Marine Chemistry 79 (2002) 133 – 153 www.elsevier.com/locate/marchem
Observations on trace element hypersaline geochemistry in surficial deposits of evaporation ponds of Exportadora de Sal, Guerrero Negro, Baja California Sur, Mexico Evgueni Shumilin a,*, Mario Grajeda-Mun˜oz a, Norman Silverberg a, Dmitry Sapozhnikov b a
Centro Interdisciplinario de Ciencias Marinas-IPN, Av. IPN s/n, Col. Playa Palo de Santa Rita, Apdo postal 592, La Paz, Baja California Sur 23096, Mexico b V.I. Vernadsky Institute of Geochemistry and Analytical Chemistry, Russian Academy of Sciences, Moscow, Russia
Abstract Trace element concentrations were determined for 28 samples of surficial deposits collected over a salinity gradient of 39x to approximately 250xin evaporation ponds of a large industrial NaCl production facility located near Guerrero Negro (Peninsula of Baja California, Mexico). Grain size and lithology are used to describe the mixed chemical – detrital material. Major earth and trace elements in the sediments were measured by a combination of an inductively coupled plasma mass spectrometry (ICPMS) for As, Ba, Cd, Cu, Mg, Ni, Pb, U, and Zn, flame atomic absorption spectrophotometry for Al, Fe, and Mn, isotope dilution ICPMS for Hg, and instrumental neutron activation analysis (INAA) for Cs, Ca, Sr, Cr, Sb, and Sc. Changes in grain-size, mineralogy, and chemical composition suggest that suspended particulate matter supplied to the system, mainly by seawater pumped in from the adjacent Ojo de Liebre Lagoon, and to a smaller extent, carried into the ponds by wind from adjacent dunes and by wave-induced erosion of retaining dikes, dominates the sediment in the first two concentration ponds. Freshly deposited fine sediments in these ponds are enriched in organic matter, biogenic carbonates, Al, Fe, Mn, and most of the trace elements, and are similar to surface sediments of the adjacent lagoon. In ponds 3 and 4, intensive evaporationinduced precipitation of calcium and magnesium carbonates occurs causing scavenging of Al, Fe, Mn carried by colloidal particles. The sediment in ponds 4 and 5 includes a thick microbial mat layer, which is enriched in organic carbon, and the accumulation of As and Cd is observed there. In ponds 7 through 11, the deposits generally are low in organic matter and trace elements (except for Sb) and are formed by intensive gypsum and anhydrite precipitation, with an admixture of terrigeneous material in ponds 8 – 10 supplied by a dyke erosion or carried in with wind from adjacent sand dunes. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Geochemistry of hypersaline environments; Trace elements; Surficial deposits; Baja California Peninsula; Guerrero Negro salt works
1. Introduction *
Corresponding author. Tel.: +52-612-112-53-44; fax: +52612-112-53-22. E-mail address:
[email protected] (E. Shumilin).
