Past mining activities in the Vosges Mountains (eastern France): Impact on vegetation and metal contamination over the past millennium

Abstract

Coring was carried out in a soligenous marsh in the Vosges Mountains in the past mining district of Sainte-Marie-aux-Mines (eastern France). High-resolution palynological, non-pollen-palynomorph, and geochemical analyses were performed along the core. Correlations between the herbal composition of the landscape and trace metals in the core reveal a specific palynological pattern during mining activities. Two main periods of anthropogenic impacts on vegetation and trace metal contamination are shown: during the 16th–17th centuries, for mining and smelting activities, and the beginning of the 20th century, for smelting and the Industrial Revolution. No drastic deforestations occurred near the study site, contrary to historical descriptions and prints of the valley. Controlled forest practices were implemented from the beginning of the record, that is, since cal. AD 1000, so the impact of mining activities seems to be less significant than expected near mining sites. We demonstrate that the minerotrophic characteristics of the record closest to past mining sites allows for (1) the description of the landscape associated with anthropogenic activities and (2) the recording of past trace metal emissions without post-depositional mobility.

Keywords

geochemistry lead metallurgy palynology soligenous marsh Vosges Mountains

Introduction

Mining and smelting activities were the most important anthropogenic sources of trace metal (TM) emissions before the Industrial Revolution and the use of leaded gasoline (Nriagu, 1996). Pollution sources are multiple, and emissions of TMs contaminate water, soil, and the atmosphere. The temporal trends of TM emissions in the environment are documented in various archives around the world such as ice cores (Hong et al., 1994), peat bogs (De Vleeschouwer et al., 2010; Küttner et al., 2014), lacustrine sediments (Renberg et al., 1994; Schindler and Kamber, 2013; Thevenon et al., 2011), estuarine sediments (Alfonso et al., 2001), and other biological archives such as trees (Watmough and Hutchinson, 1996; Xu et al., 2014). Geochemical analyses of Swiss and French peat bogs reveal that 20% of lead (Pb) emissions occurred before Roman Times, 50% before the 18th century, and 90% during the past millennium (Monna et al., 2004; Shotyk et al., 2000). The nature and intensity of perturbations on biogeochemical cycles depend on the historical period of mining exploitation, which is linked to the type(s) of ore(s) smelted and extraction and smelting procedures. TM emissions are not the only impact of mining/smelting activities. By combining geochemical data with palynological analyses in peat bogs or other sedimentary archives, several studies have demonstrated that mining and smelting activities are linked to opening/clearing of the vegetation because of wood requirements for the smelting process and the intensification of agricultural production owing to increased populations (Bindler, 2006; Breitenlechner et al., 2010).

Peatlands are good terrestrial archives for reconstructing past environmental changes because of their ability to conserve physical, chemical, and biological indicators. TMs are conserved in ombrotrophic peat with very low mobility because they are mostly bound with humic acids (Dudare et al., 2011). Ombrotrophic peatlands, that is, peat bogs, receive all their water from precipitation; so, they have been widely studied to reconstruct TM atmospheric emissions over time throughout the world (Shotyk et al., 1996). Minerotrophic marsh trap TMs brought with water from local sources in adjacent areas as they are topographically supplied, and their water table is controlled by regional underground water resources. Although several studies underline the potential mobility of Pb in water-saturated peat deposits (MacKenzie et al., 1998; Urban et al., 1990), an experimental study (Vile et al., 1999) demonstrates that Pb is nonmobile in 95% of input despite changing redox conditions and drastic fluctuation of the water table.

Here, we present a multi-proxy paleo-ecological study from prominent ancient mining sites in the Vosges Mountains (eastern France) which were exploited at least during the last millennium. Poly-metallic ores from the Vosges Mountains contain among others Pb, silver (Ag), copper (Cu), arsenic (As), and zinc (Zn; Fluck, 2000b). In addition, this is an attractive region for smelting activities because of its dense forest cover. The regional impact of mining and smelting activities from this region has already been studied in two ombrotrophic peat bogs located 30 km from the main mining districts (Forel et al., 2010), but to our knowledge, no paleo-environmental study has been carried out close to archaeological mining sites. The aim of this study is to assess the local signal of past mining and smelting activities on the landscape and TM contamination at the scale of a mining district. Sainte-Marie-aux-Mines is a town where many mining and smelting archaeological sites were discovered (Fluck, 2000b), indicating intense mining/smelting activities for several centuries since, at least, the 10th century. We investigated a minerotrophic marsh representing 1000 years of sediment accumulation and situated just 3 km from the first mining/smelting archaeological sites. Lead is used to reconstruct past mining activities because of its presence as the element of interest in galena (PbS), the main ore exploited in the mining district of Sainte-Marie-aux-Mines, or as an element present in the mineral matrix of many poly-metallic ores. Moreover, Pb is considered to be nonmobile in peat and sediment cores, so it is considered the best TM to reconstruct past TM emissions (Benoit et al., 1998; Shotyk et al., 1997). Contrary to most paleo-pollution studies, seven TMs, comprising Pb, Ag, cadmium (Cd), Cu, mercury (Hg), As, and Zn, are analyzed here owing to the geological context. The past vegetation around the coring site was reconstructed using palynological and non-pollen-palynomorph (NPP) analyses. NPPs contain microfossils such as fungi spores or algae which can trace, for example, settlement activities as pastoralism, changes in local vegetation, or modifications in local trophic conditions (Van Geel et al., 2003). Associated pollen and NPP analyses enhance our understanding of local and regional landscapes because of the complementarity of these two proxies. The timing and intensity of local TM contamination and vegetation disturbance are also compared to historical mining events well documented by archaeological research or other historical sources.

Material and methods

Study area

Located in the Vosges Mountains (north-eastern France, Figure 1), in the department of Haut-Rhin (68), Sainte-Marie-aux-Mines was one of the biggest mining districts in France for Pb, Ag, and Zn during the 15th–16th centuries (Fluck, 2006). Ores from the valley contain Pb and Ag which are the main extracted metals. The valley is called the ‘Val d’Argent’, in reference to mining activities based on Ag and archaeologically dated from at least AD 937 (Fluck, 2000b). Extraction and transformation of ores continued for 1000 years and stopped in the mid-20th century. Two main parts of the valley are known to be exploited at different periods and are called the Altenberg (southeast of Sainte-Marie-aux-Mines) and the Neuenberg (south of Sainte-Marie-aux-Mines). The Altenberg was the only zone exploited up until 1550 and then the Neuenberg was exploited after the discovery of Ag veins.

Figure 1.

Location of the Le Hury site and map of archaeological mining/smelting sites.

