Middle- to late-Holocene storminess in Brittany (NW France): Part I – morphological impact and stratigraphical record

Abstract

Our study aims to understand the recurring climatic conditions prevailing during the largest storms reaching NW France (Brittany). These storms are responsible for the breaching of coastal barriers and major flooding of lowlands. In a first part of our work, we examine the morphological impact and stratigraphic record of storm events along Western Brittany rocky coasts, with a special focus on the southern coast of the Bay of Audierne, the most exposed coast of the region. In a second paper (‘Middle- to Late-Holocene Storminess in Brittany (NW France): Part II’), we shall focus on the chronology of storm events and their climate forcing conditions. Drilling transects and stratigraphic analyses were first undertaken to constrain chronology, strength and wind direction during the main Holocene storm events. New dates, observations and a relative sea-level (RSL) curve were then used to inform discussion of the necessary climatic and morphologic conditions leading to destructive storm events. Most recorded events appear to be linked with cooling episodes of the Holocene and a RSL close to present. Some storms are clearly responsible for breaching and dune building or remobilisation. We demonstrate that storm frequency and intensity appear to rise in a stepwise manner during the late Holocene. Maximum efficiency is reached during the ‘Little Ice Age’ with clustered events probably lasting several days, but major storms also occurred immediately prior to the ‘Medieval Warm Period’. We suggest that recent coastal dune building from c. ad 1100 until now, despite a sea level close to present and continuously rising, may be a direct consequence of the restoration of beaches after periods of recurrent storminess. This building activity often occurred during dry negative North Atlantic Oscillation (NAO) events, in connection with the available sedimentary supply.

Keywords

Brittany dunes erosion North Atlantic Oscillation sea level storminess

Introduction

Generalities

Major coastal flooding risk in western Europe is attributed to storminess (Wolf, 2009) and climate change. A slight trend towards weakening of storm activity has been observed over the past century (Bijl et al., 1999). Several authors argue for a change in global circulation based on wind data (Clarke et al., 2002; Dawson et al., 2007; Dickson et al., 2000; Pirazzoli, 2000; Pirazzoli et al., 2004; Pye and Neal, 1994) and on higher frequencies of the positive North Atlantic Oscillation (NAO) mode (Visbeck et al., 2001). Indeed, observed European climate variability during the winter season (in terms of wind and storminess) may be largely explained through the NAO index pattern (Alexander et al., 2005; Burningham and French, 2012). Furthermore, a fair correlation has been observed between solar activity, the NAO index and cold winter conditions occurring in the northern Atlantic sector (Ineson et al., 2011). However, the studies cited above essentially deal with regional or global frequency, but not with specific destructive events induced by major storms. Morphogenic storm events are only recorded along the coast and are not necessarily the result of major storms of intensity similar to the Xynthia event (in 2010) or the Great Storm of 1987; their impact also depends on additional circumstances such as sediment starvation. Major storms may induce even greater morphogenetic impacts if high tide, storm surges and high amplitudes are combined (Wolf, 2009).

Over longer periods, discontinuous instrumental data and proxy records indicate that the cool ‘Little Ice Age’ (LIA; c. ad 1300–1900) included periods of enhanced storminess compared with today (Bärring and Fortuniak, 2009; Clarke and Rendell, 2009; Lamb and Frydendahl, 2005). Previous studies on north-western European coasts suggest that over the same period, the increase in storminess has resulted in flooding as well as in numerous episodes of sand drift, leading to the construction of coastal dunes (Clarke and Rendell, 2009; Lamb and Frydendahl, 2005). Moreover, following Orford et al. (2000), and Lamb (1995), dune building seems linked to periods of limited marine regression (c. 50 cm). In contrast with this traditional interpretation of coastal dune establishment (e.g. Pye and Tsoar, 2009), Regnauld et al. (1996) suggested that dune building may preferentially occur when sea level is rising. Recently, Mauz et al. (2013) demonstrated that dunes may develop in connection with both regressive and transgressive sea level and as a function of sedimentary supply. However, studies carried out on western European sand drift have provided extended Holocene records and demonstrated that stormy conditions generally prevailed during Northern Hemisphere (NH) cooling intervals (Meurisse-Fort, 2009; Orford et al., 2000), especially during the Sub-Boreal (middle Holocene). These intervals would bring with them associated episodic flooding linked with combined storm surges. At times, heavy rainfalls and a high groundwater table (Gandouin et al., 2007; Meurisse-Fort, 2009) may limit sand drift.

Middle-Holocene climate deterioration is evident from the re-extension of NH glaciers (Denton and Karleń, 1973; Long et al., 2009; Magny and Haas, 2004) and from a southern shift in baroclinic depression main tracks. Furthermore, an increase in storminess during the Sub-Boreal period may be linked with both the weakening of the Atlantic Meridional Overturning Circulation (AMOC) and a transition towards higher frequencies of negative NAO conditions (Dawson et al., 2007; Magny et al., 2003; Trouet et al., 2009, 2012), induced by Arctic cooling and a larger degree of sea-ice cover. On the contrary, when the AMOC efficiency was stronger, sea-surface warming occurred in the North Atlantic Ocean. According to Sorrel et al. (2009), this oceanic behaviour would be regulated through the Holocene by a 1500-year cyclicity without any clear evidence for solar forcing.

The storm record in Brittany

Brittany, and particularly the Bay of Audierne (Figure 1), is one of the most exposed French western coasts in terms of storms and flooding risk. It is also one of the best places in France for recording major storm impacts and for the reconstitution of meteorological conditions that occurred when coastal barriers were destroyed.

On Brittany’s coasts, storm impacts can only be traced in written archives back to the 16th century (Hallégouët and Hénaff, 1993). The coastal barrier of the Bay of Audierne is one of the best sites in the region for analysing the geological record of past storminess.

The Bay of Audierne is mostly characterised by bioclastic sands with a limited silicoclastic supply on a very flat polygenetic platform. This NNW–SSW coastline is exposed to the Western Atlantic fetches, particularly during SW and NW storms. For instance, during the October 1987 hurricane, 15-m-high waves were recorded offshore Penmarc’h (cf. Figure 1). Furthermore, waves strongly impacted the crests of Kerity dunes (cf. Figure 1) at high tide during the storms experienced on 7 January 1839 (Monfort, 1985) and 9 January 1924 (waves of 8 m height, 2.10 m storm surge; Bresson, 1924). Geomorphological studies carried out in Brittany have highlighted important transformations sustained by this susceptible coastline during storms, notably around ad 1000, as well as a consequence of sea-level rise. This is particularly so since the 1950s (e.g. Guilcher and Hallégouët, 1991; Hallégouët et al., 1989). During the LIA, the Beg-an-Dorchenn Peninsula (Figures 1 and 2) was merging with the dune (Cassini’s map, 1786 first edition), while the Lescors and La Joie marshes (cf. Figure 2) were regularly flooded. Stratigraphic studies carried out in this region only provide a limited record of storm impact signatures (Carter et al., 1993; Haslett and Bryant, 2007). A recent campaign of deep drilling was undertaken on this exposed Brittany platform in order to analyse the preserved history of the Holocene coastal sedimentary prism. New data obtained around Brittany have been combined with older regional observations (cf. Figure 1) of storminess trends recorded after the Holocene thermal optimum in Brittany (this study).

Geological and geomorphological settings

Brittany

From a geodynamical point of view, Brittany’s peninsula is related to the post-rift subsidence of the Atlantic margin, like Devon and Cornwall, United Kingdom (Ziegler, 1992; Figure 1). At the Holocene scale, southern Brittany, south of the South Armorican Shearing Zone, has been subsiding since the opening of the Bay of Biscay (Sibuet et al., 1991), while the central part and the northern coast of Western Brittany (Finistère) is tilting to the Western Approaches of the English Channel.

Post-glacial isostatic adjustments (GIAs) are still relatively poorly constrained for the study area. Work conducted on the subject for south-western United Kingdom hint that Brittany must have been under the joint influence of both a forebulge subsidence inherited from the British Ice Sheet and long wavelength isostatic responses related to the distant, but much larger Fennoscandia and Laurentide ice sheets, as well as to hydro-isostatic effects. The RENAG database (http://webrenag.unice.fr; Z: multisatellite reference base) gives a subsidence rate of −0.3 mm/yr for the southernmost Cornwall (United Kingdom) and of 0.0 mm/yr for Brest.

Peltier’s modelling exercises for the last 250 years give −0.66 mm/yr (Penzance, United Kingdom) for Newlyn −0.62 mm/yr for Brest (France), respectively (Peltier, 2004). GIAs associated with the peripheral bulge is suspected to influence the relative sea-level (RSL) reconstruction for Brittany (Lambeck, 1997). Hence, in the absence of more precise information regarding the relative position and potential migration of the glacial forebulge, the question of Holocene vertical behaviour in the Brittany region with respect to glacio-isostatic effects remains debated. In particular, incertitude exists on the roles possibly played by deep crustal structures, such as the Lizard Suture Zone and the Ushant Fault Zone (Sibuet et al., 1991), in the mitigation of long wavelength deformations such as isostatic adjustments or reactivation of the tectonic inversion (Van Vliet-Lanoë et al., 2002). Values inferred from south-western United Kingdom must then only be taken as indicative. The recent Holocene RSL curve established for Brittany (Goslin, 2014; Goslin et al., 2013) indicates an estimated subsidence rate of c. −0.3 mm/yr for the last 2000 years (see below).

Bay of Audierne

The basement in the Bay of Audierne mainly consists of mica schist and gneiss, with some granite to the south (Figure 2). Several palaeoplatforms, that is, old marine formations, are cut into the mica schist and are mainly preserved in the northern part of the bay (Hallégouët, 1990). Rusty and stained quartz gravels, spread north of the bay close to 16–20 m Nivellement Général de la France (‘NGF’, i.e. French geodetic level), are generally attributed to Quaternary interglacials (Guilcher, 1948; Guilcher and Hallégouët, 1991; Morzadec-Kerfourn, 1974). Electron spin resonance (ERS) dating of these quartz levels proves the polygenetic character of the beaches. Combined with subsidence, this explains the preservation of a weathered low-lying platform dating back to the Neogene. Beaches along southern Brittany thus bear witness to both Pliocene low glacial sea levels and stacked Quaternary interglacial levels (Van Vliet-Lanoë et al., 2002, 2009). From Penhors to Penmarc’h (Figure 2), continuous outcrop shows a lower abrasion surface rising from 6 to 10 m NGF, weathered over approximately 2 m (i.e. coherent saprolite) by continental erosion and wave action. Waves irreversibly modify the morphology of the coast, yielding the present-day shape of the low platform. Furthermore, archaeological settlements installed at the Beg-an-Dorchenn (south of Bay of Audierne; Figure 2) supply additional information on past environments.

The active beaches comprise many cobbles north of Penhors Village, although calcareous sandy beaches prevail from Penhors to Pors Carn. Quartzite gravels derive from veins in the mica schist (Penhor’s sector, Figure 2). The present-day coast is limited by a cobble barrier from Penhors to Trunvel, followed by a dune barrier anchored in the south of the Beg-an-Dorchenn Peninsula. Today, these dune ridges are commonly eroded with a coastal cliff. The altitude of the present high water spring tide (HWST) reaches 2.64 m NGF, while the tidal amplitude is 4.3 m (St Gwénolé harbour). Prevailing winds are commonly those of a Westerly nature, although during the 1960s and since 2002, meridian winds have been prevailing (Hénaff, 2008). The tidal platform is limited onshore by a rocky palaeocliff. Its sedimentary cover is limited and inland varies from 5 to 15 m in thickness. Local river incisions exist, thus allowing for the development of back-barrier marshes. On the Penmarc’h Peninsula, the average bottoms of the Lescors and La Joie marshes are between +2 and +5 m NGF with some localised incised channels. Those two marshes correspond to a meandering valley, dammed originally by a (cobble) beach at the level of La Joie church (Figure 2).