In hot, arid climates, evaporation greatly exceeds the supply of fresh water. In flat coastal zones, which are occasionally covered by high tides, and areas with
0304-4203/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 2 0 3 ( 0 2 ) 0 0 0 6 0 - 9
134
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
restricted exchange of seawater, such as shallow lagoons or isolated seas or lakes, hypersaline conditions commonly develop. Although the sequence of precipitation of the major salts during progressive evaporation of seawater has long been known (Van Hoff, 1912, cited by Borchert, 1965; Horne, 1969), only a few natural hypersaline environments have been studied from the point of view of their trace element geochemistry. Much information is however available from the Dead Sea (e.g. Herut et al., 1997, 1998; Nissenbaum, 1974, 1980; Stiller and Nissenbaum, 1999). In this exceptional environment, natural salinity is very high (ca. 340x ). Salinity of Red Sea brines is also elevated, and their chemistry is influenced greatly by hydrothermal solutions and exchange with ancient salt deposits (Anschutz and Blanc, 1996; Degens and Ross, 1969; Hartmann et al., 1998; Schoell and Faber, 1978). Borchert (1965) and Millero (1996) described the major ion chemistry of seawater concentrated by evaporation in some natural bays and lagoons of eastern Mexico. Zherebtsova and Volkova (1966) analyzed some major components and trace elements (Sr, Rb, and Li) in Black Sea waters, as well as in Sasyk – Sivash lake brines, which had been experimentally concentrated using solar evaporation, with special attention to the late stages of evaporation. Some other major and trace element information is available from sediments of a highly saline lake (Mihelcˇic´ et al., 1996) and from evaporation ponds created for industrial production of table salt (Horne, 1969; McCaffrey et al., 1987; Varnavas and Lekkas, 1996). Modeling of chemical equilibria and precipitation during the progressive evaporation of seawater has provided a good understanding of the major components of such systems (Borchert, 1965; Garrett, 1980; Weare, cited in Millero, 1996; Eugster et al., 1980; Zdanovskii, 1979), but information is still not adequate to explain trace component behavior, either singly or their patterns of co-precipitation with the principal salt minerals crystallizing in such environments. In this study we provide new information concerning trace element geochemistry in surface deposits of the hypersaline environment of a succession of large evaporation ponds of the Exportadora de Sal salt works (Guerrero Negro, Baja California Sur, Mexico). This facility is the second largest in world production
of solar salt. Of particular interest were the ponds with salinity between 60x and 160x , where large areas of the bottom are covered by microbial mats, and the possible influence of the microbial mats on trace element geochemistry. An additional motivation was concern over possible impacts of industrial salt production on Ojo de Liebre Lagoon, which is famous for the seasonal visits of Pacific grey whales for procreation and rearing of their pups. Data on the concentration and speciation of dissolved trace elements in brines in these ponds are presented in a separate paper (Shumilin et al., in press).
2. Study area The study area is located midway along the Pacific coast of the Baja California Peninsula at the edge of the Vizcaino Desert and near the city of Guerrero Negro (Fig. 1). The shallow Ojo de Liebre Lagoon, approximately 8 30 km, supplies seawater to the salt production facility. The climate of this area is hot and very dry. Air temperature ranges from 0 to 40 jC, and the 10-year average has been 18.5 jC. Precipitation averages less than 100 mm/year, and prevailing northwestern winds occur year-round (Phleger, 1962). The area is characterized by extensive flatlands receiving only episodic freshwater inputs during rare tropical cyclones. Conditions favor the formation of natural salt flats, and hence the location of the Exportadora de Sal salt production plant. The physical layout of the plant is shown in Fig. 1. Covering an area of 30,000 ha, 13 shallow ( < 0.5 m deep) concentration ponds are used in the first stage of the process. Seawater pumped from adjacent Ojo de Liebre Lagoon is concentrated by solar- and wind-induced evaporation until it becomes saturated with respect to halite (NaCl). The bottom of evaporation ponds 4 and 5 is covered by thick layer of microbial mats composed of colonies of cyanobacteria and diatoms (Garcia-Pichel et al., 1994; Lopez Cortes, 1998; Nu¨bel et al., 1999). The second stage consists of 32 crystallization ponds with an area of 3,000 ha, where table salt (halite, NaCl) precipitation occurs. The residual brine (bitterns, rich in Mg –SO4 –K – Cl) is removed to storage ponds with
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
135
Fig. 1. Study area and location of sampling stations of surficial sediments in the concentration ponds of Exportadora de Sal, Guerrero Negro, Baja California Sur, Mexico.
136
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
Table 1 Coordinates of the sampling and textural characteristics of sediments in the concentration ponds of the Guerrero Negro salt works Sample number
Location of sampling site
Salinity, x
General features of air-dried sediment samples
Surficial sediments from Ojo de Liebre Lagoon (n = 23) 1-A
Adjacent portion of the Ojo de Libre Lagoon
38a 3040
Fine to very fine sand.