Study site

The study site (48.23801°Ν, 007.24441°E; 650 m a.s.l.), called Le Hury (Figure 1), is close to Sainte-Marie-aux-Mines, that is, approximately 2 km, and to most of archaeological mining/smelting sites in the district. It is a marsh formed in a natural subcircular topographic depression of about 0.3 ha (approximately 60 m × 50 m), sustained by rainfall and slight surface runoff from the slope. Blocks of conglomerates, gneiss, sandstone, and granite have been deposited by solifluction. Sandstone is the dominant geological formation in the catchment area. Around the coring site, the forest is mainly composed of spruce (Picea abies), fir (Abies alba), and beech (Fagus sylvatica). The site is occupied by Carex brizoides group, birch (Betula pubescens), and alder (Alnus glutinosa) in an arboreal stage.

Sampling

A 71-cm-thick long core was taken from the deepest part of the marsh using a Russian GYP-type corer. The core was packed and stored in a dark cooling chamber (4°C) until analysis. In the laboratory, the core was cut into continuous 1-cm-thick slices, with a stainless steel knife. Five subsamples were taken every centimeter in order to analyze (1) pollen and NPPs; (2) TMs; (3) bulk density, water content, and loss-on-ignition (LOI); (4) carbon/nitrogen (C/N); and (5) particle size.

Chronology

Three ¹⁴C measurements were performed on bulk sediment samples at depths of 15, 50, and 60 cm using accelerator mass spectrometry at Beta-Analytics laboratory (Florida). Radiocarbon dates were calibrated using the IntCal09.¹⁴C calibration curve. Age–depth model was obtained using the Clam program developed by Blaauw (2010). The model was run as a linear regression, and the sedimentation rate was calculated using the age–depth model.

Palynological methods: Pollen and NPPs

Subsamples (1 g fresh matter) were prepared using the standard method established by Faegri and Iversen (1989). Subsamples were filtered to 200 µm and then treated with HCl 10%, HF 40%, NaOH 10%, and acetolysis. A minimum of 400 pollen grains were counted on each level at 400× magnification and identified using the standard identification keys of the Central European pollen flora (Beug, 1961; Faegri and Iversen, 1989; Moore et al., 1991; Punt, 1976; Punt and Clarke, 1980, 1981, 1984; Punt et al., 1988; Reille, 1992, 1995) and the reference collection of the Chrono-Environment laboratory. Only NPPs with known ecological significance were counted on the pollen slides. The main NPPs used in the results and discussion are summarized in Table 1. These NPPs were counted with a minimum of 100 NPPs, at a 400× magnification. NPPs were identified using the references cited in Table 1.

Table 1.

Main NPPs and their ecological signification.

Name	Author	Ecological signification
EMA-41	Prager et al., 2006	water
Coniochaeta cf. lignaria (HdV-172)	Cugny et al., 2010	dung related
Sporormiella (HdV-113)	Cugny et al., 2010	dung related
cf. Sordaria (HdV-55B)	Cugny et al., 2010	dung related
Cercophora (HdV-112)	Cugny et al., 2010	dung related
Coniochaeta B (TM-211)	Cugny et al., 2010	dung related
Delitschia spp.	Cugny et al., 2010	dung related
Trichocladium (HdV-572)	Dietre et al., 2012	tree cover
Cirrenalia	Dietre et al., 2012	tree cover
cf. Trichocladium opacum (TM-011)	Cugny et al., 2010	beech forest
cf. Endophragmiella C (TM-227)	Cugny et al., 2010	beech forest
EMA-3	Prager et al., 2006	alder carr
LCE-26	Dietre et al., 2012	pasture woodland
Dictyosporium	Dietre et al., 2012	dead wood
Valsaria-type (HdV-263)	van Geel, 1976	wet eutrophic conditions
HdV-85	van Geel, 1976	eutrophic conditions
HdV-20	van Geel, 1976	eutrophic conditions
HdV-18	van Geel, 1976	fund on Eriophorum vaginatum
HdV-16	van Geel, 1976	mesotrophic conditions in Sphagnum peat
Gaeumannomyces	van Geel, 1976	parasites of Carex
HdV-124	van Geel, 1976	helophyte marsh with eu- to mesotrophic conditions

EMA: Ernst-Moritz-Arndt, University Greifswald, Germany; LCE: Laboratoire Chrono-Environnement, Besançon, France; HdV: Hugo de Vries Laboratory, Amsterdam, the Netherlands; TM: Université Toulouse-le Mirail, Toulouse, France.

The calculation of the pollen and the NPP diagrams was performed using the program TILIA 1.5.12 (Grimm, 1991). The pollen sum was calculated from the arboreal and nonarboreal pollen count, while water plants, spores, and local trees Betula and Alnus, and regionally allochthonous planted trees Pinus were excluded from the pollen sum.

The percentage of each NPP was calculated relative to the sum of counted NPPs. Pollen assemblage zones (PAZs) were assigned based on the visual inspection of the pollen curves because of the low contents of some herbaceous taxa, which are major indicators of vegetation changes but are not detected with the depth constrained cluster analysis (CONISS). The zonation of the pollen diagram was adopted in the NPP diagram.

Geochemical methods

All the vessels in contact with the samples were carefully acid washed beforehand (HNO₃ 10% followed by HCl 10%) and then rinsed (ultrapure water).

Mercury analyses

Mercury concentrations were determined by cold-vapor atomic absorption spectrometry (CVAAS) on freeze-dried and ground matter using an AMA254 Hg analyzer (Altec, Czech Republic). The method detection limit based on repeated blank determinations was 0.02 ng Hg. Analytical accuracy and precision were determined by regularly measuring the following standard reference materials (SRMs) with certified Hg values: ERM-CD 281 of rye grass (sample nο. 0114) and INCT-OBTL-5 of oriental basma tobacco leaves. The recovery rates of the SRM were, respectively, 97% ± 10% and 94% ± 3%. Depending on the standard deviation of Hg concentrations, two or three measurements were taken for each sample.

Other elements: Lead, arsenic, cadmium, copper, zinc, and silver

Aqua regia digestion (HCl:HNO₃, 1:1, v/v) was applied on freeze-dried and ground subsamples, and metal concentrations were measured with the inductively coupled plasma mass spectrometer (Perkin Elmer Elan 6000/9000, AcmeLabs laboratory, Canada). A blind sample INCT-OBTL-5 of oriental basma tobacco leaves was sent with samples in order to check the results. Recovery rates of the SRM were between 95% and 110%, except for Cd with 125% and As with 120%. Concentrations are given in Supplementary Data 1 (available online).

The accumulation rate (AR, µg cm⁻² yr⁻¹) was calculated for each element every centimeter using the equation from Thevenon et al. (2011) which takes into account the dry bulk density (g cm⁻³), the sedimentation rate (cm yr⁻¹), and the concentration (µg g⁻¹) of the element (Eq. 1):

AR = concentration \times sedimentation rate \times dry bulk density

(1)

Physical and chemical characterization of the archive

Bulk density and water content

For bulk density, 1 cm³ of the subsample was taken using a stainless steel punch. Subsamples were freeze-dried to obtain dry bulk density. We also measured the water content by weighing the subsamples before and after freeze-drying.