The outcropping sands along the Bay of Audierne are rather fine in grain size (Toulemont, 1964), but limited in volume (no offshore storage). The median size remains at around 135–140 µm close to Beg-an-Dorchenn, especially for the intertidal zone and the dunes. The silicoclastic sands are mostly supplied by the erosion of saprolite and quartz veins cropping out at the surface of the rocky platform or attacked in cliffs to the north of the bay, further drifted by longshore transit. A more limited proportion is inherited from the local rivers or raised beaches. Layers of coarser sands to cobbles exist from Trunvel to Beg-an-Dorchenn, and are provided during storms by a rapid longshore transit from the northern part of the Bay. Most of the sands are highly bioclastic. Within the dune’s field, laminated mud and sand attest to the decantation of finely crushed shells with fine bacterial carbonated mud, inherited from the older calcareous beach sandstones.

Methodology

All of the coastal sites presented here (cf. Figures 1 and 2) have been regularly monitored since 1980 (Hallégouët, 1981), by levelling and differential global positioning system (DGPS). Digital elevation model data (25 m grid) were provided by the French Geographical Institute (IGN). Drilling transects were chosen orthogonally to the coast, in sectors with limited erosion (Figure 2 and Table 1). These cross sections have been obtained through BRGM (French Geological Survey) mechanical drilling to a maximum depth of 15 m (until the rock basement). Stratigraphic description (accelerator mass spectrometry (AMS)-¹⁴C dates) as well as sediment analysis (including fine description of the sedimentary facies, grain size analysis and mineralogy) have been undertaken on these coastal sections. Palaeostorminess traces are marked – by barrier breaching – by washover fan deposits with or without load casting figures (liquefaction) or isolated cobbles in sands. Indeed, back-barrier lagoons may be filled by a perched rhythmic sedimentation of calcareous mud and sands or alternating brackish and terrestrial malacofauna. Storms are also recorded by shingle or grit splays truncating dune deposits, by peat truncation and burial under brackish or marine deposits (Buynevich et al., 2004; Donnelly et al., 2001; see Figure 5) or by consolidation of peat and other deposits (impact of desiccation).

Table 1.

Site locations (coordinates).

Place	Drilling	Altitude NGF	N	W
Tronoën	TR1	6	47°51′05.7″	4°20′50.0″
	TR2	5	47°51′09.0″	4°20′44.9″
	TR3	6	47°51′10.7″	4°20′39.6″
	TR4	14	47°51′21.8″	4°19′59.0″
	TR5	6	47°51′13.2″	4°20′29.8″
	TR6	9	47°51′15.2″	4°20′23.2″
Beg-an-Dorchenn	BD1	7	47°50′24.2″	4°20′40.2″
	Breach	4	47°50′25.1″	4°20′36.5″
	Burned bark	5	47°50′11.8″	4°20′54.9″
Pors Carn – Lescors	PR 5	4.5	47°49′17.1″	4°20′28.6″
	PR 5′	3.8	47°49′21.2″	4°20′30.3″
	PR 6	8	47°49′29.2″	4°20′44.4″
	PR 7	11	47°49′44.6″	4°20′43.1″
	PR 7′	9.3	47°49′39.0″	4°20′46.5″
	PR 8	8	47°49′47.5″	4°20′50.2″
	PR 9	6	47°49′48.7″	4°20′53.8″
Le Conquet	Relict raised beach	8	48°21′21.1″	4°46′42.0″

NGF: Nivellement Général de la France.

Considering the scarcity in datable material at the Bay of Audierne, we use not only our own samples but also all available information existing in the literature. In this study, 20 new AMS-¹⁴C dates (15 dates from the Poznan Radiocarbon Laboratory, Poland, and 5 dates from the ‘Laboratoire de Mesure du Carbone 14’, UMS 2572, CEA Saclay, Gif-sur-Yvette, France) have been obtained on peat, land snails and marine shells (Table 2). In this paper, we discuss 40 dates from the Audierne/Ile Tudy sectors (cf. Figure 1) and 16 other pertinent dates from Brittany (Table 2). The calibration of all dates was undertaken mostly at 2σ, using the CALIB 6.0. software (Stuiver et al., 2005) and the non-marine radiocarbon calibration (IntCal09; Reimer et al., 2004) for peats and terrestrial snails, and the marine radiocarbon calibration (Marine09; Reimer et al., 2004) for marine shells. In order to accommodate local effects for marine shells, we have introduced a regional marine reservoir correction Δr of approximately 40 ± 42 years in the area (Sein Island; Hughen et al., 2004; Stuiver et al., 2005).

Table 2.

Radiocarbon dating: local (bold) and Brittany (regular).

Location	Material	Altitude NGF (without decompaction)	Lab number	Radiocarbon age (conventional) BP	Calibrated age, best intercept (average)
Kerdrafic	Shells (Zirfaea crispata) in sandstone ΔR −40 ± 42	−2 m	SacA-23972	565 ± 30	ad 1705–1857 (ad 1781 ± 40)^a	This paper
Tronoën (coast)	Shells in dune with ΔR −40 ± 42 (pre-STS)	+4.50 m section	Beta-145238	580 ± 120	ad 1677–1898 (ad 1787 ± 110)^a	Haslett and Bryant (2007)
La Joie Marsh	Buried peat summit BT 1/1 (pre-STS)	−4.25 m	Beta-57656	140 ± 70	ad 1719–1778 (ad 1750 ± 30)^a	Carter et al. (1993)
Anse du Verger	Buried peat summit (pre-STS)	+6.5 m	?	113 ± 100	ad 1753–1762 (ad 1757± 10)^a	Regnauld et al. (1996)
Beg-an-Dorchenn (south)	Shell (Littorina) with ΔR −40 ± 42 (intradune gravel)	−2 m	Gif-238	520 ± 70	ad 1671–1861 (ad 1766 ± 90)^a	Giot (1966)

Lescors marsh	Buried peat summit (BH 2; pre-STS)	−2.89 m	Beta-57653	370 ± 60	ad 1572–1629 (ad 1600 ± 30)	Carter et al. (1993)
La Joie marsh	Flooded peat summit BT1-2 (pre-STS)	−3.90 m	Beta-57657	350 ± 60	ad 1556–1632 (ad 1594 ± 40)	Carter et al. (1993)
Mt St Michel Bay	Marine shell with ΔR −40 ± 42 (pre-STS)	+3 m	Poz-6666	750 ± 30	ad 1541–1650 (ad 1595 ± 55)	Billard et al. (2013)

Lescors marsh	Buried humic soil summit (BH 1; pre-STS)	−1.22 m	Beta-59650	810 ± 50	ad 1152–1284 (ad 1218 ± 70)	Carter et al. (1993)
Mt St Michel Bay	Peat (post-STS)	+2.9 m	Ly-13288	975 ± 35	ad 1012–1156 (ad 1084 ± 75)	Billard et al. (2013)

St Urnel graveyard	Bones (Su 50: several remains)	+14 m	Gif-2296	970 ± 90	ad 989–1165 (ad 1077 ± 90)	Giot and Monnier (1977)
St Urnel graveyard	Bones (Su 46-III-8/18)	+ 14 m	Gif-2681	980 ± 90	ad 983–1163 (ad 1073 ± 90)	Giot and Monnier (1977),
Lescors marsh	Main peat base (BH1) post-STS	−1.82 m	Beta-57651	1010 ± 50	ad 947–1157 (ad 1052 ± 100)	Carter et al. (1993)
Maguero (Lorient)	Land snail in dune (pre-STS)	+ 7 m	Poz-13366	1025 ± 30	ad 990–1024 (ad 1007 ± 20)	Meurisse-Fort (2009)

Beg-an-Dorchenn breach	Dunes snail in lagoon infill (post-STS)	+ 4 m	Poz-52797	1105 ± 35	ad 937–982 (ad 960 ± 25)	This paper
Fogéo (Rhuys)	Peat (post dune)	+ 6 m	A9258	1095 ± 35	ad 888–1017 (ad 952 ± 65)	Visset and Bernard (2006)
Beg-an-Dorchenn North drill	Shells in a lower beach ΔR −40 ± 42	−1.4 m	Poz-37202	1460 ± 30	ad 872–998 (ad 935 ± 70)	This paper
Beg-an-Dorchenn North drill6	Shells in a lower beach ΔR −40 ± 42	−0 m	Poz-36728	1435 ± 30	ad 836–971 (ad 903± 70)	This paper
Tronoën TR6	Marine Shells in sand (Cerastoderma edule) with ΔR −40 ± 42	−2 m	Poz-31327	1620 ± 30	ad 743–875 (ad 809 ± 70)	This paper

Kerdrafic	Shells (Mytilus edulis) in sandstone ΔR −40± 42 (beach)	−4 m	Poz-34212	1605 ± 30	ad 703–957 (ad 830 ± 130)	This paper
Pors Carn	Shell with ΔR −40 ± 42 (intradune gravel; pre-STS)	+6 m	Gif-891	1280 ± 170	ad 614–899 (ad 756 ± 150)	Giot (1998)
Beg-an-Dorchenn North drill	Shells in a lower beach (base) ΔR −40 ± 42 (pre-STS)	−3 m	Poz-36729	1600 ± 30	ad 688–795 (ad 741± 60)	This paper

Pors Carn	Cultivated humic soil within the dunes; drilling 6	+6.50 m	Poz-31323	1355 ± 30	ad 648–676 (ad 662 ± 20)	This paper
Beg-an-Dorchenn North drill	Shells in a lower beach	−3.5 m	SacA-23973	1390 ± 30	ad 632–664 (ad 648 ± 20)	This paper
Pors Carn	Terrestrial snails, peaty sandy soil at the base of the dune	+3.50 m	Poz-31324	1380 ± 40	ad 622–671 (ad 646 ± 60)	This paper

Tronoën (coast)	Terrestrial snails (dune section)	+ 4.50 m	SacA-13344	1550 ± 30	ad 435–491 (ad 463 ± 30)	This paper
Lescors marsh	Lower peat summit (BH1; pre-STS)	−0.25 m	Beta-57652	1710 ± 60^a	ad 312–400 (ad 356 ± 45)	Carter et al. (1993)
St Jean Thomas (Cotentin)	Wood (pre-STS)	+ 1.55 m	Ly-12464	1700 ± 60	ad 317–409 (ad 363 ± 50)	Billard et al. (2013)
La Joie Marsh	Peat summit MJ 6/1 (pre-STS)	+ 0.82 m	Beta-57654	1890 ± 60	ad 59–178 (ad 118 ± 60)	Carter et al. (1993)
Mt St Michel Bay	Marine shell with ΔR −40 ± 42 (pre-STS)	−0.1 m	Poz-6667	1920 ± 35	ad 54–126 (ad 90 ± 50)	Billard et al. (2013)

