is to display Holocene fluctuations of the tree limit. It gives a first chronology of the upper tree limit variations based on AMS dates in the northern French Alps subalpine vegetation belt. The distinction between a climatic or anthropogenic causality for these fluctuations remains difficult to appreciate. Nevertheless, the time interval of 200 years between the opening of the vegetation cover and the litter supply and the correspondence of these events with glacial advances throughout the Alps suggest a general climatic trend. In no case has  the problem concerning the dating and the estimation of the different events been underestimated here, nor is it presented as definitely solved. As a matter of fact, problems concerning the duration of the Venediger phase, the beginning of the Frosnitz phase, the end of the Rotmoos 2 phase or the chronology of the Göschenen 2 phase (Patzelt, 1973; Zoller, 1977; Burga, 1979, 1988; Gamper and Suter, 1982) result either from the absence of radiocarbon dates or from divergences of estimation due to the use of non-homogeneous methods, e.g. fossil soils and woods (Holzhauser, 1984; R6thlisberger, 1986), pollen analysis, glacial mire stratigraphy (Patzelt, 1973; Zoller, 1977; Burga, 1987, 1991).

At least, this study offers some directions for further inquiries and must be considered as a state of art.








Sensibilité du milieu à des paramètres climatiques

Informations complémentaires (données utilisées, méthode, scénarios, etc.)


The research area is located in the subalpine vegetation belt on the eastern slope of Montagne de Cherferie, a N S-oriented range at the northwestern border of the Massif de la Vanoise in the French Alps. The dominant vegetation on the surrounding slopes is a dwarf shrub tundra with Rhododendron ferrugineum, but in this vegetation belt all the steps between Larix decidua-Pinus cembra woods and acidophilous pasture with Nardus stricta can be observed. Above the upper tree limit, a dwarfshrub tundra with Rhododendron exists: Juniperus nana, Vaccinium uliginosum and Empetrum hermaphroditum are well represented as are acidophilous species (e.g. Geum montanum, Potentilla aurea, Plantago alpina, Campanula barbata). Loiseleuria procumbens appears and dominates in the alpine vegetation belt. At the boundary with the mountain belt, Picea abies dominates, accompanied by Sorbus aucuparia. Vaccinium myrtillus, V. vitis idaea, Lonicera nigra, Melampyrum sylvaticum and Filicales (Polystichum spinulosum, Aspidium lonchitis, Dryopteris linnaeana) dominate the lower layer (Gensac, 1967).

A small marsh (15 m in diameter) called Plan des Mains, 2050 m high in altitude, was sampled with a Russian corer. The total thickness of the sediment (detritus peat) was 300 cm. Pollen samples were taken each 2 cm and processed by standard methods. Absolute concentrations were quantified by the volumetric method (Clerc, 1988). A minimum pollen sum of thousand 'land pollen' (TLP) was adopted which excludes pollen of characteristic aquatic plants and all spores. A relative pollen diagram was produced and divided into six pollen zones.

Eight wood samples collected from the core were dated by AMS (Arizona MSF Facility) and four coarser samples were dated at the laboratory of Isotope Geochemistry, Tucson Arizona. The calibrated dates (Stuiver and Reimer, 1993) were used to calculate the rate of sedimentation and annual pollen accumulation. Due to their imprecision, the four decay-counting radiocarbon dates have not been used in these calculations.

In order to compare pollen influx of very different pollen productive taxa, the values of each taxon are expressed in percentage of the cumulated flux of the concerned taxon for the whole sedimentary sequence. This type of representation also permits comparison of pollen flux variation between different sites (David and Barbero, 1995).

Stomata were counted during pollen analysis. Only Pinus stomata were found. Six periods of stomata accumulation were distinguished. The duration of each period was estimated in the light of the AMS dates assuming continuous sedimentation. For easier comparison with previous published data, all the dates mentioned here are uncalibrated dates.


(3) - Effets du changement climatique sur l'aléa










Paramètre de l'aléa

Sensibilité des paramètres de l'aléa à des paramètres climatiques

Informations complémentaires (données utilisées, méthode, scénarios, etc.)





(4) - Remarques générales

Tree limit history: which approach?

Variation in the AP/NAP ratio and occurrences of distinctive upper tree stand taxa were often presented as the main tools for describing upper tree limit variations. This approach was particularly successful in Swiss and Austrian sites at the transition between the subalpine and alpine vegetation belt (2000-2200m) (e.g. Zoller et al., 1966; Bortenschlager and Patzelt, 1969; Bortenschlager, 1972). Nevertheless there are limitations to this approach:

(1) Definition of the ecological transition can be difficult and imprecise. The limit between the subalpine and alpine belt varies in altitude throughout the Alps, depending on local and/or regional factors. Generally, vegetation belts are higher in the internal Alps than in the external Alps (Ozenda, 1985) but human activity often has locally lowered this limit.