LOI

The percentage of organic matter (OM), or ash content, and the carbonate content of deposits were identified by the standard LOI method according to Heiri et al. (2001). Subsamples were dried at 105°C during 20 h. All subsamples were cooled in a desiccator before burning. After weighing, subsamples were burned at 550°C during 4 h and weighed to calculate OM content.

C/N

Subsamples were dried at 60°C and ground in an agate mortar. The C and N content was determined by a VarioMAX CNS analyzer (Elementar). Two measurements were performed on sample every 10 measurements to check analyses.

Granulometric analyses

OM was removed from subsamples by H₂O₂ hot oxidation. Samples were sieved to 2 mm, and particle sizes were analyzed using a laser particle sizer Beckman Coulter LS230.

Statistical methods

Correlations between LOI and water content were made with the Spearman correlation test, as well as correlations between concentrations of the different TMs. Correlations between TM concentrations, some palynological taxa, pollen diversity, and LOI were also made with the Spearman correlation test. Pollen diversity was estimated by rarefaction analysis (Birks and Line, 1992) with ‘vegan’ R package (Oksanen et al., 2012). This method standardizes sample size and does not consider the abundance of different pollen types. Statistical analyses were performed using R.2.15.1 statistical software (R Core Team, 2012).

Results

Chronology

The age–depth model (Table 2, Figure 2) reveals that the sequence covers approximately the past millennium, that is, from cal. AD 1000 to the present day. The sedimentation rate is approximately 0.04 cm yr⁻¹ from a depth of 70 to 50 cm, 0.07 cm yr⁻¹ from 50 to 15 cm, and 0.25 cm yr⁻¹ in the uppermost 15 cm.

Table 2.

Radiocarbon dates of the Le Hury core.

Laboratory code	Depth (cm)	Material	¹⁴C-age (yr BP)	Calibrated calendar years (AD) 2σ calibration	Calibrated calendar years (BP) 2σ calibration
Beta-345126	15	Bulk	0 ± 20	Modern	Modern
Beta-345127	50	Bulk	420 ± 30	Cal. AD 1440–1520	Cal. BP 510–430
Beta-345127				Cal. AD 1570–1590	Cal. BP 380–360
Beta-345127				Cal. AD 1590–1630	Cal. BP 360–320
Beta-334184	60	Bulk	810 ± 30	Cal. AD 1220–1280	Cal. BP 740–670

Figure 2.

Age–depth model of the Le Hury core based on radiocarbon calibrated dates.

C/N, LOI, water content, dry bulk density, and particle size composition

At the opening of the core in the laboratory, sedimentary evolution was hard to describe: the sediment was dark brown along the core with coarse mineral fraction. In the absence of any real transition, several physical and chemical analyses were, therefore, undertaken to determine the nature of the sediment along the core and its changes.

In the deepest part of the core, that is, STC-1, the sediment is mostly composed of coarse mineral fraction (80%) with a high dry bulk density around 0.9 g cm⁻³ (Figure 3). C/N ratios are from 20 to 16 along the zone, and coupled with LOI measurements lower than 10%, indicate a greater OM decomposition. Water content is around 45% up to cal. AD 1400 where it decreases to 30% up to the end of the zone. This decrease corresponds to a rapid increase in small particle size fractions with silts and clays accounting for 40% and 4%, respectively.

Figure 3.

Physical and chemical results for the Le Hury core comprising LOI 550°C (%), C/N, dry bulk density (g cm⁻³), water content (%), clays (%), silts (%), and sands (%).

In the second zone, STC-2, LOI increases to 11% and water content rises to the same percentage as in the deepest part of the core, before the clay cap in STC-1. Dry bulk density decreases to 0.7 g cm⁻³. A slight increase in C/N occurs at the beginning of the zone, up to 20 (Figure 3).

In the third zone, STC-3, LOI increases up to 20%, while C/N remains stable (Figure 3). Water content increases up to 60% while dry bulk density decreases to 0.4 g cm⁻³. The percentage of silts and clays increases slightly while that of sands decreases to 65%.

STC-4 is characterized by a constant LOI, water content, and dry bulk density, but the particle size composition changes. A sharp increase in silts and clays was noted in the middle of the zone, while sands decrease (Figure 3). Two sharp increases of C/N up to 23 and 26 were noted during the period but do not correspond to the change in particle size composition.

The zone between cal. AD 1900 and 1950, that is, STC-5, presents a sharp increase in LOI (up to 50%) and water content (up to 80%). C/N and dry bulk density decrease, respectively, to 14 and 0.3 g cm⁻³. The mineral matter mostly contains sands up to 80% (Figure 3).

The last zone STC-6 is characterized by rapid and sharp changes. Particle size composition has a high proportion of silts and clays, more than 85% cumulated (Figure 3). LOI shows a particular pattern with a rapid increase (60%), decrease (30%), and increase (50%). C/N increases up to 24 and dry bulk density to 0.4 g cm⁻³. Water content is positively correlated to LOI along the core (ρ = 0.88, p < 0.001).

Pollen, NPP, and geochemical results

PAZs were noted from STC-1 to STC-6 from the base to the top of the core (Figure 4a and b). TM ARs in the Le Hury core are shown in Figure 5. To describe the evolution of ARs throughout the core, the same zonation as for the palynological diagram was used. The main palynological, NPP, and geochemical results are summarized in Table 3. A Spearman correlation matrix shows that all TMs are significantly positively correlated (ρ between 0.73 and 0.95, Table 4).

Figure 4.

(a) Simplified palynological diagram for the Le Hury core. Percentages were calculated relative to the amount of pollen excluding water plants (Sagittaria saggitifolia, Elatine, Nymphea, Typha, and Callitriche), spores, Betula, Alnus, and Pinus that are presented in number of pollen grains or spores counted. Empty curves represent exaggeration ×5. (b) Non-pollen-palynomorph diagram for the Le Hury core. Percentages were calculated relative to the amount of total NPP counted per sample. Pollen assemblage zones are taken from the pollen diagram.

Figure 5.

Trace metal ARs for the Le Hury core in full lines and exaggeration ×5 in dotted lines. Results are given in microgram square centimeter per year.

Table 3.

Pollen, NPP, and geochemical results for the Le Hury core.

