Thủy ngân trong một đầm bùn ở Tây Ban Nha: Hồ sơ của biến đổi khí hậu và sự lắng đọng kim loại trong khí quyển
Tóm tắt
Một mảnh lõi bùn từ một đầm lầy ở tây bắc Tây Ban Nha cung cấp một bản ghi về sự tích lũy khí quyển của thủy ngân kể từ 4000 năm trước đây. Nghiên cứu cho thấy khí hậu lạnh thúc đẩy sự tích lũy gia tăng và sự bảo tồn thủy ngân có độ ổn định nhiệt thấp, trong khi khí hậu ấm áp có đặc điểm là sự tích lũy thấp hơn và sự chiếm ưu thế của thủy ngân có độ ổn định nhiệt từ trung bình đến cao. Bản ghi này có thể được chia thành các thành phần tự nhiên và nhân tạo. Thành phần thủy ngân nhân tạo đáng kể bắt đầu từ khoảng 2500 năm trước, gần thời điểm bắt đầu khai thác thủy ngân ở Tây Ban Nha. Thủy ngân nhân tạo đã chiếm ưu thế trong bản ghi lắng đọng kể từ thời kỳ Hồi giáo (thế kỷ 8 đến 11 sau Công nguyên). Các kết quả được trình bày ở đây có những tác động đến chu trình thủy ngân toàn cầu và cũng gợi ý rằng độ labil nhiệt của thủy ngân tích lũy có thể được sử dụng không chỉ để định lượng các tác động của hoạt động con người, mà còn như một công cụ mới cho việc tái tạo nhiệt độ cổ đại một cách định lượng.
Từ khóa
#thủy ngân #khí hậu #tích lũy #lắng đọng #đầm bùn #Tây Ban NhaTài liệu tham khảo
; S. A. Norton P. J. Dillon R. D. Evans G. Mierle J. S. Kahl in Acidic Precipitation vol. 3 of Sources Deposition and Canopy Interactions S. E. Lindberg A. L. Page S. A. Norton Eds. (Springer-Verlag New York 1990) pp. 73–102;
; J. M. Benoit W. F. Fitzgerald A. W. H. Damman in Mercury Pollution: Integration and Synthesis C. J. Watras and J. W. Huckabee Eds. (Lewis Boca Raton FL 1994) pp. 187–202;
; Geol. Surv. Can. Bull. 429 1 (1996);
Rasmussen P. E., Friske P. W. B., Azzaria L. M., Garrett R. G., Geosci. Can. 25, 1 (1997).
Martı́nez-Cortizas A., Pontevedra-Pombal X., Nóvoa Muñoz J. C., Garcı́a-Rodeja E., Water Air Soil Pollut. 100, 387 (1997);
. The PVO bog is located at the southernmost limit of ombrotrophic peat formation in Europe.
Mercury concentrations were measured in wet and dried samples with a LECO-ALTEC AMA-254 Hg analyzer (Altec Praha Czech Republic) connected to an automatic sampler. This equipment is a single-purpose atomic absorption spectrophotometer for Hg determination in solid and liquid samples without sample chemical pretreatment. It uses a Hg vapor generation technique and attains exceptionally high sensitivity. The detection limit is 0.01 ng of Hg and the working range is 0.05 to 600 ng. Each sample was measured in triplicate. Differences between replicates never exceeded 0.5 ng g –1 being always less than 1% of the mean value of the sample. Standard reference materials (SRMs) were run within each set of analysis. All the determined values were in the precision range for each SRM and mean reproducibility for triplicates was 5.2%. Radiocarbon age dating was previously described (4). All radiocarbon age dates presented here are given as conventional radiocarbon years before present ( 14 C yr B.P.).
Mercury and Hg minerals have been exploited for centuries in Spain with the first reference dating to 430 B.C. in the region of Sisapo (a Celtic word that means “the mine”) later identified as the Almadén area where vermilion (HgS) of first quality was extracted; mining began to be intensive during the Roman times (first century B.C. to fourth century A.D.) and metallurgy was introduced during the Islamic period (eighth and ninth centuries A.D.) (18). Available data on Hg production extend to 500 years ago.