Fogéo (Rhuys)	Peat summit under marine transgressive mud	−0.82 m	A9259	2085 ± 40	121–51 bc (86 ± 35 bc)	Visset and Bernard (2006).
Anse du Verger	Shells (kitchen middens in dunes, pre-STS) with ΔR −40 ± 42	+5 m	Pa-1200	2460 ± 80	220 bc–ad 20 (120 ± 100 bc)	Regnauld et al. (1996),
Beg-an-Dorchenn North drill	Shell (Scrobicularia plana) ΔR −40 ± 42	−4,4 m	SacA-23974	2475 ± 30	191–44 bc (117 ± 30 bc)	This paper
Tronoën TR5	Shell (Cerastoderma edule) ΔR −40 ± 42 (channel)	−6.5 m	SacA-23975	2475 ± 30	191–44 bc (117 ± 30 bc)	This paper
Fogéo (Rhuys)	Peat under dune (pre-STS)	−0.2 m	A 9260	2125 ± 75	209–47 bc (128 ± 80 bc)	Visset and Bernard (2006)

Beg-an-Dorchenn North	Shells with ΔR −40 ± 42 (sandstone with cobbles)	−1.5 m	Gif-1100	2670 ± 110	bc 524–194 (bc 359 ± 170)	Giot (1981)
Nerizellec Marsh	Truncated peat summit (pre-STS)	−1.25 m	Beta-57647	2390 ± 60	541–395 bc (468 ± 70 bc)	Carter et al. (1993)
Beg-an-Dorchenn breach	Terrestrial and marsh snails, interstratified with dunes peaty soil	+ 6 m	Poz-31321	2375 ± 30	538–390 bc (464 ± 75 bc)	This paper
Lescors Marsh	Organic colluvium (pre-STS)	+ 2 m	Poz-31322	2400 ± 35	513–404 bc (458 ± 55 bc)	This paper

Kermor (Ile Tudy)	Organic mud (post-STS)	0 m	Poz-36730	2465 ± 30	670–481 bc (575 ± 90 bc)	This paper
Nerizellec Marsh	Truncated peat (pre-STS)	−1.45 m	Beta-57648	2420 ± 80	775–388 bc (581 ± 200 bc)	Carter et al. (1993)

Anse du Verger	Sandy peat summit (BH2; pre-STS)	+ 2 m	?	2630 ± 130	930–728 bc (829 ± 100 bc)	Regnauld et al. (1996)
La Joie Marsh	Peat MJ 6/2 (pre-STS)	− 0.48 m	Beta-57655	2820 ± 70	1057–897 bc (977 ± 80 bc)	Carter et al. (1993)
Nerizellec Marsh	Peat base	−1.32 m	Beta-57649	2890 ± 70	1134–977 bc (1055 ± 80 bc)	Carter et al. (1993)
Lampaul-Plouarzel-Trézien	Floated wood (pre-STS)	+ 1.5 m	Gif-712	3660 ± 115	1409–1125 bc (1267 ± 140 bc)	Morzadec-Kerfourn (1974)

Gwendrez	Humic matter: palaeosol below the dune	+ 35 m	Beta-11497	3950 ± 70	2595–2210 bc (2402 ± 200 bc)	Haslett et al. (2000)
Kermor (Ile Tudy)	Thin shore peat	0 m	Poz-36731	4530 ± 30	3241–3103 bc (3172 ± 70 bc)	This paper
Pen Hat	Terrestrial snail (base of dune)	+ 9 m	Hv-25113	4590 ± 110	3635–3013 bc (3323 ± 310 bc)	Meurisse-Fort (2009)
Beg-an-Dorchenn south	Charcoal (fire place)	+ 4.5 m	Gif-5063	5140 ± 110	4050–3782 bc (3916 ± 135 bc)	Peuziat and Giot (1983)

Tronoën TR6	Truncated peaty sandy soil summit	−4 m	Poz-31325	6170 ± 40	5180–5063 (5121 ± 60 bc)	This paper
Tronoën TR5	Sandy peat soil on tidal mud	−5.50 m	Poz-31326	6260 ± 40	5299–5217 bc (5258 ± 40 bc)	This paper

Porz Carn	Burned bark in sandy colluvium	+5 m	Poz-34213	6750 ± 40	5724–5617 bc (5670 ± 60 bc)	This paper
Beg-an-Dorchenn kitchen midden	Shells ΔR −40 ± 42	Excavation base	Gif-6858	7580 ± 65	6218–5894 bc (6056 ± 160 bc)	Dupont et al. (2010)
Beg-an-Dorchenn kitchen midden	Shells ΔR −40 ± 42	+6 m	Ly-12290	7195 ± 70	5755–5602 bc (5678 ± 80) bc	Dupont et al. (2010)
Anse du Verger	Muddy peat	−8 m	?	7310 ± 80	6236–6072 bc (6154 ± 80 bc)	Regnauld et al. (1996)
Succinio (Morbihan)	Mud	−4 m	A 9833	6600 ± 85	5669–5461 bc (5565 ± 105 bc)	Visset and Bernard (2006)

NGF: Nivellement Général de la France; STS: storm surge.

Radiocarbon dates calibrated using the CALIB 6.0 programme (Stuiver et al., 2005) that uses IntCal05 (Reimer et al., 2004) and regional ΔR values (marine carbon reservoir; Hughen et al., 2004; Stuiver et al., 2005).

a¹⁴ C plateau effect.

The description follows the final stratigraphic labelling (see Table 1: coordinate, Table 2: dating, Table 3: stratigraphy) and is summarised in a compound log (Figure 4), which compiles whole field observations, drillings and dating. Complementary data were extracted from the report of Carter et al. (1993). They concern mostly data obtained by drilling in the marshes of Lescors, La Joie and Nerizellec. For other sites in Brittany, the selected dating corresponds to events susceptible to track RSL (discontinuity, marine invasion, etc.) or storm impact.

Table 3.

Stratigraphic description for the Penmarc’h–Lescors–Pors Carn sector and the Tronoën–Beg-an-Dorchenn sector.

Unit	Altitude (m NGF)	RSL^a (m)	Description	Dates^b (cal.)	Interpretation
Pors Carn–Lescors–La Joie (sections and drilling)
E3			Thin, massive, calcareous fine sand sheet		Recent (coastal management)

E2		−0.)	Thin dark-brown soil on the last dune, partly burying D5d1	Seaweed furnaces (late 19th to early 20th century), now displaced (2013)	Cultivated from c. 1850 to 1940

D3d		0	Stratified sand, grit and gravels, rich in quartz	1677–1898 cal. ad; Beta-145238, Haslett and Bryant (2007)	Storm terrace
		0	Stratified sand, grit and gravels, rich in quartz	1671–1886 cal. ad; Gif-238, Giot (1966)	‘Little Ice Age’

D3b			Sand invasion on Lescors marsh	1152–1264 cal. ad	Third major dune building

D3a	+5.50		Stratified sand, grit and gravels, rich in quartz	614–899 cal. ad on shell (recalibrated Gif-891; Giot, 1998)	Storm terrace
D3a	+5.50		Stratified sand, grit and gravels, rich in quartz	614–899 cal. ad on shell (recalibrated Gif-891; Giot, 1998)	c. 900 ad

D2			Organic soil on dune (land snails)		Stabilisation, limited ETP and high precipitation

D1c			Thick stratified calcareous fine sand		Second dune building

D1b	+1		• Organic soil on dune (land snails)	622–671 cal. ad (Poz-31324)	Stabilisation, limited ETP and high precipitation
D1b	+1		• Lacustrine peat (Lescors)	648–676 cal. ad (Poz-31323)	Stabilisation, limited ETP and high precipitation

D1a	+0	−0.40	• D1a2 first layer of calcareous sandy• D1a1Top of the peat in Lescors marsh	Old quay of St Gwénolé (around ad 400; Giot, 1984)	First dune building (Late Roman)
D1a	+0	−0.40		59–178 cal. ad (recalibrated; Beta-57654; Carter et al., 1993)	First dune building (Late Roman)

C1/C3	(+6)	−0.75	• Loamy ‘kaki’ colluvium to organic colluvium, sandy, including charcoal and shells, +4.5 m NGF• Peat in Lescors marsh	475–0 bc salt pan	Second Iron Age clearance and agriculture; limited ETP and high precipitation
C1/C3	(+6)	−0.75		513–404 cal. bc (Poz-31322)

	Base at		Cliff incision, buried by C1		Storm incision

B2c	+6.50		Beach gravel, including quartz, quartzite and flint		Storm terrace

B1b	10–25 cm thickAt + 1 m above the HWM	−1.00	• Base of the peat at La Joie marsh• Humic loamy coarse sand including charcoal and shells	1057–897 cal. bc (recalibrated Beta-57655; Carter et al., 1993)	Cultivated horizon (Ap)
		−1.00			First Iron Age
		−8.00		5724–5617 cal. bc (Poz-34213)	Mesolithic

B1a			Loamy coarse sand up to 1m thick

A1			Saprolite on granite, variable in thickness

A0			Rock basement: granite
Beg-an-Dorchenn–Kerdrafic Tronoën
E3		+.0.3	Present-day accumulation of dune sands		Coastal dune management

E2		0	Stabilisation loamy ground on the back of the dune barrier, enriched in organics and various anthropic debris	Archives: Atlantic wall (1942–1945)	Formed during the Second World War

E1		−0.5	Loamy organic sands forming a stratified deposit or in an Ap soil horizon	Seaweed furnaces (late 19th to early 20th century)	Cultivated from c.1850 to 1940

D3d	+0 NGF	0	Recent ‘piddock pits’ of Zirfaea crispata (shellfish, Nordic species carved in D2B2)	ad 1705–1857 (SacA-23972)^b	Storm abrasion surface with important dune retreat

D3b	+8	0	3-m-thick stratified coarse sand, in back-barrier position, pending towards inland, including grit, somewhat compacted with irregular load casting at the base, alternating with finer sandy layers with ripple marks (D3b2 and D3b3). Its base is erosive and is locally occupied by alluvial sands (D3b1); turns inland to lagoonal stratified calcareous silts and sands. Some of the layers are slightly organic (palaeosols).Dunes at St Urnel: very large splay of grit at the land surface, up to 300 m from the present-day coastline	ad 1677–1898 (Beta-145238)^b	Stacked storm washover fans and ridges separated by stabilisations
D3b	+3.6			ad 1677–1898 (Beta-145238)^b	‘Little Ice Age’.