PAZ	Cal. AD date	Pollen results	NPP results	Geochemistry results
STC-1	1000–1500	Dominance of Fagus and Abies with some deciduous taxa dominated by Corylus. Heliophilous taxa dominated by Urticaceae, Rumex and Plantago lanceolata. Cerealia-type is present with 13% such as Cannabis/Humulus and Centaurea cyanus. Large representation of monoletes spores. Pollen grains of water plants (Callitriche), alder carr indicators are present	Dominance of NPP forest indicators (50–70%). Presence of LCE-26 and HdV-85 indicating pasture woodlands and eutrophic conditions	TM fluxes are stable. No clear change in TM concentrations
STC-2	1500–1600	Arboreal taxa increase (55%), while herbaceous taxa decrease. Abies increases while Corylus and Fagus steadily decrease. Heliophilous taxa and agricultural indicators are still present but number of monoletes spores is divided by 1.5 along the period	Increase in NPP forest indicators (80%). Eutrophic condition indicators decrease	Increase in TMs with Pb AR between 0.9 and 2.5 µg cm⁻² yr⁻¹
STC-3	1600–1750	Arboreal taxa decrease while heliophilous taxa double. Potentilla increases to reach a level of 1%. Pinus increase. Monoletes spores continue to decrease, while triletes spores greatly increase in a short time	Decrease in forest indicators (20%). Increase in wetland indicators and HdV-16	ARs increase again during this period. Lead AR increases to 3.3 µg cm⁻² yr⁻¹ during 1 century (cal. AD 1650–1750)
STC-4	1750–1900	Arboreal taxa such as Corylus, Tilia, and Abies increase during this period. Heliophilous and herbaceous taxa which increased during STC-3 (10% and 45%, respectively) decrease here (5% and 30%, respectively). Presence of pollen grains of water plants	Coniochaeta B increases, and represents 55% of the NPP sum counted. Presence of trophic conditions indicator such as HdV-20	ARs return to values similar to STC-1, for all TMs
STC-5	1900–1950	Corylus, Abies and Picea decrease. The same heliophilous taxa as STC-3 increase. Some herbaceous taxa such as Potentilla, Polygonum aviculare, and Valerianaceae increase sharply in a short time	Pasture indicators decrease (5%) while NPP forest indicators increase (70%). HdV-16 and HdV-18 increase again as in STC-3	Rapid increase in ARs for all TMs. This leads to multiply the Pb AR by 15 in two decades; the maximum in Pb AR recorded is 215 µg cm⁻² yr⁻¹
STC-6	1950–present	Arboreal taxa increase with Picea and Fraxinus but Abies decreases. Agricultural indicators decrease	Increase in forest indicators	ARs decrease sharply but remain at levels higher than those of STC-1. ARs in all TMs increase again and decrease during the recent years, except for Zn

NPP: non-pollen-palynomorph; PAZ: pollen assemblage zone; TM: trace metal; AR: accumulation rate; HdV: Hugo de Vries Laboratory, Amsterdam, the Netherlands; LCE: Laboratoire Chrono-Environnement, Besançon, France.

Six PAZs are described according to Figures 4a and b and 5.

Table 4.

Spearman correlation coefficients between TM concentrations in the Le Hury core.

	Pb	Hg	As	Cd	Cu	Zn	Ag
Pb		0.81	0.85	0.80	0.92	0.91	0.96
Hg	***		0.73	0.78	0.80	0.78	0.77
As	***	***		0.84	0.91	0.91	0.91
Cd	***	***	***		0.90	0.87	0.84
Cu	***	***	***	***		0.91	0.94
Zn	***	***	***	***	***		0.95
Ag	***	***	***	***	***	***

TM: trace metal.

***

p < 0.001.

Statistical correlations between TMs and percentages of pollen

Because of correlations between TM concentrations, only Pb concentrations were used for further statistical analyses. The correlation is positive between Pb concentrations and pollen percentages for Potentilla (Figure 6), Polygonum aviculare, Hypericum, Poaceae, Cyperaceae, Fraxinus, Carpinus, Alnus, Salix, and Pinus. The correlation is negative for Tilia, Cannabis/Humulus, Cerealia-type, and Corylus (Figure 6). The correlation between TM and pollen diversity (by rarefaction analysis) is negative but not significant, and the same result was observed for the pollen percentage of tree taxa (Figure 6). LOI is positively and significantly correlated to Pb concentrations (Figure 6).

Figure 6.

Plots of correlations between Pb concentrations, Corylus, Potentilla, crop indicator taxa, tree taxa, pollen diversity (rarefaction analysis), LOI 550°C (%), and associated Spearman tests.

Discussion

General characterization of the archive: An efficient record to reconstruct paleo-pollution events

Ash content varies from 5% to 49% and is much higher than ombrotrophic peat bog values, such as those measured by Martı́nez Cortizas et al. (1997), who found an ash content varying between 1% and 8%. Moreover, the marsh is mainly composed of coarse mineral fraction, except in the top layer. The physical and chemical characteristics of the archive indicate that the Le Hury site is composed of a minerotrophic marsh. Concentrations of TMs in the sequence are greater at the surface than in deeper parts, thereby confirming that TM inputs come mainly from atmospheric deposition and not from the geological substratum (Shotyk, 2002). Most of the TM concentrations recorded have the same pattern as Pb but with different concentration levels. As all the TM concentrations are positively correlated, it can be considered that poly-metallic ore mining and smelting led to the emission of a set of TMs. Lead is considered to be nonmobile in sediment sequences, while other TMs could be mobile. Compared to other sequences from minerotrophic bogs (Monna et al., 2004; Shotyk, 1996), no post-depositional mobility was observed, even for TMs other than Pb. This indicates that the TMs did not move in depth in our core and that this minerotrophic marsh is, therefore, a good archive to reconstruct paleo-pollution events. TMs in the upper part of the profile are positively related to LOI because of the high affinity of TMs, particularly Pb, for OM (Shotyk et al., 1997). Nevertheless, the increase in TMs during the 14th–18th centuries does not correspond to an increase in LOI, as observed by Martı́nez Cortizas et al. (1997). OM content does not influence TM mobility in the Le Hury core which will be considered as an efficient record of past TM emissions.

Mining activities in the district of Sainte-Marie-aux-Mines: A chronology of vegetation changes linked to past industrial events

Cal. AD 1000–1500: The beginning of mining activities

This part of the archive has a high density because it is mostly minerotrophic because of the presence of a small stream supplying abundant mineral material at the beginning of the development of the archive. These sediments are separated from the water flow by a clay cap at 60 cm depth, that is, cal. AD 1400. The minor part of OM in the bottom of the marsh is greatly decomposed. Between cal. AD 1000 and 1500, the area was a marsh with fresh water, as indicated by the presence of aquatic plants dominated by Callitriche, a taxon with mostly amphibious species growing in eutrophic stagnant waters. Eutrophic conditions were also shown in NPP assemblages with the presence of HdV-85 (Van Geel, 1976).