27j39.371VN; 114j57.386VW
44:2 3042
1-B
27j36.402VN; 114j57.386VW
49:1 3042
1-C
27j35.631VN; 113j54.378VW
50:4 3042
2-A
27j35.535VN; 113j54.263VW
50:5 5261
2-B
27j34.342VN; 113j53.905VW
61:6 5261
2-C
27j34.639VN; 113j53.218VW
62 5261
3-A
27j34.766VN; 113j53.090VW
74:3 6174
3-B
27j35.013VN; 113j52.798VW
69:8 6174
3-C
27j35.385VN; 113j53.478VW
64:8 6174
4-A
27j35.890VN; 113j53.481VW
64:8 7996
4-B
27j38.671VN; 113j55.116VW
67:1 7996
4-C
27j41.628VN; 113j54.803VW
72:8 7996
5-A
27j41.863VN; 113j54.994VW
79:2 100117
5-B
27j42.961VN; 113j54.379VW
90:2 100117
5-C 6-A
27j45.047VN; 113j55.563VW 27j44.880VN; 113j55.420VW
97:2 100117 102:8 123143
6-B
27j45.699VN; 113j55.250VW
112:3 123143
Very fine silty sand, with a slight odour of H2S. Its top is covered by a greyish-brown coloured layer. Fine black sand. Mean diameter = 0.21 mm. 92.9% sand; 7.1% fines. A slight odour of H2S. Soft short, elastic black fine sand. Mean diameter = 0.21 mm. 93.3% sand; 6.7% fine silt. Soft short, elastic black very fine silty sand. The top of the sample has a greyish-brown coloured layer. Coarse sand. Mean diameter = 0.21 mm. 97.0% sand; 3.0% fines, black coloured, with hard grains inside with a size of 3 – 4 mm. (probably carbonates or gypsum concretions). Strong odour of H2S. Black silty sand, pressed into thin organic mat (carpet) about 3 mm. thick. Very coarse sand. Mean diameter = 1.30 mm. 98.0% sand; 2.0% fines. Black material pressed to thin organic mat (carpet) about 5 cm thick, with incorporated hard large particles, probably gypsum. Fine silty sand. Black material pressed into thin organic mat (carpet) with a thickness of ca. 6 cm. Upper layer (2 – 3 cm) is very elastic. Elastic, black bacterial mats. Can be easily divided into separate elastic layers with a thickness ca. 1 cm each. Upper layer is light-brown with an admixture of green colour. Grey – black elastic coarse sand. Mean diameter = 0.63 mm. 96.4% sand; 3.6% fines silt with an admixtute of hard grains (gypsum?) up to 6 mm. in size. Odour of H2S is present. Bacterial mats, but can be easily ground. Coarse sand after grinding. Mean diameter = 0.7 mm. 90.9% sand; 9.1% fines Top layer is thin grey-light brown. Odour of H2S is present. Grey – brown bacterial mats, harder to grind and rather easily separating into individual layers, in particular, upper layer of 2 – 3 cm thickness. After grinding: very coarse sand. Mean diameter = 1.2 mm. 98.1% sand; 1.9% fines. Elastic, evidently laminated bacterial mats looking like annual growth rings with the alternation of brown-red and green layers. Twenty layers can be clearly distinguished. The upper layer (cover) is grey – yellow – green . It is firm and flexible as a rubber sole of a boot. After grinding: silty sand. Laminated bacterial mats, but resolution is worse than for the 5-A sample. After grinding: very coarse sand. Mean diameter = 1.98 mm. 99.9% sand; 0.1% fines. Brown fragile bacterial mats with a consistence of a gelatin. Brown gelatin-like mash with a large quantities of grains of gypsum with a diameter up to 8 mm. After grinding: very coarse sand. Mean diameter = 1.01 mm. 93.9% sand; 6.1% fines. Large crushed pink grains of gypsum and an admixture of a brown material. After grinding: silty sand.