D3a2	+ 3.6		At the coast: an overconsolidated massive fine sand body which consists primarily of fine whitish calcareous sands, including reworked land snails	Dune snails 937–982 cal. ad (Poz-52797)	Compound washover fan (upper part of storm beach ridge) re-trimming dune and beach sands, associated with storm erosion of an early storm ridge (Iron Age) c. 1000–1050 ad (Capetian)
			10 m inland: the sedimentation evolves into slightly stratified, and sometimes slightly humic, pending inland alternating planar calcareous silts, up to 4-cm-thick and thin sheets of oxidised fine sand with shell fragments, brackish to freshwater snails (Hydrobia, Lymnaea, Planorbis) and terrestrial snails, to laminated sands, with local ripples and dune snails alternating with calcareous silts	Reworked terrestrial snails ‘538–390 cal. bc’ (Poz-31321)
	+0.5		Two erratic burial at St Urnel graveyard	Bones 989–1165 cal. ad (Gif-2296)
			Two erratic burial at St Urnel graveyard	Bones 983–1163 cal. ad (Gif-2681)

D3a1	+0.5		3.5 m grey coarse stratified silicoclastic sands including shell fragments and grind, slightly calcareous (drilling)	0 m NGF 872–998 ad (Poz-37202)	Re-trimming and storm erosion of a compound storm ridge (washover fan?)
				−1.4 m NGF ad 836–971 (Poz-36728)
				−2.3 m NGF ad 632–664 (SacA-23973)
	−3.0			–3 m NGF ad 668–795 (Poz-36729)	c. 900 ad (Capetian)

D2	+4.3		Fine yellowish brown sand, unconsolidated, most likely corresponding to the nearby Kerdrafic sandstone: a bright yellowish brown, slightly calcareous, stratified sandstone without shells, with some rare cobbles, with humic palaeosoil close to the top; a sandy aggradation preserved in a tidal channel at Tronoën (drilling)	Mussel in lower beach sandstone (Kerdrafic):	Alluvial fan on a storm erosion surface, topped dunes sands: storm ridge c. 800 ad (Carolingian)
				703–957 ad (Poz-34212)
				743–875 ad (Poz-31327)	Dune and storm ridge form
	−2	−0.30	D2a: compacted yellow sand interstratified with dune and two humic palaeosoils	435–491 ad (SacA-13344) on land snails	c. 400

C3	−4.4		Dark grey muddy fine sand, including marine shell fragments and gravels with few cobbles (Beg-an-Dorchenn, BD1 ‘Trangressive’ fine sand, coarsening to the top (Tronoën drilling))	Reworked shells: pre-STS 191–44 cal. bc (SacA-23974).	Storm erosion surface, storm ridge; breaching of an old coastal ridge with the formation of a tidal channel and lagoon at the base (Tronoën)
C3	−6.5	−0.55		Reworked shells: pre-STS 191–44 cal. bc (SacA-23975)	c. ad 200 (late Roman)

C2	(−1.5)		Shelly, well-consolidated whitish calcareous sandstone, excavated by recent ‘piddock pits’ of Zirfaea crispata	Reworked shells: pre-STS 524–194 cal. bc (recalibrated Gif-1100)	Storm erosion surface and storm ridge
C2	−5	−0.80			Storm erosion surface and storm ridge

C1			Basal organic soils (colluvium or gyttja) in drilling (Tronoën)	5180–5063 cal. bc (‘kaki mud’, Poz-31325)

B2			B2b is a gravel with a sandy matrix		Early transgression (tidal marsh followed by the lower part of a beach ridge)
B2			B2a is a mud grey sand (drilling)

B1		−6.00	Soil developed at the surface of the saprolite	5299–5217 cal. bc (Poz-31326)

A1			Weathered mica schists		Old tidal platform, rather regular

RSL: relative sea level; NGF: Nivellement Général de la France; STS: storm surge; ETP: evapotranspiration.

RSL (Goslin, 2014): deduct 2.4 m to obtain the NGF local value.

See Table 2 for details.

Bold lines represent storm abrasion surfaces.

The RSL curve used within this study (see Figure 5) was built on a regional scale, following a morpho-sedimentological approach (Goslin, 2014; Goslin et al., 2013). It is presented in an age/altitude diagram against the actual HWST level.

The regional RSL curve used within this study (Goslin, 2014; Goslin et al., 2013) was constructed using a multiproxy analysis based on the conjoint use of geochemical indicators (following the approach of Kemp et al. (2012) and Engelhart et al. (2013)), micromorphological indicators (pyrite; Dellwig et al., 2001, 2002; Enio et al., 2011) and plant macro-remains in order to determine the original position of the studied deposits within the tidal frame, with reference to the HWST level. Reconstructed RSL evolution is guided by six basal sea-level index points, obtained from basal peat deposits sampled at their very base and for which indicators clearly testified to brackish forming conditions. The global RSL envelope (grey envelope shown in Figure 5) was derived from RSL predictions produced by GIA models (model ICE-5G of Peltier (2004) and model of Kuchar et al. (2012)), whose results were adjusted by best-fit polynomial curves. The linear terms of the latter were then modified to fit regional basal deposits as suggested by Horton and Shennan (2009).

Storm events are extracted from the stratigraphy (see Table 2). Truncated peat summits are considered as lower limits on age for storm events (pre-storm surge or pre-STS). Post-dating sedimentary units or soils are considered to be upper age boundaries for storm events (post-storm surge or post-STS). Locally, those discontinuities may last up to 1000 years (see ‘Pors Carn – Lescors marsh and Lajoie marsh’ and Figure 5) As 36 of 56 dates employed were derived in the literature, their positioning relative to erosional discontinuities needs to be viewed with caution. Marine shell dates are used as lower age limits as they are commonly reworked. Nevertheless, whole valve preservation is a fair argument for limited displacement. Despite this, results herein demonstrate that fragile shells such as dune gastropods may survive a major storm and thus be responsible for age anomalies.

Stratigraphy at the Bay of Audierne

Stratigraphic context is first required to understand the setting of the coastal deposits and the impact of recurrent storms on the record (Table 3; Figures 3 and 4). It includes our own data and older data from which stratigraphical positions have been reassigned thanks to our projections. Stratigraphic units are identified with their final labelling, considering the complexity of the record and truncation at a local scale related to basement irregularities. The presented transects are the deepest for the whole Bay of Audierne. Other attempts at drilling failed.

Basement rocks are considered here as unit A0. Weathered zones are often shaped in a palaeoplatform. The saprolite corresponds to unit A1. This is a constant for most of the sections and drilling while apparently it is weathered by the Holocene soil (Unit B1a) and occasionally cultivated early (Ap horizon, Unit B1b).

Pors Carn – Lescors marsh and Lajoie marsh

The coastal section at Toul Gwin (St Gwénolé) is present at the top of the bedrock and attests to Mesolithic occupation with agriculture during the Atlantic (Table 3 and Figures 3 and 4), at +1 m above the HWM. It is covered by quartzite gravels and grit layers, is slightly organic, and commonly observed along the coast interbedded or topping aeolian sands. On top of the dune and the gravel’s splays, several seaweed furnaces (from the end of the 19th century) are preserved with a very late thin sand cover.

Pors Carn – Lescors drilling transect

Pors Carn – Lescors drilling transect from 2008 starts at the coastal sections of bay (Figures 2 and 3). This transect is based on seven drillings reaching the rock basement, from the coast to the middle of the Lescors marsh. It has been supplemented with data from Carter et al. (1993; Figure 2). Our transect crosses a basement relief outcropping at +6.5 m NGF. This rock is eroded to the NW by a palaeocliff in sands, 4 m high, pasted with organic grounds.

Beg-an-Dorchenn breach, BD1 drilling and Tronoën drilling transect

The lowermost outcropping sandy unit (C2b) is usually preserved between Beg-an-Dorchenn and Tronoën, at the level of Kerdrafic place (Table 3 and Figure 3). Beach sandstones were regularly observed prior to 1998 on the lower shore and beach north of Beg-an-Dorchenn during stormy periods, especially at Prat ar Hastel (D2). These sandstones crop out as isolated blocks of calcareous beach conglomerate that is matrix supported and includes quartzite and flints gravels (first generation), or as slightly consolidated sands (second generation). Shell remains are rather weathered, particularly Cerastoderma edule, although mussels’ shells are well preserved at the base of the second generation of sandstones (D2).

At approximately 1 km to the north of Beg-an-Dorchenn, new investigations were undertaken from summer 2004 to spring 2010 along an intermittent rivulet, the Poulguillochet. Since August 2009, a breach has been excavated both by flooding and wave action. It is constituted by stratified coarse sands (units D3b) cropping out at the base of coastal sections over approximately 2 km, from Beg-an-Dorchenn to the north of Tronoën. These observations were completed 30 m landward by the deep drilling BD1. Towards the north, half-way to the Tronoën transect, two gravel layers end the section (units D3a and D3d) that are interbedded with dune sands. They crop at +8 m NGF. A similar record exists in several places but is less complete. At least four successive storm ridges are stacked at the coast close to the Beg-an-Dorchenn breach, from c. ad 700 (D2) to the 18th century (D3b), contemporaneous with the building of the Pors Carn dunes (Figure 3).

Tronoën transect

The Tronoën transect (Figure 3) is somewhat comparable to those of Beg-an-Dorchenn and Pors Carn with twinned gravel layers topping the dune barrier and the German occupation ground. The base of the transect preserves basal ‘peaty soils’ and colluvium. Inland, a gravel beach polluted by loams and resting on the saprolite is preserved at +12 m NGF (TR4), below a sandy orange palaeosoil (Figure 3). Located close to the base of the palaeocliff, it probably represents some remains of the last interglacial complex.

North Bay of Audierne

At the North Bay of Audierne, the data of Carter et al. (1993) for the Nerizellec marsh attest to an older stabilisation close to −1 m NGF, with peat growth fitting our observations from the Pors Carn – Lescors transect. It implies the building of a barrier from c. 1100 bc, considered here as unit C2b. It most likely reactivated the dune of Gwendrez, built from c. 3500 bc (Haslett et al., 2000) and considered here as unit C2a. This barrier probably migrated from the north. The coast is today in erosion from Trunvel breach to the north, with a historical stratified storm cobble splay directly covering the early cultivated soils (Penhors).

Archaeological remains

Archaeological remains provide complementary information. They are well preserved in the sector of Beg-an-Dorchenn and St Urnel (present-day isolated St Saturnin chapel, +20 m NGF), especially since the Mesolithic (Giot and Monnier, 1977, 1978). An excavation of late-Mesolithic kitchen midden on Beg-an-Dorchenn yielded dates of 4640–5550 cal. bc (Dupont et al., 2010). Several Neolithic and Chalcolithic settlements and fire places were dated on Beg-an-Dorchenn and surroundings, as close to Trunvel in age, attesting to a coastline further to the west than today. Sand dunes further invaded St Urnel in late Roman times (Giot and Monnier, 1978), constituting the ‘old dunes’. A second dune invasion seems to have discretely developed first, before being interrupted by palaeosoils between ad 800 and 1000 (Giot and Monnier, 1977), and mostly during the second half of the 20th century.

Discussion

Regional sea-level evolution combined with storminess record

Dates in BP will now be used for comparison purposes with long coastal and climate records: conversions are available in Table 2. Interpreted trends in Holocene sea-level change have been controversial for some time. One group following Fairbridge (1961) considered that sea level reached its maximum at approximately 6000 BP, about 1 m above the present-day level (Guilcher, 1948), and moreover, that it presents an oscillatory pattern (Behre, 2007; Fairbridge, 1961; Morzadec-Kerfourn, 1974). Others consider a continuous rise under decelerating rates (Baeteman, 2008; Bungenstock and Weerts, 2010; Clavé et al., 2001; Jelgersma, 1961; Orford et al., 2000; Pontee et al., 1998; Shennan and Horton, 2002). Today, numerous studies indicate that large-scale (pluri-decimetric to metric) oscillations observed in sea-level reconstructions are the expression of disturbances induced by local morpho-sedimentological activity and post-depositional lowering processes (such as compaction), and hence do not reflect real changes in RSL (Baeteman, 1999; Bungenstock and Schäfer, 2009; Bungenstock and Weerts, 2010; Gandouin et al., 2007; Horton and Shennan, 2009; Massey et al., 2006). Effects of storminess on Holocene sedimentary sequences are particularly important from the middle-Holocene period, where RSL reached a position close to the actual one. Meurisse-Fort (2009) showed that, in northern France, Holocene RSL rise was mostly achieved around 5900–5850 BP and that the subsequent rising of sea level was perturbed by the increased strength of storm surges, particularly after the Roman times. In short, reconstructed RSL for the Western Brittany region shows progressively decelerating rates of RSL rise, with clear inflexion points around 7000, 6000, 4500 BP and around the 3000–2000 BP period, without any distinct oscillation or stepwise character (Figure 5). Altitudinal discrepancies between the sea-level index points were found to be closely linked to compaction (Goslin, 2014).