Mining activities have been conducted in the Altenberg since at least cal. AD 937, based on radiocarbon dates of mined wood. This period corresponds to the establishment of Echery monastery where monks opened mines (Fluck, 2000b). During medieval times (10th–14th centuries), a continuous but low-intensity mining activity has been archaeologically dated in the Altenberg region where 13 mines were found. The Pb concentrations recorded at the onset of the sequence (23 µg_Pb g⁻¹) probably do not correspond to the natural TM background. A natural background in the Vosges Mountains was estimated by Forel et al. (2010) at less than 1 µg g⁻¹ for Pb in the deepest part of ombrotrophic peat bogs (situated on the crest of the Vosges Mountains). The concentration found in the bottom of the Le Hury core is high, indicating that the local geochemical background is influenced by the presence of metalliferous veins in the catchment area and also by the global metal pollution (Nriagu, 1996). No marked increase in TM ARs was noted during this time period, just some fluctuations. Forel et al. (2010) noted an increase of Pb in peat bogs around cal. AD 1100, indicating that Pb recorded in the Le Hury core derived from local and not from regional sources, highlighting the low-intensity mining activities in the Altenberg. Concerning the vegetation, pollen and NPP records indicate open mixed beech-fir woodland with the presence of pioneer taxa such as Corylus and fern spores. This vegetation picture is explained by forest grazing, as indicated by the presence of a NPP, LCE-26, linked to pasture woodlands (Dietre et al., 2012). Pollen representation of crop indicator taxa, Cannabis/Humulus and Centaurea cyanus, increased between cal. AD 1000 and 1500, probably because of crops in the valley because of the establishment of Echery monastery (Fluck, 2000b). The high presence of Alnus is because of a local alder carr, as shown by the presence of EMA-2, EMA-3, EMA-36, and EMA-38 in the NPP diagram (Barthelmes et al., 2012; Prager et al., 2006).

Cal. AD 1500–1600: The sharp increase in mining activities in the valley

The sediment characteristics indicate a rapid increase in C/N at the beginning of STC-2 because of the input of OM linked to small-sized particle matter, that is, clays and silts, by erosional process. A switch to mesotrophic conditions is observed with the alder carr in regression as indicated by NPPs, for example, HdV-16. These two events corroborate a period of leaching of less decayed OM from the marsh.

In cal. AD 1520, a sharp increase in mining activities is historically attested: in cal. AD 1545, 3000 miners were working in the ‘Val d’Argent’, with 12 smelters and 150 km of galleries (Fluck, 2000a). Moreover, during the 16th century, a new method of Ag extraction was used, the saigerprozess, which consisted in extracting Ag from tetrahedrites with Pb addition. These local mining activities were recorded with an increase in TM ARs. The related increase in Pb, up to 75 µg g⁻¹, was also reported by Forel et al. (2010) in two Vosgian ombrotrophic peat bogs with Pb concentrations up to 106 µg g⁻¹ in the Rossely core and 18 µg g⁻¹ in the Gazon-du-Faing core. Rossely is, like Le Hury, situated near a past mining district but in the southern part of the Vosges Mountains. The highest Pb concentrations in these two sites reflect the vicinity of TM sources. While historical records indicate that in cal. AD 1538, the forest of ‘val de Saint-Philippe’, in the mining district of Sainte-Marie-aux-Mines, was completely deforested by smelting activities, pollen records do not depict this situation. The study site remained forested and even more closed than previously, as indicated by an increase in the Abies curve, the disappearance of LCE-26 (a NPP pasture woodland indicator), and a decrease in fern spores. This shortage of wood is not noticeable in the palynological diagram of the Le Hury core, unlike in Galop et al. (2001) in the Pyrenees Mountains. Agricultural indicators are still present, indicating a double activity in the valley, that is, mining/smelting and crops.

Cal. AD 1600–1750: A period of fluctuating mining activities

This time period is archaeologically characterized by fluctuating mining activities (Fluck, 2000b) corresponding to a progressive increase in TM ARs in the Le Hury core. It also corresponds to the opening of the forest at the end of the 16th century. The increase in herbaceous taxa concerns in particular some taxa, such as Potentilla, Polygonum aviculare, Valerianaceae, and Brassicaceae. The forest seems to have been cleared with the creation of meadows as herbaceous taxa, and fern spores generally increase, while NPP forest indicators sharply decrease (Figure 4). The increase in Pinus representation is because of the openness of the vegetation which allows a better interception of pollen from very productive taxa. This opening probably results from wood exploitation for mining activities, as confirmed by the statistical correlation between Potentilla and other heliophilous taxa and Pb. The slight change in particle size composition could correspond to this forest opening with an increase in erosional process. Although the use of wood was high in the mining district (Fluck, 2000b), the regional pollen signal of beech-fir forest might also be high enough to conceal the local deforestation signal from the valleys (Mighall et al., 2002). Wood could have been cut in other valleys and transported to the mining district (Fluck, 2000b).

Cal. AD 1750–1900: Mining activities were still present in the valley

TM ARs decrease during the 18th century and return to values similar to those at the base of the core. Historically, this time period is characterized by a decrease in the intensity of mining and smelting activities because of property issues and difficulties in accessing the mine (Fluck, 2000b). During this period of apparent calm, however, there were two revivals in mining activities in cal. AD 1805–1812 and cal. AD 1822–1826 observed in the sequence by a slight increase in all TMs. Changes in particle size composition could denote an increase in erosional activity in the catchment area of the site, but this increase does not seem to be linked to a decrease in vegetation cover, as indicated by the increase in arboreal pollen representation, mainly Corylus and Fagus. NPP data show the opposite trend to pollen results: NPP pasture indicators increase, probably indicating very local pastures. The 18th century is marked by an increase in the forest pollen representation which corresponds to a decrease in mining activities.

Cal. AD 1900–1950: A decrease in mining activities during the Industrial Revolution

The organic composition of the marsh changed rapidly because of the input of decomposed OM. This part corresponds to the water supply zone of the marsh vegetation, with the highest water content measured a few centimeters below the surface.

The first half of the 20th century shows a sharp increase in all TM ARs. The maximum is reached in cal. AD 1953. Mining activities were still present in the district up to the first half of the 20th century but the number of mines decreased. During the first half of the 20th century, the palynological diagram shows a return to an open forest with the decrease in Fagus and Corylus and the increase in heliophilous herbaceous taxa (e.g. Potentilla). Abies increases probably because it is not exploited as heavily as Fagus, the preferred species for charcoal burning. Indeed, studies from the Vosges Mountains related to charcoal kilns from the same period demonstrate that Fagus was the preferred taxon for charcoal burning (Nölken, 2005). This time period is marked by the two World Wars (1914–1918; 1939–1945) which have been linked to the opening in the forested landscape, for example, by bombing, as shown by photographs illustrated by Balmier and Roess (2002).