Mercury pyrolysis studies of contaminated soils and synthetic standards [
Biester H., Zimmer H., ibid. 32, 2755 (1998);
] have demonstrated that Hg° is the only Hg species that has significant thermal desorption below 100°C whereas HgCl 2 and Hg bound to humic acids have maximum Hg releases at much higher temperatures (200° to 300°C).
I. Font Tullot Historia del Clima de España: Cambios Climáticos y sus Causas (Instituto Nacional de Meteorologı́a Madrid 1988); J. C. Garcı́a Codrón Un Clima para la Historia Una Historia para el Clima (Servicio de Publicaciones de la Universidad de Cantabria Lecciones 2/96 Santander Spain 1996).
J. M. Grove The Little Ice Age (Methuen London 1988)
It has been suggested that mosses may accumulate metals in at least three different ways: by accumulation over their entire surface by retention on cell walls through ion exchange and by intracellular uptake [D. H. Brown in Bryophyte Ecology A. J. E. Smith Ed. (Chapman and Hall New York 1982) pp. 383–444;
Gstoettner E. M., Fisher N. S., Water Air Soil Pollut. 93, 321 (1997);
]. Our results for cold climates may be explained in terms of enhanced dry deposition of Hg that is accumulated on the surface of the mosses largely as particulate noncomplexed Hg (Hg°) of low thermal stability.
The shift to more thermally stable species with increasing humidity (net surface moisture) can be the result of an increase in the input of Hg chlorides. The bog is located only 20 km south of the coast and more precipitation would correspond with an enhanced marine influence (especially the ionic strength of wet deposition).
Mackay D., Wania F., Schroeder W. H., ibid. 80, 941 (1995).
; W. Fitzgerald and R. Mason in Global and Regional Mercury Cycles: Sources Fluxes and Mass Balances vol. 21 of NATO ASI Series 2 W. Baeyens R. Ebinghaus O. Vasiliev Eds. (Kluwer Academic Norwell MA 1996) pp. 85–108.
To obtain this function we performed a regression analysis (SPSS version 6.1 SPSS Chicago IL) on 24 samples with ages older than 2200 years which is before the beginning of the Roman exploitation of Almadén. A stepwise regression method was used with F probabilities of 0.05 and 0.10 (in and out respectively). Analysis of variance and Student t analyses indicated that the regression function and the regression coefficients of independent variables are significant ( P < 0.01). The correlation coefficient is 0.97 and the standard error of the estimate is 2.2 ng g −1 .
The Hg flux calculation is based on accumulation and has to be considered as a crude estimation of the actual Hg flux because accumulation is not exclusively the result of deposition processes but is the result of a balance between inputs to and outputs from the bog. Our data suggest that part of the deposited Hg is reemitted to the atmosphere with the proportion increasing with increasing temperature.
Hernández Sobrino A., Geohistoria 7, 67 (1993).
C. Domergue Les Mines du la Péninsule Ibérique dans l'Antiquité Romaine vol. 127 of Collection de l'Ecole Française de Rome (l'Ecole Française de Rome 1990).
___ Ed. Minerı́a y Metalurgia en las Antiguas Civilizaciones Mediterráneas y Europeas (Ministerio de Cultura Madrid 1989); A. Hernández Sobrino Actas de las I Jornadas sobre Minerı́a y Tecnologı́a en la Edad Media Peninsular (Fundación Hullera Vasco-Leonesa y Sociedad Española de Esudios Medievales León 1996) vol. I pp. 384–394; A. Matilla Desde 1646 a 1799 vol. II of Historia de las Minas de Almadén (Minas de Almadén y Arrayanes–Instituto de Estudios Fiscales Ministerio de Economı́a y Hacienda Madrid 1987);
Maturana S., Hernández Sobrino A., Bocamina 1, 38 (1995).