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
137
Table 1 (continued) Sample number
Location of sampling site
Salinity, x
General features of air-dried sediment samples
6-C
27j45.648VN; 113j56.717VW
122:7 123143
7-A
27j45.619VN; 113j56.709VW
119:8 124152
7-B
27j44.833VN; 113j56.965VW
124:1 124152
7-C
27j44.909VN; 113j57.289VW
129:3 124152
8-A
27j44.812VN; 113j57.555VW
122:9 132162
8-B
27j45.866VN; 113j57.051VW
121:7 132162
8-C
27j45.867VN; 113j58.946VW
122:1 132162
9-C
27j47.873VN; 113j59.134VW
147:1 153190
10-A
27j47.873VN; 113j59.134VW
146:5 170214
10-C
27j49.811VN; 113j57.319VW
170:6 170214
11-A
27j49.811VN; 113j57.319VW
170:8 223247
Large grains of gypsum with a size of 1 – 3 cm. Admixture of a brown material. After grinding: very coarse sand. Mean diameter = 1.06 mm. 95.4% sand; 4.6% fines. Small and large pieces of pink – white gypsum. After grinding: very coarse sand. Mean diameter = 1.63 mm. 98.9% sand; 1.1% fines. Upper layer is the crust of dirty grey – yellow material with a thickness of ca. 0.7 cm. Below this are pink – white large and small grains of gypsum. After grinding: very coarse sand. Mean diameter = 1.52 mm. 99.9% sand; 0.1% fines. Thick, heavy flat crusts of gypsum wth thickness 1 – 2 cm. Grey coarse sand is accumulated below them. After grinding: sandy silt. Pink gypsum grains and granular crusts, not very firm. After grinding: coarse sand. Mean diameter = 0.8 mm. 98.6% sand; 1.4% fines. Grey coarse sand without gypsum, may be from excavation of dikes or of aeolean origin. After grinding: medium sand. Mean diameter = 0.27 mm. 98.7% sand; 1.3% fines. Grey grains adhering together. Looks like brain-shaped corals. After grinding: coarse sand. Mean diameter = 0.78 mm. 97.3% sand; 2.7% fines. White, very firm crust of gypsum precipitate, sometimes it covers stones. The crust of grey – white dense, firm gypsum precipitate with pink incrustations. White, lusterless, firm crystals adhering together and stone-like crusts of gypsum precipitate. Flat, gray gypsum crusts of gypsum precipitate.
a
Salinity in the point during sediment sampling : Multiannual range of salinity in the pond ðmin and max averages for the years 1990 1992; 1998; 1999Þ
an area of 2000 ha. About 7 million metric tons of table salt is collected annually and shipped worldwide.
3. Materials and methods Twenty-eight sediment samples were taken from the concentration ponds using a Van Veen grab (Fig. 1, Table 1) and overlying waters were hand-collected at each location in high pressure, polyethylene bottles, previously rinsed with de-ionized water and then by sample water. The density of the brines was determined at the Exportadora de Sal’s analytical laboratory by standard
techniques, using 100-ml calibrated pipettes with salinity calculated on a volume basis. Sediment samples were divided for lithological and chemical analyses. Grain-size distribution of the sand fraction was determined with calibrated sieves according to Folk (1974), after washing of finer particles through a 63-Am screen. The weight of the < 63-Am material was added to the bottom pan material recovered after sand sieving to calculate the total fine fraction. The 63- to 125-Am fraction was used for characterization of the lithology. Chemical analyses were performed on whole sediment subsamples. Carbonate content was determined by titration of the excess of HCl with 0.5 M NaOH following its addition to sediment subsample
138
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
(Rauret et al., 1988). Organic carbon content was then determined in the remaining material by potassium dichromate oxidation and back-titration, after removal of chloride by washing the sample prior to analysis. The determination of Al, As, Ba, Cd, Cu, Fe, Mg, Mn, Ni, Pb, U, V, and Zn was performed at the analytical laboratory of the Skidaway Institute of Oceanography (Savannah, GA,) according to standard procedures (Alexander and Windom, 1999; Windom et al., 1989). Total digestion was carried out in Teflon beakers using 250 mg of sediment that had been dried at 80 jC for 10 h and homogenized by grinding with an agate mortar and pestle. The samples were treated with a mixture of 10 ml