Cal. AD 1950–2012: A record of modern industrial activities

The upper part of the marsh shows a high level of OM because of the presence of the local litter and vegetation. The beginning of the zone is characterized by an increase in LOI and a low C/N, indicating a large quantity of well-decomposed OM. The particle size composition changes with an input of smaller particles, probably because of the hydrological functioning of the marsh. This change in the physical and chemical composition of the marsh is synchronous with the sharp decrease in TM ARs, except Cd.

The last increase in ARs after cal. AD 1970 is synchronous with the C/N increase. These two events can be linked because TMs are generally linked to OM. However, we cannot exclude the possibility that the increase in Pb may be because of the use of leaded gasoline and the end of the Industrial Revolution. The sharp increase and decrease in TM ARs toward the surface concern all TMs, except Zn and Cd. Zinc is an essential element for living plants which probably explains its mobilization in the upper part of the archive (Martı́nez Cortizas et al., 1997; Rausch et al., 2005; Shotyk, 1996). This Cd pattern was recorded in peat cores in Europe (Coggins et al., 2006; Martı́nez Cortizas et al., 1997) and may be because of its high mobility in the top of the marsh in relation to water table movement (Cd solubility) and biological activity (Cd is analogous to calcium). Indeed, water content in the upper centimeters of the core is high, highlighting the importance of the solubility of some TMs, for example, Zn and Cd, and their concentration profile along the core. Even Ag seems to have been emitted at high levels during the Industrial Revolution and recent times as it was also recorded by Strnad et al. (2008) in a minetrophic peat near a historical Pb-Ag mine in the Czech Republic. Strnad et al. (2008) hypothesized that this recent increase in Ag was because of the global development of recycling precious metals from electronic components, such as computers. We cannot exclude the role of local industrial activities in the valley during the 20th century with a smelter at the beginning of the century and textile manufacturers. Concerning the vegetation, the transition to the last 50 years is characterized by the return to a dense forest. Crop indicators disappear as a result of changes in agricultural practices in the region: agricultural pressure seems to be lower and replaced by forestry exploitation. Land organization changed and crops were relocated to lower altitudes, while cattle breeding was translocated to mountainous areas (Schnitzler and Muller, 1998). Picea and Pinus increased during this time period because of high-altitude plantations for wood production (De Klerk, 2014).

Mining district of Sainte-Marie-aux-Mines: An ambiguous impact on vegetation

The correspondence between the archaeological and geochemical data of the Le Hury core validates the age–depth model. Indeed, data on the chronology of mining activities in the Sainte-Marie-aux-Mines district confirm the accuracy of the chronology of the sequence. Even if TM deposits are known to be generally linked with OM, it can be concluded from the correlation between archaeological and geochemical data that there was no post-depositional mobility of TMs in the marsh because of changes in its functioning. The positive correlation between LOI and Pb concentrations indicates that mining activities in the valley induced changes in land use that may have affected the functioning of the site. These changes in the functioning of the archive are synchronous with changes in mining activities and, therefore, in land use, reflecting the impact of human activities on the valley.

Regionally, over the past millennium, palynological and NPP diagrams depict a local landscape consisted of a beech-fir forest. This is the most common landscape described in the Vosges Mountains over the past three millennia (Kalis et al., 2006). However, the absence of intense deforestation related to mining activities, as reflected by the Le Hury record, has already been observed, for instance, during the Bronze Age in Mid-Wales (Mighall et al., 2002) or during the Middle Ages in the Alps (Breitenlechner et al., 2010). Several factors can explain the image given by the diagram of the Le Hury core. First, the surface of the archive is small, and this could lead to overestimation of the forest cover if the vegetation near the archive was significantly denser than the rest of the valley (Jacobson and Bradshaw, 1981). However, the exclusion from the pollen sum of the local trees which are, most of the time, overrepresented in the pollen diagram (Soepboer et al., 2010) will reduce this effect. Second, wood for smelting and carpentry may have been taken from other sides of the catchment, or the archive may have been a private or managed (rotation logging) plot with no or low clearing. Third, the species of trees in the plot may not correspond to the preferred species for charcoal burning (beech; Nölken, 2005). The time period of intensive mining activities is perceptible in the palynological diagram through the decrease in arboreal pollen and the increase in herbaceous and arboreal heliophilous taxa, as demonstrated by statistical correlations between TM concentrations and pollen percentages. Positive correlations between TMs and pollen taxa concern heliophilous taxa and provide evidence of human presence, as demonstrated in Ireland by Lomas-Clarke and Barber (2004) with the establishment of monks and the development of agricultural practices in the Middle Ages. In Mid-Wales during the Bronze Age and the Iron Age, there were several declines in arboreal pollen related to the increase in Cerealia-type, Plantago lanceolata, Rumex-type, and Potentilla (Mighall et al., 2002). Plantago lanceolata and Poaceae are linked to erosional signals, because of settlement activities, in Great Britain and Ireland (Lomas-Clarke and Barber, 2007). The similar palynological pattern found in the Le Hury core emphasizes interesting statistical correlations with a mining activity proxy, that is, TMs. The negative correlations concern forest pioneers and crop indicators. This indicates that the landscape opened during phases of atmospheric TM emissions. Conversely, when TM ARs decrease, forest pioneers increase, indicating the decrease in anthropogenic pressure and the increase in forest cover. Differential preservation of pollen taxa could also influence the picture given by the palynological diagram. Indeed, the high presence of spores could indicate the deterioration of other pollen taxa, principally in the deepest part of the core. However, spores are considered to come from the site, so their high presence does not influence the interpretation at the scale of the valley. Although the pollution-tolerance or pollution-sensitivity of these taxa have not been studied, the observed palynological pattern is probably not a response to TM contamination but rather a response to land-use modifications in the mountainous area because of changes in the needs of human populations depending on their activities. The palynological pattern related to mining activities is composed of pollen taxa which represent small percentages, for example, Potentilla. Statistical correlations combining pollen and geochemical analyses make it possible to identify these taxa which highlight changes in land use (Lomas-Clarke and Barber, 2007).

Conclusion

This multi-proxy study of a minerotrophic marsh has enabled the reconstruction of the history of the landscape linked to mining/smelting activities. The impact on the vegetation was not as pronounced as expected, in relation to ancient prints and other historical documents. Deforestations are absent in the vicinity of the archive indicating probable forest management in the catchment area for at least one millennium. However, the more open landscape is shown by the recrudescence of heliophilous taxa during the increasing phase of TM contents. In the historical archives, these phases correspond to intensified mining activities between cal. AD 1650–1750 and cal. AD 1900–1950 and, more recently, to industrial activities. The absence of high TM AR between cal. AD 1360 and 1500 and the concomitant increase in forest cover may correspond to agricultural decline because of the Hundred Years’ War. Combining palynological and geochemical methods at a local scale brings to light a paleo-ecological pattern specific to metallurgical activities. However, the forest cover increased in density between mining phases and the ecosystem recovered. While perturbations of the forest ecosystem by mining and smelting activities probably slowed down its natural evolution, it, nonetheless, seems to be resilient.

