150
peat embedding tephra attributed to this volcano; their mean
age was considered as denoting the Pavin eruption. In the
local context, it was estimated that one centimetre of
Holocene peat represents 25–50 years. Therefore, when
about 10 cm of peat was sampled beneath the tephra only,
instead of beneath and above the tephra, the measured^14 C
age predated the tephra by about 125–250 years as was the
case for Montcineyre.
Table 8.2 gives for information the age results which were
not used in the process. Most of them were measured many
years ago for studies on palaeosols. Such^14 C dating applied
to sediments and palaeosols was possibly biased (Goh et al.
1977 ; Geyh 1983 ; Hétier et al. 1983 ); the authors of the ini-
tial application of^14 C dating method to the Pavin confi rmed
such diffi culties (Delibrias 1979 ). Also,^14 C measurements
for palaeosols sampled at other sites during the same cam-
paign as for the MEMP group and published in the same
articles, were later invalidated (for example, for the
Gravenoire lava fl ow in the Royat valley: Huxtable et al.
1978 ; Boivin et al. 2009 ). The date for the Pont de Gouelle
(Montchal, Table 8.2 ) was rejected although it seems to
really concern carbonized plant remains and although it has
been repeated twice, because the authors of the work were
not sure of the origin of the correlated layer of tephra. Since
then, the corresponding outcrop has been identifi ed and a
new study is now planned.
New^14 C dates were obtained a few years ago for samples
extracted out of sediments drilled in the bottom of Lake
Pavin (Chapron et al. 2010 ). Of them we considered only
two results (Table 8.1 ) related to tree-leaves because the
sampling levels were suffi ciently close to the base of the lake
sediments for producing useful ages. A^14 C age measured in
the same work for undetermined organic material at the very
base of the log was not used despite the fact that it was in
agreement with the other data (Poz-27052: 6090 ± 40 B.P.)
because the particular conditions of the environment induced
a “reservoir effect” (Albéric et al. 2013 ). Sampling ‘terres-
trial’ plant macrofossils, such as leaves or pollen, obviates
such effects (e.g., Newnham et al. 2007 ; Staff et al. 2013 ).
Indirect age evaluation by palynostratigraphic correlation
was not accounted for in the calculations because it necessi-
tates further theoretical investigation and because it has been
demonstrated that pollen features are diachronous even at the
regional scale (Juvigné et al. 1988a ; 1996 ).
Strictly speaking, calibrating conventional^14 C ages will
result in probability distributions (Fig. 8.5 ) but for sake of
simplifying their expression in the text, we opted for quoting
the median ages (Table 8.3 ). The proposed age for the Pavin
crater is fi nally 4720 ± 170 B.C., i.e. a current age of
6730 ± 170 year. Thanks to the stratigraphic constraints, the
RenDateModel numerical method allowed us to estimate the
age of the maar of Estivadoux without any direct measure-Table 8.2 Published age data, not used in the present work
References
(Articles) Volcano and site MaterialAge reference
(Lab. No)14 C result
(B.P.)Calibrated age B.C.
interval (95 %) Comments
Pavin
Brousse et
Delibrias ( 1969 )Ciberons, near
Chaumianepalaeosoil Gif 1496 3450 ± 110Montchal
Brousse et
Delibrias ( 1969 );
Brousse et al.
( 1969 )Banks of Couze
Pavin river, upstream
to Gouelle bridgeCharred plant
remainsGif 1164 6670 ± 160 5890 5320 Attribution to
Montchal must be
confi rmedBrousse et
Delibrias ( 1969 );
Brousse et al.
( 1969 )Banks of Couze
Pavin river, upstream
to Gouelle bridgeCharred plant
remainsGif 1191 6650 ± 160 5830 5320 Attribution to
Montchal must be
confi rmedBrousse et
Delibrias ( 1969 )Banks of Couze
Pavin river, upstream
to Gouelle bridgeCharred plant
remains? 6660 ± 140 Attribution to
Montchal must be
confi rmed
Estivadoux
Camus et al.
( 1973 )1 km North of Les
Costespalaeosoil Gif 2350 6760 ± 130 Ex. “maar des
Costes”
Montcineyre
Brousse et
Horgues ( 1969 )CD 36 Road, near
Anglardspalaeosoil Gif 1278 5750 ± 150 4960 4330For Montchal, the 6660 B.P. result seems to be averaged from two distinct measurements, 6650 B.P. and 6670 B.P. (Brousse et Delibrias 1969 )
E. Juvigné and D. Miallier