(2) In the northern French Alps the zone 2000 2200 m was colonized by trees later in the  Holocene than in the Austrian and Swiss Alps and was clearly situated above the upper tree limit during the late Holocene (David, 1993a).

(3) Over-representation of long distance pollen (e.g. Pinus) in unforested mountain areas has been demonstrated for past and present times by studying sites on altitudinal transects (Fall, 1992; David, 1993b; Kullman, 1991). This over-representation makes it difficult to assess the representation of poorly productive local taxa in relative pollen diagrams.

These three points led the author to search for other indicators for the upper tree limit. Stomata analysis during pollen counting revealed to be a simple and quick method to indicate conifer stands around the sites that were studied. However, a recent synthesis showed little direct relationship between pollen percentages and stomata (Ammann and Wick, 1993). Stomata can be associated with either low or high pollen percentages. In fact, pollen percentage variations are mainly linked to high pollen productive taxa which in the studied sites (high in altitude) are also long distance pollen productive taxa (Pinus). Only the use of pollen influx allows us to trace individual taxa fluctuations and can give a more accurate idea of forest development (Hyvfirinen, 1976; David, 1992, 1993a,b; Hicks, 1994). Thus, a comparison between stomata deposition and pollen data seems to be useful. The calculation of absolute pollen frequency requires sufficient precise datings and a conversion of 14C chronology in calendar chronology. The fluctuation source (climatic, anthropogenic or both) and fluctuation extent (local, regional or global) of the vegetation cover can be evaluated in the light of previous works.

Plant nomenclature is following Flora Europaea.


(5) - Syntèses et préconisations


Références citées :

Hicks, S., 1994. Present and past pollen records of Lapland forests. Rev. Palaeobot. Palynol. 82, 17 35.

Hyväarinen, H., 1976. Flandrian pollen deposition rates and tree-line history in northern Fennoscandia. Boreas 5,163-175.

Ammann, B., Wick, L., 1993. Analysis of fossil stomata of conifers as indicators of the alpine tree line fluctuations during the Holocene. In: Frenzel, B. (Ed.), Oscillations of the Alpine and Polar Tree Limits in the Holocene. Palaeoclim. Res. 9, 164 175.

Bortenschlager, S., 1972. Der pollenanalytische Nachweis von Gletscher- und Klimaschwankungen in Mooren der Ostalpen. Ber. Dtsch. Bot. Ges. 85, 105 137.

Bortenschlager, S., 1978. Ursachen und Ausmass postglazialer Waldgrenzschwankungen in den Ostalpen. In: Frenzel, B. (Ed.), Dendrochronologie und Postglaziale Klimaschwankungen in Europa. Steiner Verlag, Wiesbaden, pp. 260 266.

Bortenschlager, S., 1982. Chronostratigraphic subdivisions of the Holocene in the Alpes. In: Mangerud, J., Bucks, H.J.B., Jager, K.D. (Eds.), Chronostratigraphic Subdivisions of the Holocene. Striae 16, 75-79.

Bortenschlager, S., Patzelt, G., 1969. W~:irnlezeitliche Klimatund Gletscherschwankungen im Pollenprofil eines hochgelegenes Moores 2.270 m der Venedigergruppe. Eiszeitalter Ggw. 20, 116 122.

Burga, C.A., 1979. Postglaziale Klimaschwankungen in Pollendiagrammen der Schweiz. Vierteljahrschr. Naturforsch. Ges. Zarich 124, 265-.283.

Burga, C.A., 1987. In: Wetter, W., Gletscherschwankungen in  Mont Blanc-Gebiet. Gebo Druck AG, Z~rich, 267 pp.

Burga, C.A., 1988. Swiss vegetation history during the last 18,000 years. New Phytol. 110, 581-602.

Burga, C.A., 1991. Vegetation history and paleoclimatology of the middle Holocene: pollen analysis of alpine peat bog sediments, covered formerly by the Rutor Glacier, 2510m (Aosta Valley Italy). Global Ecol. Biogeogr. Lett. 1, 143 150.

Burga, C.A., 1993. Pollen analytical evidence of Holocene climate fluctuations in the European Central Alps. In: Frenzel, B. (Ed.), Oscillations of the Alpine and Polar Tree Limits in the Holocene. Palaeoclim. Res. 9, 138-163.

Carcaillet, C., Thinon, M., 1996. Pedoanthracological contribution to the study of the evolution of the upper treeline in the Maurienne Valley (North French Alps): methodology and preliminary data. Rev. Palaeobot. Palynol. 91,399 416.