As far as low storm abrasion surfaces are concerned, compaction may have a null or negligible impact as most of these surfaces developed within sand layers. It is not the case for storm terraces often cut in peat steps, for which the timing of the compaction process remains uncertain. In particular, it can be advocated that wave-breaking load on storm-revealed peat outcrops are a probable agent of over-consolidation of these deposits, with desiccation resulting from aerial exposure.

The stratigraphic sequence of the Bay of Audierne is rather discrete due to the recurrent storm abrasion surfaces and sediment reworking. Most low levels remain within the tidal amplitude (c. ≤−4 m NGF). This partly fits the observation of Lamb (1995), who argued for only c. 50 cm regression during cold events, as also shown for the period 1803–1945 by Jevrejeva et al. (2006) and as far as back as 1500 (Jevrejeva et al., 2008, 2009).

Extreme storm events are not only responsible for major breaking and erosion processes, especially in the intertidal zone (Fruergaard et al., 2013; Harris and Coleman, 1998; Houser et al., 2008; Sedgwick and Davis, 2003), but can also provoke thick sediment deposition on the upper shoreface over the course of a single extreme event as well as erosive process on the coast by offshore exportation (Goff et al., 2004). Bertin et al. (2012) have shown that the most aggressive storms (sting storms) form rapidly with young, short but energetic high waves, enhancing by 30% the potential for storm surge. These are probably the most efficient in terms of sediment exportation.

These could be responsible for major morphological changes, notably on the former back-barrier zone deposit from Beg-an-Dorchenn and other places in Brittany, from which information is today retrieved. From field and drilling observations, the former ‘high stands’ (Table 3 and Figure 6) seem to correspond to storm terraces that may form as a result of dispersive swash action (McKenna et al., 2012) from 1 to 10 m above the HWM in combination with high tides. In association with storm surge and also continental flooding, high levels are accumulated in lagoons in the form of washover fans, leading to pseudo-‘high’ stands, often consolidated by overwash (load casting, unit D3) or with a desiccation cracking pattern as occurs in south of Beg-an-Dorchenn.

Similarly, a lower altitude for storm ridges, as in BD1 drilling, may represent another feature inherited from storm impact (Table 3). Retallack and Roering (2012) suggested that storm-cut platforms develop at the level of the low water mark in soft material (LSWM; saprolite, shales or sediments). We may interpret the −5 m NGF (D1a) ‘low surface’ at the base of BD1 as a storm abrasion surface, formed during major storm(s) as the result of sediment exportation (Table 3; Figures 5 and 6). It promotes by enhancing the slope (lowering of the tidal flat) and by local liquefaction due to water-table resurgence (loss in cohesion), inland regressive erosion in rivers or estuaries and tidal scouring due to stronger tidal current that is also enhanced by renewed tidal flooding of back-barrier marshes after breaching (more important volume for ebb flow; Figure 6). Recurrent severe storms may accelerate the process (clustered storms). A rapid succession of major storms may maintain the low storm-cut platform long enough to favour a deep upstream regression along the rivers. After the storm, the restoration of the coastal ridge and of the normal beach profile may take some time and promote an infill and burial of the scours with a ‘transgressive’ sedimentary signature. Sediment exportation to the shelf by storms is responsible for starving the sediment coastal supply and thus promoting an inland retreat of the coastline. This clearly seems to be the case during the LIA.

During cold events, as during negative NAO events, precipitation is usually reduced in Europe (Hurrell, 1995) while aeolian aggradation may result from moderate but frequent storminess, favouring the development of coastal dunes on storm ridges (Hallégouët, 1981; Houser et al., 2008; Mauz et al., 2013). This should also be consistent with a limited regression (≤0.5 m; Jevrejeva et al., 2006; Lamb, 1995).

The stratigraphic record: Storminess versus sea level

From all field data as well as all archaeological information, south of the Bay of Audierne, the earliest record of transgression at the coast occurred between 7250 and 7050 cal. BP (recalibration of Dupont et al. (2010)) for kitchen midden deposits at Beg-an-Dorchenn. At this time, the HWM is considered between −7 and −2.4 m NGF following these authors. This confirms the occurrence of a peat at −5 m NGF dated 7150 cal. BP at Tronoën. On the northern coast of Brittany at the Anse du Verger, a salt marsh at −8 m NGF yielded c. 7750 cal. BP (Regnauld et al., 1996), corresponding to a HWM close to −7 to −6.5 m NGF. A tidal flat existed at 6600 cal. BP at −5 m NGF along the Rhuys Peninsula (Morbihan; Visset and Bernard, 2006). In sum, the results of this paper attest of a RSL close to −7/−6 m below the present level around 7000 cal. BP and close to −5/−4 m below the present level around 6500 cal. BP (Goslin, 2014). From a climatic point of view, this period is consistent with the thermal ‘Mid-Holocene Optimum’ (Viau et al., 2002). The site was temporarily abandoned before Neolithic occupation, from c. 7150 to c. 6850 cal. BP (recalibration of Dupont et al. (2010)), probably in relation to a period of enhanced storminess and the true entrance of the sea at Pors Carn in consideration of the new RSL curve. From these data and levelling, this situation favoured the development of a mudflat and salt marsh within the bay of Pors Carn that was exploited by Neolithics living on the Beg-an-Dorchenn Peninsula (kitchen middens). This seems somewhat earlier than the official slowdown of the Holocene transgression (see ‘Regional sea-level evolution combined with storminess record’; 5850–5800 cal. BP).

A gravel ridge reached +6.5 m NGF at Pors Carn (PC8). This ridge linked St Gwénolé Island to the continent by way of a gravel accumulation (Figure 2). This terrace is connected with a HWM close to + 2 m NGF (base of the palaeocliff, Figure 3). This ridge corresponds to a storm terrace (McKenna et al., 2012) formed between 7650 and 7050 cal. BP, and more likely closer to 7150 cal. BP, synchronous with the abandonment of Beg-an-Dorchenn. Following this event, the apparent RSL rise seems to slow down. From that period, the RSL envelope (Goslin, 2014; Goslin et al., 2013) indicates a progressively decelerating rise. From 5850 to 2950 BP, the record is poorly preserved in the south of Bay of Audierne. It seems that storminess as well as soil erosion have evacuated most of the pre-existing deposits (marine and loess), probably enhanced by the clearance of forest by Bronze Age cultivators and resulting groundwater logging (peat formation). Most of the coastal landscape was capped by thin ‘kaki’ loamy colluvium prior to the dunes or the marine invasion. It attests to bare soil erosion due to high precipitation. Despite being dated locally at 7050 cal. BP, it is more likely diachronic as it locally dates up to 3890–3981 cal. BP. The onset of dune aggradation is located in the Sub-Boreal at Gwendrez (north of Bay of Audierne; Figures 1 and 4), by dating of a humic soil buried below the dune to younger than 4950 cal. BP (Beta-114971; Haslett et al., 2000). In addition, terrestrial snails (Pomatia elegans) preserved at the base of a bioclastic dune at Penhat (Crozon Peninsula) attest to some Sub-Boreal dune activity (5585–4963 cal. BP; Hv-25113; Meurisse-Fort, 2009) due to local watershed conditions in relation with cooler and dryer climate conditions and/or active storminess. In the inner bay of Brest, the RSL seems to reach c. −1 m NGF for the HWM between 6000 and 4800 cal. BP (Stephan, 2011a, 2011b). It is also recorded in Morbihan (c. −1 m NGF HWM; Visset and Bernard, 2006). In the Mt St Michel’s Bay (Tessier et al., 2010), an erosion surface is located at −12 m NGF, fitting a +1.35 NGF HWM and is dated younger than c. 6450–5650 cal. BP, eroding the Holocene main transgressive system tracks. In the Kermor marshes close to the Ile-Tudy, a ‘brackish’ sandy peat resting on a freshwater peat yields 5191–5053 BP (Poz-36731) at +0.7 m NGF, perhaps attesting to the breach of coastal barriers by storms. The new RSL curve yield −2.4 m for this period or c. 0 NGF. This pseudo ‘regression’ (in the sense of a drop in RSL) could, in fact, result from strong storminess and deeper wave action. The Sub-Boreal peat development in St Omer marshes (northern France) is clearly linked to very wet and cool conditions (with evidence of frost activity synchronous with sedimentation; Gandouin et al., 2007) and regionally to a ‘marine regression’ along the Aa River.

This cooling and storminess seems to have been forced by the onset of the Neoglacial between approximately 5200 and 4200 BP with a glacial extension in Greenland and Scandinavia (Long et al., 2009; Nesje et al., 2008). This will be thoroughly discussed in part II of this paper.

A stable period is evidenced by the development of coastal marshes with continuous occupation at c. +1 m. NGF is observed close to Trunvel (Leroux, 1981) during the late Neolithic, the Chalcolithic and the Bronze Age. A low surface is recorded near −2.5 m NGF at c. 3950 BP by Morzadec-Kerfourn (1974) at Lampaul-Plouarzel-Trézien. This is followed regionally by a marine ‘invasion’ reaching an altitude close to the present mean sea level (MSL) c. 4050 BP, only preserved in the Mt St Michel Bay (Billard et al., 2013); it should also correspond to storms’ shaping of the topography as the new RSL attests to an altitude of −2 m or c. +0.5 m NGF at that time. Most of the ‘rapid sea-level variations’ seem mostly controlled by climate instabilities and storm recurrence.

Onset of the recent dune field (2850–450 BP): The ‘old’ dunes

From 2950 BP, the climate becomes unstable and stormy (Lamb, 1981). During the Hallstatt Celtic Iron Age (2750–2400 BP), the climate is rather fresh: a sharp, solar-forced cooling occurred, followed by an equatorward relocation of mid-latitude storm tracks (Beer and Van Geel, 2008). This corresponds to the ‘Homeric’ solar minimum (2900–2750 BP; Stuiver et al., 1998). The Beg-an-Dorchenn and St Urnel sites, located on highs, were occupied (Giot, 1974). At other places in Brittany, between 2800 and 2500 BP, Morzadec-Kerfourn (1974) and Stephan (2011a, 2011b) observed a potential ‘regression’ of −4 and −2 m, respectively, in non-tuned RSL. Again, this ‘regression’ may be interpreted in terms of a storm abrasion platform favouring regressive incision in unconsolidated sediments, and thus an artificial lowering of the RSL record (Goslin, 2014; Goslin et al., 2013).