concentrated HNO3 and 10 ml concentrated HF; allowed to digest overnight at ambient temperature; 3 ml of concentrated HClO4 were then added; and the sample was heated to near-dryness at 120 jC. The residue was twice treated with 1 ml concentrated HNO3 and heated to dryness to remove chloride ions. The final residue was dissolved in 20 ml of 1% HNO3. If necessary, an internal standard was added. Analyses for As, Ba, Cd, Cu, Ni, Pb, U, V, and Zn were made with an inductively coupled plasma mass spectrometer (VG Plasma Quad). Atomic absorption spectrophotometry (Perkin Elmer, Analyst 800) was used to measure Al, Fe and Mn. Reagent blanks and standard reference material
Table 2 Results from the analysis of the standard reference materials of the marine sediment MESS-2 (NRC Canada), of the estuarine sediment SRM 1646 a (NIST, USA) (as average F standard deviation) and of polluted marine sediment IAEA-356 (in the numerator: ranges; in the denominator: average) Element content
MESS-2 passport value
MESS-2 values determined by ICP-MS technique (n = 4)
ES 1646 a passport values
ES 1646 a values determined by INAA (n = 6)
Aluminum, Amol/g
3180 F 80 Certified value
3140 F 40
–
–
276 F 11 Certified value
275 F 1
2.46 Noncertified value 83 F 3 Certified value
3.3 F 1.6 108 F 33
7.33 F 0.17 2.14 F 0.09
1.53 Noncertified value –
2.04 F 0.44 –
– 246 F 14
– 787 F 37 Certified value 85 Noncertified value
– 750 F 77 122 F 76
603 F 46
–
–
806 F 70
358 Noncertified value
403 F 21
994 F 5
–
–
– 14.7 F 1.7 4.67 F 0.14
111 Noncertified value – –
2.57 F 0.09
0.75 F 0.02 Certified value
Antimony, nmol/g Arsenic, nmol/g Barium, Amol/g Cadmium, nmol/g Calcium, Amol/g Chromium, nmol/g Cobalt, nmol/g Copper, nmol/g Iron, Amol/g Lead, nmol/g Manganese, Amol/g Mercury, nmol/g Nickel, nmol/g Scandium, nmol/g Uranium, nmol/g Vanadium, Amol/g Zinc, Amol/g
2.14 F 0.09 Certified value – 234 F 24 Certified value 618 F 31 Certified value 780 F 40 Certified value 805 F 6 Certified value 6.64 F 0.38 Certified value 0.46 F 0.04 Certified value 840 F 24 Certified value – 4.95 F 0.20 Certified value 2.63 F 0.25 Certified value
IAEA-356 passport values
IAEA-356 values determined by INAA (n = 4)
20242415 2213
2320 F 100
403449 432
489 F 233
116 F 4 – –
150221 153
– –
158 F 3 – –
0.77 F 0.12
–
–
6.55 F 0.13 0.45 F 0.01 821 F 24
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
MESS-2 were incorporated into the sample run to maintain continuous calibration. Mercury was determined in the deposit samples by the isotope dilution inductively coupled plasma mass spectrometry (ICPMS) also in the Skidaway Oceanographic Institute’s analytical laboratory (Smith, 1993). Subsamples were weighed, spiked with 201Hg, and digested with 5.0 ml of a mixture of concentrated HNO3 and HCl (1:1) on a steam bath for 1 h using Teflon containers. After the samples have cooled for 30 min at ambient temperature, they were poured into 20-ml clean polyethylene liquid scintillation vials. The special sample cell of total volume of 100 ml was used for mercury reduction of sediment digests. A 0.5-ml aliquot of the digest was diluted with approximately 40 ml of water in the sample cell. The
139
sample cell was purged with argon (0.7 ml/min), from the nebulizer supply, for 15 s before the output from the cell is connected to the torch to remove atmospheric oxygen, which would quench the plasma. After the sample cell output was connected to the torch, borohydride was injected into the side arm cell via a hypodermic needle connected to teflon microbore tubing and peristaltic pimp. A suitable flow rate for the reductant was 0.5 ml/min and liberated mercury was measured by the inductively coupled plasma mass spectrometer (Fisons Model PQII+) equipped with a standard interface and oil diffusion pumps. The procedure detection limit was 10 pmol/g using 0.5 –g sample. Quantitative recovery was obtained on NRC reference standard (BEST-1), which is certified at 459 F 45 pmol/g.