Ramil P., Gómez L., Muñóz C., Fervedes 3, 33 (1996);
; A. Pérez Antelo ibid. p. 151.
Discriminant analysis was performed with the SPSS software package version 6.1. After assigning the climate class to each sample we ran significance tests ( F and Wilk′s lambda) for the equality of class means for each variable which resulted in significant differences between climate classes for H Hg L Hg L Hg / H Hg and M Hg / H Hg ( P < 0.01). Three canonical functions were needed to explain the variance with the first function ( F 1) accounting for 69% the second one ( F 2) accounting for 27.5% and the third one ( F 3) accounting for 3.5%. Canonical correlations for these functions are 0.95 0.89 and 0.59 respectively; thus the first two functions account for almost all the variance (96.5%) and have correlation coefficients suggesting that an optimum classification can be obtained with them. The structure matrix revealed that correlations between the discriminant variables and the canonical functions are –0.93 and –0.73 for L Hg and L Hg / H Hg respectively to F 1 and 0.73 and 0.67 for M Hg / H Hg and H Hg respectively to F 2. Using the discriminant scores and applying Bayes' rule we estimated the probability of each sample belonging to a climate class. Five of 80 samples have a discriminant score for which the probability of belonging to a climate class that is different from the one it was previously assigned is highest; these samples are misclassified. The other 75 samples have very large probabilities of belonging to the class that they were assigned and are considered to be correctly classified.
This result may be associated with the decrease in the water content of the atmosphere as temperature drops; therefore the colder the climate the drier it tends to be. But this cannot be fully assessed because PVO is located in an oceanic region only ∼20 km from the sea and there is no clear evidence of an intense humidity decrease in northwestern Spain during the cold spells of the late Holocene. For example during the LIA (the most recent cold period) extreme humid and dry conditions seem to have been of common occurrence (21) and on the basis of the results shown in Fig. 3 the average trend reflected by the F 2 scores is one of humidity that is similar to or only slightly higher than that at the present.
The TI values were calculated as ( F 1 i / F 1 1 )-1 in which F 1 i is the score of F 1 for each individual peat sample and F 1 1 is the score of the uppermost sample of the peat. The latter is a sample with a thickness of 2 cm which represents the past 25 to 30 years of peat accumulation based on the growth rate. At the place where PVO is located the mean annual temperature for this period is 10.3°C (5.4°C in winter and 16.5°C in summer) and annual precipitation is 1600 mm (575 mm in winter and 170 mm in summer) [
Martı́nez-Cortizas A., Castillo F., Geographicalia 33, 127 (1996)].
Andreu J. M., Rev. Acad. Galicia Cienc. 12, 119 (1993);
; J. M. Raso in Advances in Historical Climatology in Spain J. Martı́n-Vide Ed. (Oikos-Tau Barcelona 1997) pp. 91–102.
It is not certain whether the TI of the peat-vegetation interface sample is a correct value because it may be not in equilibrium with present conditions and it is still subjected to Hg volatilization. In this case the TI would be lower and a more reasonable estimate for the change during the present century would be an increase in TI on the order of 0.5 to 1.0 units.
Intergovernmental Panel on Climate Change El Cambio Climático (Ministerio de Obras Públicas y Transportes Instituto Nacional de Meteorologı́a—OMM/PNUMA Madrid 1991). The longest temperature record of the closest meteorological station A Coruña shows an average increase of 1.15°C since the beginning of the 20th century.
We thank Á. Hernández Sobrino and C. Llana for their fundamental help on the history of Hg exploitation in Spain and B. Blanco Pequeno for his helpful comments. The financial support of the Xunta de Galicia (project XUGA-2001B98) is sincerely appreciated. Special thanks to W. Chesworth and C. Schlüchter for reviewing an earlier version of this manuscript and thanks to two anonymous referees whose editorial technical and scientific comments greatly improved the manuscript.