Following this, the climate improved with the late Hallstatt Optimum. Beaches are essentially shelly to biodetritic and were lithified after 2143–2473 cal. BP (Gif-1100: Delibrias et al., 1971). It is located today at the outlet of rivers, close to −1.2 m NGF. These observations allow us to constrain in time the onset of calcareous dune building to at least from 2450 to 2350 cal. BP (Gif-1100 and Poz-31321), with a RSL estimated to have been only around 0.7 m below present (c. 1.7 m NGF; Goslin, 2014). This is somewhat later than as suggested by Guilcher et al. (1992; ‘from c. 1100 to 700 bc’). The largest dune body was formed at this time, reaching at least +6 m NGF in altitude. In parallel with main calcareous dune aggradation responsible for the construction of an important coastal barrier, colluvium (Poz-31322; 2353–2462 cal. BP) and peat began to accumulate in depressions such as the marshes of Lescors, La Joie and Nerizellec. Following this, the peat was truncated by an important storm surge c. 2350 cal. BP (pre-storm surge or STS, Table 2, Beta 57648; Carter et al., 1993), high upstream in the Grand Loch of Guidel (Morbihan) and also at Santec on the northern Brittany coast (Morzadec-Kerfourn, 1974). Most of these marshes turn brackish, as in Picardy (Meurisse-Fort, 2009) and East Anglia (Funnel and Pearson, 1989), with increasing storminess.

The onset of coastal barrier degradation and its relation with storminess (2400 BP – today)

Along the southern shore of the Bay of Audierne, a second phase of dune building seems to develop after onset of the La Tène Iron Age, c. 2400 BP. St Urnel church, Beg-an-Dorchenn and St Gwénolé harbour were still occupied (Giot, 1974). The new RSL is 0.7 m below the present one. This was in relation to a brief climate degradation, responsible for the formation of a storm ridge and for dune aggradation (Poz-31321: 2339–2487 cal. BP) with further reworking in Medieval times. Breaching of the coastal barrier and flooding reappear close to 1993–2140 cal. BP (pre-STS; SacA-23974 and SacA-23975) at Beg-an-Dorchenn and Tronoën. In the drillings TR1, TR5 and TR6 at Tronoën, a surface consolidated by storm swash is observed in an embayment close to 0 m NGF. A similar observation can be made in Morbihan at Fogéo (A9260, 1996–2158 cal. BP; Visset and Bernard, 2006).

After this dune aggradation, peat reappears in depressions during the Roman Optimum, such as in the marshes of Lescors and La Joie (Figure 2), attesting of the closure of the coastal ridge. In the Tronoën embayment, transformed at that time into a sandy lagoon protected by the coastal barrier, no evidence for peat formation is preserved. This is probably also the case for the Nerizellec marsh and other regional estuaries (Figure 1). A similar situation exists in the Canche estuary in northern France (Meurisse-Fort, 2009). During the Roman Empire Period, the system seems stable, with a RSL only slightly lower than today.

Lescors’ marsh presents a renewed breaching after 1638–1550 cal. BP (ad 312–400, pre-STS, Beta 57652; Carter et al., 1993) associated with washover fans. Sand drift fills the Tronoën lagoon from 1515 to 1459 BP (ad 435–491; pre-STS; SacA-13344). Sand drift is also responsible for burying St Urnel’s graveyard in late Roman times (money and ceramics dating to the 3rd century; Giot and Monnier, 1978). A burial place and a wooden chapel were regularly used at St Urnel from c. 1550 to 1150 BP (ad 400–800), despite some sand invasion (Giot, 1998; Giot and Monnier, 1978). This is also the onset of the most important migration of dunes inland, especially to the southeast, forming a wide splay reaching the Lescors’ marsh to the south. A storm terrace (+6.5 m NGF), abraded by swash, truncated the units A and B at the south of Beg-an-Dorchenn before being buried by dunes. N–S longshore currents raised by NW prevailing winds also fed sand in the Pors Carn bay. Dunes accumulated at Pors Carn, occasionally before 1328–1289 BP (ad 622–671) to c. 1050 BP (ad 800) with a short stabilisation close to ad 622–676 (Poz-31323; Poz-31324). The base of these Merovingian dunes in the Scilly Isles yields 1510–1630 BP (ad 380 ± 60, optically stimulated luminescence (OSL) date; Big Pool, St Agnes; Banerjee et al., 2001).

The shore was again eroded at Beg-an-Dorchenn. Located at a river outlet, the D2 beach rock fossilising a middle beach facies close to −5 m NGF on the shore of Kerdrafic yielded 1247–993 cal. BP (ad 703–957, Poz-34212; Figure 7). Similar assays (ad 688–795; Poz-36729) at −4 m NGF were obtained from the BD1 drill for lower beach sand resting on Bronze Age mud, also close to the Poulguillochet outlet. From this time, dunes were also built at Prat ar Hastel, cropping out up to +4.3 m NGF and presently lithified on the beach, including some drifted beach cobbles from the north of the bay. This ‘regression’ is mentioned from 1241 BP in NE Brittany (ad 709; Pelé, 1973), associated with the reclamation of the Marais de Dol. It seems more likely associated with a shift in continental conditions due to the successive stacking of storm ridges, around 1200–1150 BP (ad 750–800), the RSL being close to −0.3 m at that time.

After this period of littoral aggradation, successive storm terraces developed at c. 1050–1020 BP (ad 900–930), forming a washover fan (D5a1) in the BD1 drilling (MFS3). It was further buried by massive sand (D5a2), partly reworked from the c. 2475 BP ‘old’ dunes, which formed a storm barrier c. 1020 cal. BP (Poz-37202 and Poz-36728); the lagoon remained active until c. 1050–1010 BP (c. ad 900–960). This major storm was responsible in western Cotentin for a Viking ship wreck. Another major storm is recorded in Southampton in 1015 BP (ad 935; Marusek, 2010). Back-barrier lagoons were active and recurrently flooded with the construction of washover fans north of Tronoën, attesting to a sea level close to the present one. It was rapidly interrupted by two or three very important storm surges c. 900 BP (ad 1050; Figure 3), prior to the Medieval Warm Event.

These are well preserved on the isthmus and south of Beg-an-Dorchenn, invading the Pors Carn dune field to the SE over more than 600 m (grind and gravel splay) and up to 11 m in altitude. It probably breached the existing coastal barrier at different places, with a storm surge reaching +6 m NGF and with higher run up (load casts in section). Storm surges flooded the Poulguillochet and Lescors’ marshes. Coherent with these observations, slope deposits enriched in loess are preserved only above 9 m NGF on the Beg-an-Dorchenn Peninsula. These events are associated with sand drift, mostly during the second half of the 10th century, thus confirming the deduction of Giot and Monnier (1978).

A renewed stabilisation is recorded at Lescors during the ‘Medieval Warm Period’ (MWP). After this main phase of dune aggradation, the coastline mostly moves back inland (Figure 7). Dunes continue to migrate during the 17th–18th centuries in other places, particularly in NW Brittany, the Crozon Peninsula and the Loire Atlantique department (Guilcher and Hallégouët, 1991). A later series of storms is recorded in the 18th century. The present-day sea-level rise is mentioned regularly from 1899 at Penmarc’h (Monfort, 1895), consistent with information from tide gauges (Jevrejeva et al., 2006). The gravel ridge of the Bay of Audierne seemed to reactivate after 1775 (Hallégouët and Hénaff, 1993) in relation to recurrent great storms and N–S longshore drift. Furthermore, comparison between maps produced between 1771 and 1785 (1/40,000; Ingénieurs géographes) shows that the Beg-an-Dorchenn rock was practically an island in 1785, with the coastline slightly inland compared with the present-day situation (Figure 7). The pits of piddock shellfishes carved in the oldest beach sandstone at Beg-an-Dorchenn attest that an important coastal retreat took place during this period (c. ad 1800; ¹⁴C plateau effect). A cadastral map of 1828 also attests to a rapid retreat of the coast in the northern part of the bay.

Today, the sea level continues to rise (Cazenave et al., 2009), and coastal erosion continued between 1966 and 1988 at a rate of approximately 1 m/yr in the studied sector (Faye et al., 2007). Between 2000 and 2006, the absence of winter storminess limited coastal retreat (Hallégouët and Hénaff, 1993). Since 2007, the reappearance of important storms reactivated erosion south of the Bay of Audierne and tidal drift of sands around Beg-an-Dorchenn, allowing for the dune aggradation at Pors Carn. This was the motivation for the present study.

Storm events, extracted from this work, occurred c. 7150 BP, 5100 BP, 3500 BP, 2700 BP, c. 2350 BP, 2060 BP, 1550 BP (ad 400), 1150 BP (ad 800), 1050 BP (ad 900), 1020 BP (ad 930), 900 BP (ad 1050) and between ad 1700 and 1800. These events can be seen in Figure 5.

An accurate analysis at the scale of Brittany (see ‘Middle- to Late-Holocene Storminess in Brittany (NW France): Part II’) shall allow us to more precisely depict the historical events (ad 125 ± 50, ad 310 ± 50, ad 400 ± 50, ad 640 ± 50, ad 800 ± 50, ad 900 ± 50, ad 1010 ± 50, ad 1050 ± 50, ad 1100 ± 50, ad 1192 ± 50, ad 1335 ± 50, ad 1597 ± 50, ad 1703 ± 50, ad 1750 ± 50 and ad 1806 ± 50; ¹⁴C plateau, forced by local archives), although storm ages of about: 6600 ± 50 cal. BP, 6050 ± 50 cal. BP, 5500 ± 50 cal. BP, 4650 ± 50 cal. BP, 3400 ± 100 cal. BP, 2750 ±50 cal. BP, 2535 ±50 cal. BP, 2360 ± 50 cal. BP and 2060 ± 30 cal. BP may be extracted from an enlarged regional record (also reported in Figure 5). This will in turn allow us to target specific storm events from the historical archives.

Conclusion

Storm record preservation in Brittany is limited by subsidence and by a continuously rising sea level, which mostly truncated continental formations. The record is only discrete and residual for older periods of the Holocene. Most of the coastal sedimentation in the south of Bay of Audierne represents a polygenetic construction by storms, mostly from late Roman times with an apparently ‘unstable’ regional RSL. It is in fact mostly controlled by storm abrasion surfaces: the RSL rise seems regular, but decelerating at the regional scale. Furthermore, because of a limited sand supply, the sedimentary mass is reworked several times from the Iron Age, sometimes inland from the present coastline: highly disturbed regional RSL data from the Merovingian time tend to indicate the imprint of storm events on sedimentary system functioning.

Most large storm events fit with marked cooling periods. Most dunes form from the deflation of storm ridges and the refill of the beaches during post-storm restoral. This occurs with persistent winds from NW and dry conditions, similar to negative winter NAO. This is particularly true for the Merovingian Period and the LIA. Rainy periods associated with positive NAO events do not lead today to dune building. Indeed, this is because of the fact that prior to ad 400 or during the Middle Age Warm Period (MWP) a set of millennial storms was recorded, with one occurring immediately prior to the MWP, modifying considerably the coastal morphology of Brittany. At first glance, storm frequency appears to rise and reach a maximum during the 18th century, but the appearance of this trend is in fact biased by continual sea-level rise and the discrete preservation of the deposits (see part II). The maximum morphological efficiency of storms was reached during the LIA. Efficiency is raised by clustered storm events and durations of several days that act to increase the depth of wave action. This apparent enhancing of storminess is in agreement with orbital forcing and Holocene glacial history, although the record is biased by the limited historical archives and the erasing of storm traces induced by the rising RSL.

Footnotes

Acknowledgements

We thank the BRGM for its technical support for deep drillings and the CONSERVATOIRE DU LITTORAL and NATURA 2000 for allowing this study on protected natural sites. We also thank Stefan Lalonde greatly for the language improvement of this paper.

Funding

This study was funded by the French Research Program (ANR) COCORISCO. This study was also supported by the Program PHILTRE funded by the Brittany Region, and by the National Dating Program ARTEMIS.