Footnotes

Acknowledgements

We thank the reviewers for their comments which helped to improve the manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This work was supported by a grant from the French ‘Agence de l’Environnement et de la Maîtrise de l’Energie’ (ADEME) and the Conseil Régional de Franche-Comté and a funding from the Universities of Franche-Comté and Bourgogne (BQR PRES).

References

Alfonso

Grousset

Massé

. (2001) A European lead isotope signal recorded from 6000 to 300 years BP in coastal marshes (SW France). Atmospheric Environment 35: 3595–3605.

Balmier

Roess

(2002) Scènes de tranchées dans les Voges. Strasbourg: Editions du Rhin.

Barthelmes

de Klerk

Prager

. (2012) Expanding NPP analysis to eutrophic and forested sites: Significance of NPPs in a Holocene wood peat section (NE Germany). Review of Palaeobotany and Palynology 186: 22–37.

Benoit

Fitzgerald

Damman

AWH

(1998) The biogeochemistry of an ombrotrophic bog: Evaluation of use as an archive of atmospheric mercury deposition. Environmental Research 78: 118–133.

Beug

(1961) Leitfaden der Pollenbestimmung für Mitteleuropa und angrenzende Gebiete. Stuttgart: Gustav Fischer Verlag.

Bindler

(2006) Mired in the past – Looking to the future: Geochemistry of peat and the analysis of past environmental changes. Global and Planetary Change 53: 209–221.

Birks

HJB

Line

(1992) The use of rarefaction analysis for estimating palynological richness from Quaternary pollen-analytical data. The Holocene 2: 1–10.

Blaauw

(2010) Methods and code for ‘classical’ age-modelling of radiocarbon sequences. Quaternary Geochronology 5: 512–518.

Breitenlechner

Hilber

Lutz

. (2010) The impact of mining activities on the environment reflected by pollen, charcoal and geochemical analyses. Journal of Archaeological Science 37: 1458–1467.

10.

Coggins

Jennings

Ebinghaus

(2006) Accumulation rates of the heavy metals lead, mercury and cadmium in ombrotrophic peatlands in the west of Ireland. Atmospheric Environment 40: 260–278.

11.

Cugny

Mazier

Galop

(2010) Modern and fossil non-pollen palynomorphs from the Basque mountains (western Pyrenees, France): The use of coprophilous fungi to reconstruct pastoral activity. Vegetation History and Archaeobotany 19: 391–408.

12.

De Klerk

(2014) Palynological research of the Vosges Mountains (NE France): A historical overview. Carolinea 72: 15–39.

13.

De Vleeschouwer

Le Roux

Shotyk

(2010) Peat as an archive of atmospheric pollution and environmental change: A case study of lead in Europe. The Holocene 51: 11–19.

14.

Dietre

Gauthier

Gillet

(2012) Modern pollen rain and fungal spore assemblages from pasture woodlands around Lake Saint-Point (France). Review of Palaeobotany and Palynology 186: 69–89.

15.

Dudare

Purmalis

Klavins

(2011) A study of major and trace element accumulation in humic acids. In: Proceedings of the 8th International Scientific and Practical Conference on Environment, Technology and Resources, Rēzekne, 20–22 June, vol. 1, pp. 76–82.

16.

Faegri

Iversen

(1989) Textbook of Pollen Analysis (4th Edition by Faegri Kaland

Krzywinski

). Chichester: Wiley.

17.

Fluck

(2000a) La lente agonie des mines de Sainte-Marie-aux-Mines (1633–1637). In: Colloque International ‘Les mines en Europe du Moyen-Age au XXe siècle’, Institut Catholique de Paris et UMR 858. Available at: https://hal.archives-ouvertes.fr/hal-00442268/document.

18.

Fluck

(2000b) Sainte-Marie-aux-Mines ou Les mines du rêve. Une monographie des mines d’argent. Soultz: Les Editions du Patrimoine Minier.

19.

Fluck

(2006) Les mines de plomb, cuivre et argent de la Renaissance ont généré des paysages contrastés. L’exemple de l’Alsace et des Vosges (France de l’Est). In: XIIIe Congrès International du TICCIH (The International Committee for the Conservation of the Industrial Heritage), Terni, 14–20 September.

20.

Forel

Monna

Petit

. (2010) Historical mining and smelting in the Vosges Mountains (France) recorded in two ombrotrophic peat bogs. Journal of Geochemical Exploration 107: 9–20.

21.

Galop

Tual

Monna

. (2001) Cinq millénaires de métallurgie en montagne basque. Les apports d’une démarche intégrée alliant palynologie et géochimie isotopique du plomb. Sud-Ouest Européen 11: 3–15.

22.

Grimm

(1991) Tilia and Tilia*Graph. Springfield, IL: Illinois State Museum.

23.

Heiri

Lotter

Lemcke

(2001) Loss-on-ignition as a method for estimating organic and carbonate content in sediments: Reproducibility and comparability of results. Journal of Paleolimnology 25: 101–110.

24.

Hong

Candelone

Patterson

. (1994) Greenland ice evidence of hemispheric lead pollution two millennia ago by Greek and Roman civilizations. Science 265: 1841–1843.

25.

Jacobson

Bradshaw

RHW

(1981) The selection of sites for paleovegetational studies. Quaternary Research 16: 80–96.

26.

Kalis

van der Knaap

Schweizer

. (2006) A three thousand year succession of plant communities on a valley bottom in the Vosges Mountains, NE France, reconstructed from fossil pollen, plant macrofossils, and modern phytosociological communities. Vegetation History and Archaeobotany 15: 377–390.

27.

Küttner

Mighall

De Vleeschouwer

. (2014) A 3300-year atmospheric metal contamination record from Raeburn Flow raised bog, south west Scotland. Journal of Archaeological Science 44: 1–11.

28.

Lomas-Clarke

Barber

(2004) Palaeoecology of human impact during the historic period: Palynology and geochemistry of a peat deposit at Abbeyknockmoy, Co. Gaiway, Ireland. The Holocene 14: 721–731.

29.

Lomas-Clarke

Barber

(2007) Human impact signals from peat bogs – A combined palynological and geochemical approach. Vegetation History and Archaeobotany 16: 419–429.

30.

MacKenzie

Logan

Cook

. (1998) Distributions, inventories and isotopic composition of lead in ²¹⁰Pb-dated peat cores from contrasting biogeochemical environments: Implications for lead mobility. Science of the Total Environment 223: 25–35.