Clerc, J., 1988. Recherches pollenanalytiques sur la paléoécologie tardiglaciaire et holocène du Bas-Dauphiné. Thèse es Sciences, Aix-Marseille Ilk 179 pp.

David, F., 1992. Dynamique de la végétation darns les Alpes françaises du nord. Abstr. 8th Int. Palynological Congr., Aix en Provence, September 6 12, 1992, p. 31.

David, F., 1993a. Evolutions de la limite supdrieure des arbres dans les Alpes franc;aises du nord depuis la fin des temps glaciaires. Th6se es Sciences Universite Aix-Marseille 1II, 94 pp.

David, F., 1993b. Altitudinal variation in the response of the vegetation to Late-glacial climatic events in the northern French Alps. New Phytol. 125, 203-220.

David, F., 1995. Vegetation dynamics in the northern French Alps. Hist. Biol. 9, 269 295.

David, F., Barbero, M., 1995. De l'histoire du genre Betula dans les Alpes Fran~aises du Nord. Rev. Palaeobot. Palynol. 89, 455 467.

E1-Moslimany, A.P., 1987. The late Pleistocene climates of the lake Zeribar region ( Kurdistan, western lran) deduced from the ecology and pollen production of non arboreal vegetation. Vegetation 72, 13 l- 139.

Fail, P., 1992. Spatial patterns of atmospheric pollen dispersal in the Colorado Rocky Mountains, USA. Rev. Palaeobot. Palynol. 74, 293 313.

Gamper, M., Suter, J., 1982. Postglaziale Klimasgeschichte der Schweizer Alpen. Geogr. Helv. 2, 105-114.

Gensac, P.0 1967. Feuille de Bourg-Saint-Maurice ( XXXV-31 ) et de Mofitiers (XXX-32). Les groupements v6g6taux au contact des Pessieres de Tarentaise. Doc. Carte Vdg. Alpes V, 7-62.

Holzhauser, H., 1984. Zur Geschichte der Aletschgletscher und des Fieschergletscher. Phys. Geogr. 13, 552 pp.

Kullman, L., 1991. Pattern and process of present tree limit in the Täirna region, southern Swedish Lapland. Fennia 169 (1), 25 38.

Le Roy Ladurie, E., 1967. Histoire du climat depuis Fan rail. Flammarion, Paris, 376 pp.

Magny, M., 1995. Une histoire du climat, des derniers mammouths au si6cle de l'automobile. Ed. errance, 176 pp.

Markgraf, V., 1980. Pollen dispersal in a mountain area. Grana 19, 127 146.

Ozenda, P., 1985. La vdg6tation de la chaine alpine dans l'espace montagnard europ6en. Masson, 331 pp.

Patzelt, G., 1973. Die postglazialen Gletscher- und Klimaschwankungen in der Venedigergruppe (Hohe Tauern, Ostalpen). Z. Geomorphol. NF Suppl. 16, 25 72.

Pfister, C., 1988. Une rétrospective météorologique de l'Europe. Un système de reconstitution de l'évolution du temps et du climat en Europe depuis le Moyen-age central. Hist. Mesure II1, 313-358.

Ponel, P., De Beaulieu, J.-L., Tobolski, K., 1992. Holocene palaeoenvironments at the timberline in the Taillefer Massif, French Alps: a study of pollen, plant macrofossils and fossil insects. Holocene 2 (2), 117 130.

Röthlisberger, F., 1986. 10,000 Jahre Gletschergeschichte der Erde. Verlag Sauerlander, Aarau, 449 pp.

Stuiver, M., Reimer, P.J., 1993. Radiocarbon 35, 215-230.

Wegmüller, S., 1975. Les défrichements à l'étage subalpin dans la région de Valmenier et de la vallée de la Valloire (Haute-Maurienne, Savoie). In: L'Environnement et l'Homme. INQUA, Montpellier, pp. 309 315.

Wegmaller, S., 1977. Pollenalytische Untersuchungen zur spat und postglazialen Vegetationgeschichte der franz0sischen Alpen (Dauphine). Haupt, Bern, 185 pp.

Zoller, H., Schindler, C., Röthlisberger, H., 1966. Postglaziale Gletscherstande und Klimaschwankungen im Gotthardmassir und Vorderrheingebiet. Verh. Naturforsch. Ges. Basel 77, 97 164.

Zoller, H., 1977. Alter und Ausmass postglazialer Klimaschwankungen in der Schweizer Alpen. In: Frenzel, B. (Ed.), Dendrochronologie und Klimaschwankungen in Europa. Steiner, Wiesbaden, pp. 71-281.