References

Alexander

Tett

Jonsson

(2005) Recent observed changes in severe storms over the United Kingdom and Iceland. Geophysical Research Letters 32: L13704.

Baeteman

(1999) The Holocene depositional history of the Ijzer palaeovalley (western Belgian coastal plain) with reference to the factors controlling the formation of intercalated peat beds. Geologica Belgica 2: 39–72.

Baeteman

(2008) Radiocarbon-dated sediment sequences from the Belgian coastal plain: Testing the hypothesis of fluctuating or smooth late-Holocene relative sea-level rise. The Holocene 18: 1219–1228.

Banerjee

Murray

Foster

IDL

(2001) Scilly Isles, UK: Optical dating of a possible tsunami deposit from the 1755 Lisbon earthquake. Quaternary Science Reviews 20: 715–718.

Bärring

Fortuniak

(2009) Multi-indices analysis of southern Scandinavian storminess 1780–2005 and links to interdecadal variations in the NW Europe–North Sea region. International Journal of Climatology 29: 373–384.

Beer

Van Geel

(2008) Holocene climate change and the evidence for solar and other forcings. In: Battarbee

Binney

(eds) Natural Climate Variability and Global Warming: A Holocene Perspective. Hoboken, NJ: Wiley-Blackwell, pp. 138–162.

Behre

K-E

(2007) A new Holocene sea-level curve for the southern North Sea. Boreas 36: 82–102.

Bertin

Bruneau

Breilh

. (2012) Importance of wave age and resonance in storm surges: The case Xynthia, Bay of Biscay. Ocean Modelling 42: 16–30.

Bijl

Flather

de Ronde

. (1999) Changing storminess? An analysis of long-term sea-level data sets. Climate Research 11: 161–172.

10.

Billard

Bernard

Bouffigny

A, et al

. (2003) La Techniques et modes d’exploitation des pêcheries sur le littoral normand (France): Un essai de bilan de 10 années de travaux archéologiques. In: Daire

Dupont

Baudry

. (eds) Ancient maritime communities and the relationship between people and environment along the European Atlantic coasts / Anciens peuplements littoraux et relations homme/milieu sur les côtes de l’Europe atlantique, proceedings of the HOMER 2011 Conference, actes du colloque HOMER 2011, Vannes, 28 septembre - 1er octobre 2011. Oxford: Archaeopress, British Archaeological Reports, International Series, pp. 139–150.

11.

Bresson

(1924) Le « raz de marée » du 9 janvier sur les côtes de France. Anales de Géographie 33(182): 186–188.

12.

Bungenstock

Schäfer

(2009) The Holocene relative sea-level curve for the tidal basin of the barrier island Langeoog, German Bight, Southern North Sea. Global and Planetary Change 66: 34–51.

13.

Bungenstock

Weerts

HJT

(2010) The high-resolution Holocene sea-level curve for northwest Germany: Global signals, local effects or data-artefacts? International Journal of Earth Sciences 99: 1687–1706.

14.

Burningham

French

(2012) Is the NAO winter index a reliable proxy for wind climate and storminess in northwest Europe? International Journal of Climatology. Epub ahead of print 15 August. DOI: 10.1002/joc.3571.

15.

Buynevich

FitzGerald

van Heteren

(2004) Sedimentary records of intense storms in Holocene barrier sequences, Maine, USA. Marine Geology 210: 135–148.

16.

Carter

RWG

Regnauld

Jennings

. (1993) The impact of recent sea-level and climatic changes of the coast of northwest France. CEE Environment Programme, EPOCH project no CT 90 0015, pp. 1–37.

17.

Cazenave

Do Minh

Guinehut

. (2009) Sea-level budget over 2003–2008: A re-evaluation from GRACE space gravimetry, satellite altimetry and Argo. Global and Planetary Change 65: 83–88.

18.

Clarke

Rendell

(2009) The impact of North Atlantic storminess on western European coasts: A review. Quaternary International 195: 31–41.

19.

Clarke

Rendell

Tastet

J-P

. (2002) Late Holocene sand invasion and North Atlantic storminess along the Aquitaine coast, southwest France. The Holocene 12: 231–238.

20.

Clavé

Massé

Carbonel

. (2001) Holocene coastal changes and infilling of the La Perroche marsh (French Atlantic coast). Oceanologica Acta 24: 377–389.

21.

Dawson

Hickey

Mayewski

. (2007) Ice cores; climatic changes; deuterium; global temperature changes; atmospheric circulation; North Atlantic Oscillation; atmospheric pressure; ocean-atmosphere interaction; North Atlantic region. The Holocene 17: 425–432.

22.

Delibrias

Giot

Gouletquer

. (1971) Evolution de la ligne de rivage le long du littoral armoricain depuis le Néolithique. Quaternaria 14: 175–179.

23.

Dellwig

Böttcher

Lipinski

. (2002) Trace metals in Holocne coastal peats and their relation to pyrite formation (NW Germany). Chemical Geology 182: 423–442.

24.

Dellwig

Watermann

Brumsack

H-J

. (2001) Sulphur and iron geochemistry of Holocene coastal peats (NW Germany): a tool for palaeoenvironmental reconstruction. Palaeogeography, Palaeoclimatology, Palaeoecology 167: 359–379.

25.

Denton

Karleń

(1973) Holocene climatic variations – Their pattern and possible cause. Quaternary Research 3: 155–205.

26.

Dickson

Osborn

Hurrell

. (2000) The Arctic ocean response to the North Atlantic Oscillation. Journal of Climate 13: 2671–2696.

27.

Donnelly

Roll

Wengren

. (2001) Sedimentary evidence of intense hurricane strikes from New Jersey. Geology 29: 615–618.

28.

Engelhart

Horton

Vane

. (2013) Modern foraminifera, δ13C, and bulk geochemistry of central Oregon tidal marshes and their application in paleoseismology. Palaeogeography, Palaeoclimatology, Palaeoecology 377: 13–27.

29.

Dupont

Marchand

Carrion

. (2010) Beg-an-Dorchenn: Une fenêtre ouverte sur l’exploitation du littoral par les peuples mésolithiques du sixième millénaire dans l’ouest de la France. Bulletin de la Société Préhistoric Française 107: 227–290.

30.

Enio

MKS

Shanshuddin

Fauziah

. (2011) Pyritization of the Coastal Sediments in the Kelantan Plains in the Malay Peninsula during the Holocene. American Journal of Agricultural and Biological Sciences 6: 393–402.

31.

Fairbridge

(1961) Eustatic changes in sea-level. Physics and Chemistry of the Earth 7: 99–185.

32.

Faye

Hénaff

Hallégouët

(2007) Evolution récente de la ligne de rivage en Baie d’Audierne: De Penhors à la Pointe de la Torche. Penn ar Bed 199–200: 50–61.

33.

Fruergaard

Andersen

Johannessen

. (2013) Major coastal impact induced by a 1000-year storm event. Scientific Reports 3: 1051.

34.

Funnell

Pearson

(1989) Holocene sedimentation on the north Norfolk barrier coast in relation to relative sea-level change. Journal of Quaternary Science 4: 25–36.

35.

Gandouin

Van Vliet-Lanoë

Franquet

. (2007) Analyse en haute résolution de l’enregistrement de la transgression holocène dans un secteur subsident du littoral français: Le bassin-marais de Saint-Omer (Pas-de-Calais, France). Géologie de la France 1: 11–32.

36.

Giot

(1966) Chronique des datations radiocarbones armoricaines. Annales de Bretagne 73: 124–129.

37.

Giot

(1974) Chronique de préhistoire et de protohistoire finistériennes pour 1974. Bulletin de la Société Archéologique du Finistère CII: 9–12.

38.

Giot

(1981) Chronique de préhistoire et de protohistoire finistérienne pour 1980. Bulletin de la Société Archéologique du Finistère 109: 9–15.

39.

Giot

(1984) Chronique de préhistoire et de protohistoire finistériennes pour 1984. Bulletin de la Société Archéologique du Finistère 112: 9–18.

40.

Giot

(1998) La dune ancienne de la baie d’Audierne. Norois 45: 487–494.

41.

Giot

Monnier

(1977) Le cimetière des anciens bretons de Saint Urnel ou Saint Saturnin en Plomeur (Finistère). Gallia 35: 141–170.

42.

Giot

Monnier

(1978) Les oratoires des anciens Bretons de Saint Urnel ou Saint Saturnin en Plomeur (Finistère). Archéologie Médiévale 8: 55–93.

43.

Goff

McFadgen

Chagué-Goff

(2004) Sedimentary differences between the 2002 Easter storm and the 15th century Okoropunga tsunami, southeastern North Island, New Zealand. Marine Geology 204: 235–250.

44.

Goslin

(2014) L’évolution du niveau marin relatif le long des côtes du Finistère (Bretagne, France) de 8000 B.P. à l’actuel: entre dynamiques régionales et réponses locales. PhD Thesis, IUEM, Brest University, Brest, 14 February, 252 pp.

45.

Goslin

Van Vliet-Lanoë

Stephan

. (2013) Holocene relative sea-level changes in western Brittany (France) between 7600 and 400 cal. BP.P: Reconstruction from basal-peat deposits. Géomorphologie: Relief Processus Environment 4: 54–72.

46.

Guilcher

(1948) Le relief de la Bretagne méridionale de la Baie of Audierne de Douarnez à la Vilaine. La Roche-sur-Yon: Henri Potier, 682 pp.

47.

Guilcher

Hallégouët

(1991) Coastal dunes in Brittany and their management. Journal of Coastal Research 7: 517–533.

48.

Guilcher

Hallégouët

Meur

. (1992) Exceptional formation of present-day dunes in the Baie d’Audierne, southwestern Brittany, France. In: Carter

RWG

Curtis

TGF

Sheehy-Skeffington

(eds) Coastal Dunes. Rotterdam: Balkema, pp. 15–23.

49.

Hallégouët

(1981) Les crêtes littorales dunifiées du Massif Armoricain: Formation et évolution. Géographie physique et Quaternaire 35: 205–218.

50.

Hallégouët

(1990) Les dépôts pléistocènes marins de la côte méridionale du Finistère. Revue Archéologique d l est e et du Centre-EST (Suppl. 2): 45–51.

51.

Hallégouët

Hénaff

(1993) Evolution du littoral septentrionnal du pays Bigouden entre Penhors et Pors Poulhan. Le pays Bigouden à la croisée des chemins. Revue Cap Caval (Suppl. 17): 273–280.

52.

Hallégouët

Meur

Tanguy

(1989) Évolution du littoral de la baie d’Audierne (Finistère): La brèche de Trunvel. Bulletin du Centre de Géomorphologie de Caen 36: 13–16.

53.

Harris

Coleman

(1998) Estimating global shelf sediment mobility due to swell waves. Marine Geology 150: 171–177.

54.

Haslett

Bryant

(2007) Reconnaissance of historic (post-ad 1000) high-energy deposits along the Atlantic coasts of southwest Britain, Ireland and Brittany, France. Marine Geology 242: 207–220.

55.

Haslett

Davies

Curr

RHF

(2000) Geomorphologic and palaeoenvironmental development of Holocene perched dune systems in Brittany, France. Geografiska Annaler Series A: Physical Geography 82: 79–88.

56.

Hénaff

(2008) Recherche d’indicateurs géomorphologiques de variation d’orientation des houles sur les côtes françaises de l’Atlantique et de la Manche au cours des 25 dernières années, analyse des données et comparaison avec les données de l’atlas numérique de houle EDF-LNHE. La Houille Blanche 1: 61–71.