31.

Martı́nez Cortizas

Pontevedra Pombal

Novoa Munoz

. (1997) Four thousand years of atmospheric Pb, Cd and Zn deposition recorded by the ombrotrophic peat bog of penido vello (northwestern Spain). Water, Air, and Soil Pollution 100: 387–403.

32.

Mighall

Timberlake

Clark

SHE

. (2002) A palaeoenvironmental investigation of sediments from the prehistoric mine of Copa Hill, Cwmystwyth, mid-Wales. Journal of Archaeological Science 29: 1161–1188.

33.

Monna

Galop

Carozza

. (2004) Environmental impact of early Basque mining and smelting recorded in a high ash minerogenic peat deposit. Science of the Total Environment 327: 197–214.

34.

Moore

Webb

Collinson

(1991) Pollen Analysis. 2nd Edition. Oxford: Blackwell.

35.

Nölken

(2005) Holzkohleanalytische Untersuchungen zur Waldgeschichte der Vogesen. Dissertation zur Erlangung der Doktorwürde der Fakultät für Biologie der Albert-Ludwigs-Universität Freiburg i. Br, 182 pp.

36.

Nriagu

(1996) History of global metal pollution. Science 272: 223.

37.

Oksanen

Blanchet

Kindt

. (2012) Vegan: Community Ecology Package (R package version 1.17-11) Available at: http://CRAN.R-project.org/package=vegan.

38.

Prager

Barthelmes

Theuerkauf

. (2006) Non-pollen palynomorphs from modern Alder carrs and their potential for interpreting microfossil data from peat. Review of Palaeobotany and Palynology 141: 7–31.

39.

Punt

(1976) Northwest European Pollen Flora, vol. I. Amsterdam: Elsevier.

40.

Punt

Clarke

GCS

(1980) Northwest European Pollen Flora, vol. II. Amsterdam: Elsevier.

41.

Punt

Clarke

GCS

(1981) Northwest European Pollen Flora, vol. III. Amsterdam: Elsevier.

42.

Punt

Clarke

GCS

(1984) Northwest European Pollen Flora, vol. IV. Amsterdam: Elsevier.

43.

Punt

Blackmore

Clarke

GCS

(1988) Northwest European Pollen Flora, vol. V. Amsterdam: Elsevier.

44.

R Core Team (2012) R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing.

45.

Rausch

Nieminen

Ukonmaanaho

. (2005) Comparison of atmospheric deposition of copper, nickel, cobalt, zinc, and cadmium recorded by Finnish peat cores with monitoring data and emission records. Environmental Science & Technology 39: 5989–5998.

46.

Reille

(1992) Pollen et spores d’Europe et d’Afrique du nord. Marseille: Laboratoire de Botanique Historique et Palynologie, URA CNRS 1152.

47.

Reille

(1995) Pollen et spores d’Europe et d’Afrique du Nord: Supplement 1. Marseille: Laboratoire de Botanique Historique et Palynologie.

48.

Renberg

Persson

Emteryd

(1994) Pre-industrial atmospheric lead contamination detected in Swedish lake sediments. Nature 368: 323–326.

49.

Schindler

Kamber

(2013) High-resolution lake sediment reconstruction of industrial impact in a world-class mining and smelting center, Sudbury, Ontario, Canada. Applied Geochemistry 37: 102–116.

50.

Schnitzler

Muller

(1998) Towards an ecological basis for the conservation of subalpine heath-grassland on the upper ridges of the Vosges. Journal of Vegetation Science 9: 317–326.

51.

Shotyk

(1996) Natural and anthropogenic enrichments of As, Cu, Pb, Sb, and Zn in ombrotrophic versus minerotrophic peat bog profiles, Jura Mountains, Switzerland. Water, Air, and Soil Pollution 90: 375–405.

52.

Shotyk

(2002) The chronology of anthropogenic, atmospheric Pb deposition recorded by peat cores in three minerogenic peat deposits from Switzerland. Science of the Total Environment 292: 19–31.

53.

Shotyk

Blaser

Grünig

. (2000) A new approach for quantifying cumulative, anthropogenic, atmospheric lead deposition using peat cores from bogs: Pb in eight Swiss peat bog profiles. Science of the Total Environment 249: 281–295.

54.

Shotyk

Cheburkin

Appleby

. (1996) Two thousand years of atmospheric arsenic, antimony, and lead deposition recorded in an ombrotrophic peat bog profile, Jura Mountains, Switzerland. Earth and Planetary Science Letters 145: E1–E7.

55.

Shotyk

Cheburkin

Appleby

. (1997) Lead in three peat bog profiles, Jura Mountains, Switzerland: Enrichment factors, isotopic composition, and chronology of atmospheric deposition. Water, Air, and Soil Pollution 100: 297–310.

56.

Soepboer

Sugita

Lotter

(2010) Regional vegetation-cover changes on the Swiss Plateau during the past two millennia: A pollen-based reconstruction using the REVEALS model. Quaternary Science Reviews 29: 472–483.

57.

Strnad

Mihaljevič

Ettler

. (2008) Distribution of the platinum group elements in peat deposit near a historic lead and silver mining district. Bulletin of Environmental Contamination and Toxicology 81: 159–163.

58.

Thevenon

Guédron

Chiaradia

. (2011) (Pre-) historic changes in natural and anthropogenic heavy metals deposition inferred from two contrasting Swiss Alpine lakes. Quaternary Science Reviews 30: 224–233.

59.

Urban

Eisenreich

Grigal

. (1990) Mobility and diagenesis of Pb and ²¹⁰Pb in peat. Geochimica et Cosmochimica Acta 54: 3329–3346.

60.

Van Geel

(1976) A palaeoecological study of Holocene peat bog sections, based on the analysis of pollen, spores and macro and microscopic remains of fungi, algae, cormophytes and animals. Academisch proefschrift, Hugo de Vries laboratorium, Universiteit van Amsterdam.

61.

Van Geel

Buurman

Brinkkemper

. (2003) Environmental reconstruction of a Roman Period settlement site in Uitgeest (The Netherlands), with special reference to coprophilous fungi. Journal of Archaeological Science 30: 873–883.

62.

Vile

Wieder

Novák

(1999) Mobility of Pb in Sphagnum-derived peat. Biogeochemistry 45: 35–52.

63.

Watmough

Hutchinson

(1996) Analysis of tree rings using inductively coupled plasma mass spectrometry to record fluctuations in a metal pollution episode. Environmental Pollution 93: 93–102.

64.

Tong

Stohlgren

(2014) Tree ring based Pb and Zn contamination history reconstruction in East China: A case study of Kalopanax septemlobus. Environmental Earth Sciences 71: 99–106.