57.

Horton

Shennan

(2009) Compaction of Holocene strata and the implications for relative sea-level change on the east coast of England. Geology 37: 1083–1086.

58.

Houser

Hobbs

Saari

(2008) Post-hurricane airflow and sediment transport over a recovering dune. Journal of Coastal Research 24: 944–953.

59.

Hughen

Baillie

MGL

Bard

. (2004) Marine04, Marine radiocarbon age calibration, 26–0 ka BP. Radiocarbon 46: 1059–1086.

60.

Hurrell

(1995) Decadal trends in the North Atlantic Oscillation: Regional temperatures and precipitation. Science 269: 676–679.

61.

Ineson

Scaife

Knight

. (2011) Solar forcing of winter climate variability in the Northern Hemisphere. Nature Geoscience 4: 753–757.

62.

Jelgersma

(1961) Holocene sea-level changes in the Netherlands. Mededelingen van de Geologische Stichting: Series C 6(7): 1–100.

63.

Jevrejeva

Grinsted

Moore

(2009) Anthropogenic forcing dominates sea level rise since 1850. Geophysical Research Letters 36: L20706.

64.

Jevrejeva

Grinsted

Moore

. (2006) Nonlinear trends and multiyear cycles in sea-level records. Journal of Geophysical Research 111: C09012.

65.

Jevrejeva

Moore

Grinsted

. (2008) Recent global sea level acceleration started over 200 years ago? Geophysical Research Letters 35: L08715.

66.

Kemp

Vane

Horton

. (2012a) Application of stable carbon isotopes for reconstructing salt-marsh floral zones and relative sea level, New Jersey, USA. Journal of Quaternary Science 27: 404–414.

67.

Kuchar

Milne

Hubbard

. (2012) Evaluation of a numerical model of the British-Irish ice sheet using relative sea-level data: Implications for the interpretation of trimline observations. Journal of Quaternary Science 27: 597–605.

68.

Lamb

(1981) Climate from 1000 bc to 1000 ad. In: Jones

Dimble

(eds) The Environment of Man: The Iron Age to the Anglo-Saxon Period (volume 87 of British Series). Oxford: British Archaeological Reports, pp. 53–65.

69.

Lamb

(1995) Climate, History and the Modern World. 2nd Edition. London: Routledge.

70.

Lamb

Frydendahl

(2005) Historic Storms of the North Sea, British Isles and Northwest Europe. Cambridge: Cambridge University Press, 228 pp.

71.

Lambeck

(1997) Sea-level change along the French Atlantic and channel coasts since the time of the last glacial maximum. Palaeogeography Palaeoclimatology Palaeoecology 129: 1–22.

72.

Leroux

(1981) Informations archéologiques Bretagne. Gallia Préhist (ed CNRS) 24(2): 395–523.

73.

Long

Woodroffe

Dawson

. (2009) Late Holocene relative sea-level rise and the Neoglacial history of the Greenland Ice Sheet. Journal of Quaternary Science 24: 345–359.

74.

McKenna

Cooper

GAJ

Jackson

DWT

(2012) Storm swash terraces: A previously overlooked element of the cliff-shore platform system. Journal of Sedimentary Research 82: 260–269.

75.

Magny

Hass

J-N

(2004) A major widespread climatic change around 5300 cal. yr BP at the time of the Alpine Iceman. Journal of Quaternary Science 19: 1–8.

76.

Magny

Bégeot

Guiot

. (2003) Contrasting patterns of hydrological changes in Europe in response to Holocene climate cooling phases. Quaternary Science Reviews 22: 1589–1596.

77.

Massey

Paul

Gehrels

. (2006) Autocompaction in Holocene coastal back-barrier sediments from south Devon, southwest England, UK. Marine Geology 226: 225–241.

78.

Mauz

Hijma

Amorosi . (2013) Aeolian beach ridges and their significance for climate and sea level: Concept and insight from the Levant coast (East Mediterranean). Earth-Science Reviews 121: 31–54.

79.

Meurisse-Fort

(2009) Enregistrement haute résolution des massifs dunaires; Manche, mer du Nord et Atlantique – Le rôle des tempêtes (Thèse de Doctorat, Université Lille 1, Lille, 2007) Paris: Collection Recherches – Sciences Publibook, 312 pp.

80.

Monfort

(1985) Penmarch à travers ses historiens. Pont L’Abbé: Helio-Plans, 350 pp.

81.

Morzadec-Kerfourn

(1974) Variations de la ligne de rivage armoricaine au Quaternaire: Analyses polliniques de dépôts organiques littoraux, vol. 17. Rennes: Memoires de la Société géologique et minéralogique de Bretagne, 208 pp.

82.

Nesje

Bakke

Dahl

. (2008) Norwegian mountain glaciers in the past, present and future. Global and Planetary Change 60: 10–27.

83.

Orford

Wilson

Wintle

. (2000) Holocene coastal dune initiation in Northumberland and Norfolk, eastern UK: Climate and sea-level changes as possible forcing agents for dune initiation. In: Shennan

Andrews

(eds) Holocene Land-Ocean Interaction and Environmental Change around the North Sea. London: Geological Society of London (Special Publication no. 166), pp. 197–217.

84.

Peltier

(2004) Global glacial isostasy and the surface of the ice-age earth: The ICE-5G (VM2) model and GRACE. Annual Review of Earth and Planetary Sciences 32: 111–149.

85.

Peuziat

Giot

(1983) Le Néolithique des vieux sols de la baie d’Audierne. Bulletin de la Société Archéologique du Finistérienne 12: 21–26.

86.

Pirazzoli

(2000) Surges, atmospheric pressure and wind change and flood probability on the Atlantic coast of France. Oceanologica Acta 23: 643–661.

87.

Pirazzoli

Regnauld

Lemasson

(2004) Changes in storminess and surges in western France during the last century. Marine Geology 210: 307–323.

88.

Pontee

Tastet

Masse

(1998) Morphosedimentary evidence of Holocene coastal changes near the mouth of the Gironde and on the Medoc Peninsula, S.W. France. Oceanologica Acta 21: 243–261.

89.

Pye

Neal

(1994) Coastal dune erosion at Formby Point, North Merseyside, England: Causes and mechanisms. Marine Geology 119: 39–56.

90.

Pye

Tsoar

(2009) Aeolian Sand and Sand Dunes. Berlin/Heidelberg: Springer, 458 pp.

91.

Regnauld

Jennings

Delanney

. (1996) Holocene sea-level variations and geomorphological response: An example from northern Brittany (France). Quaternary Science Reviews 15: 781–787.

92.

Reimer

Baillie

MGL

Bard

. (2004) IntCal04 terrestrial radiocarbon age calibration, 26–0 ka BP. Radiocarbon 46: 1029–1058.

93.

Retallack

Roering

(2012) Wave-cut or water-table platforms of rocky coasts and rivers? GSA Today 22(6): 4–10.

94.

Sedgwick

Davis

Jr (2003) Stratigraphy of washover deposits in Florida: Implications for recognition in the stratigraphic record. Marine Geology 200: 31–48.

95.

Shennan

Horton

(2002) Holocene land- and sea-level changes in Great Britain. Journal of Quaternary Science 17: 511–526.

96.

Sibuet

Ondreas

Duval

. (1991) New informations in the crustal structure of the Celtic Sea and western approaches from gravity data and deep seismic profiles. In: De Bois

Gariel

Sibuet

J-C

(eds) Etude de la croûte terrestre par sismique profonde: Mer Celtique – Manche et ses approches occidentales, profils SWAT et WAM/sous la dir (issue 159 of Mémoires de la Société géologique de France). Paris: Mémoires de la Société géologique de France, pp. 145–166.

97.

Sorrel

Tessier

Demory

. (2009) Evidence for millennial-scale climatic events in the sedimentary infilling of a macrotidal estuarine system, the Seine Estuary (NW France). Quaternary Science Reviews 28: 499–516.

98.

Stephan

(2011a) Colmatage sédimentaire des marais maritimes et variations relatives du niveau marin au cours des 6000 dernières années en rade de Brest (Finistère). Norois 220: 9–37.

99.

Stephan

(2011b) Les flèches de galets de Bretagne: Evolutions passées, présentes et futures. Paris: L’Harmattan, Milieux naturels et Sociétés, 264 pp.

100.

Stuiver

Reimer

(1993) Extended 14C database and revised CALIB radiocarbon calibration program. Radiocarbon 35: 215–230.

101.

Stuiver

Reimer

Braziunas

(1998) High-precision radiocarbon age calibration for terrestrial and marine samples. Radiocarbon 40: 11–27.

102.

Tessier

Billeaud

Lesueur

(2010) Stratigraphic organisation of a composite macrotidal wedge: The Holocene sedimentary infilling of the Mont-Saint-Michel Bay (NW France). Bulletin de la Société Geologique de France 181: 99–113.

103.

Toulemont

(1964) Contribution à l’étude sédimentologique du Quaternaire du littoral sud-finistérien de la pointe du Raz à l’estuaire de l’Odet. Thèse 3ème cycle. Paris: Univers. Paris, 113 pp.

104.

Trouet

Scourse

Raible

(2012) North Atlantic storminess and Atlantic Meridional Overturning Circulation during the last Millennium: Reconciling contradictory proxy records of NAO variability. Global and Planetary Change 84–85: 48–55.

105.

Trouet

Esper

Graham

. (2009) Persistent positive North Atlantic Oscillation mode dominated the medieval climate anomaly. Science 324: 78–80.

106.

Van Vliet-Lanoë

Hibsch

Csontos

. (2009) Seismically induced shale diapirism: The Mine d’Or section, Vilaine estuary, southern Brittany. International Journal of Earth Sciences 98: 969–984.

107.

Van Vliet-Lanoë

Vandenberghe

Laignel

. (2002) Paleogeographic evolution of the western Europe during the Upper Cenozoic. Geodiversitas 24: 511–541.

108.

Viau

Gajewski

Fines

. (2002) Widespread evidence of 1,500 yr climatic variability in North America during the past 14,000 yr. Geology 30: 455–458.

109.

Visbeck

Hurrell

Polvani

. (2001) The North Atlantic Oscillation, present, past and future. Proceedings of the National Academy of Sciences of the United States of America 98: 12876–12877.

110.

Visset

Bernard

(2006) Évolution du littoral et du paysage, de la presqu’île de Rhuys à la rivière d’Étel (Massif armoricain – France), du Néolithique au Moyen Âge. Archéosciences 30: 143–156.

111.

Wolf

(2009) Coastal flooding: Impacts of coupled wave-surge-tide models. Natural Hazards 49: 241–260.

112.

Ziegler

(1992) European cenozoïc rift system. Tectonophysics 208: 91–111.

113.

Marusek

(2010) A Chronological Listing of Early Weather Events, 580 pp. Available at: http://www.breadandbutterscience.com/Weather.pdf.

114.

Pelé

(1973) Dol de Bretagne, revue d’informations municipales. Available at: http://histogen.dol.free.fr/dol/com_ind/marais.htm.

115.

Peltier

PSL

(n.d.) http://www.psmsl.org/train_and_info/geo_signals/gia/peltier/.

116.

Stuiver

Reimer

MPJ

Reimer

(2005) CALIB radiocarbon calibration: Execute version 5.0.2html. Available at: http://calib.qub.ac.uk/.