Table 3 Major component and major earth element content in surficial sediments of the ponds Sample Surficial sediments of adjacent part of the Ojo de Liebre Lagoon (n = 23) 1A 1B 1C 2A 2B 2C 3A 3B 3C 4A 4B 4C 5A 5B 5C 6A 6B 6C 7A 7B 7C 8A 8B 8C 9C 10A 10C 11A
Organic carbon, %
Carbonates, Amol/g
Ca, Amol/g
Fe, Amol/g
Al, Amol/g
0.70
1490
1960
226
2560
4.33 2.43 3.48 2.16 1.49 8.93 6.17 10.12 10.42 4.68 2.46 10.54 11.81 12.58 10.89 3.77 1.99 4.43 1.40 0.79 1.41 1.29 0.72 1.90 1.24 0.15 0.26 0.09
3890 4030 7550 4840 3890 2640 3590 1940 – 2640 2600 2990 1470 – 1580 1980 2960 5210 587 2350 782 1170 1965 2520 2820 205 700 470
2500 2620 4620 3220 3290 1896 2000 1320 237 2940 1770 2120 1270 773 1073 3840 4390 4670 3870 4070 4120 4190 2400 2920 4120 3990 4540 4320
405 417 109 184 288 206 227 156 351 287 272 143 127 72 68 48 34 47 34 34 23 34 280 175 60 11 23 38
1664 1470 415 1150 1280 790 840 530 1290 1160 1800 560 367 152 222 85 7.4 74 7.4 270 196 196 1550 1020 1340 310 910 185
Mg, Amol/g
Sr, Amol/g
Ba, Amol/g
Mn, Amol/g
395
8.7
1.97
5.67
1050 1480 1010 940 1790 1660 1730 1150 1840 1310 1450 1045 1234 1053 1020 490 160 457 160 255 66 263 360 165 400 29 62 110
18.2 19.4 43.8 50.5 41.7 34.9 22.6 16.8 2.0 24.1 9.4 25.6 1.5 22.8 10.3 19.6 29.6 42.3 16.4 17.2 16.0 12.1 19.7 13.1 14.9 14.7 13.2 13.2
0.98 1.02 0.55 1.17 2.51 0.00 1.82 0.24 1.17 3.28 2.18 0.84 1.46 0.55 1.13 0.24 1.06 0.40 0.07 0.73 0.95 0.31 0.65 0.65 0.13 0.95 0.80 0.13
5.18 5.24 3.33 5.69 5.55 3.04 3.29 1.86 4.00 4.18 6.87 2.40 1.78 0.93 0.19 0.64 2.04 1.13 0.18 0.24 0.50 0.52 3.09 1.98 1.11 0.10 0.61 1.25
140
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
Cs, Ca, Sr, Cr, Sb, and Sc were measured by instrumental neutron activation analysis (INAA) at the V.I. Vernadsky Institute of Geochemistry and Analytical Chemistry (Moscow, Russia). The accuracy and precision of the INAA method were estimated on the basis of estuarine sediment SRM 1646a (NIST) standard reference material, SRMs of marine sediments SD-N-1/2, and IAEA-356 (Horvat et al., 1994), as well as Russian geological SRMs. Details of the analysis by this method, as well as detection limits for elements in sediments were presented in Shumilin et al. (2000). Satisfactory analyses were confirmed for all certified metals (Table 2).
4. Results and discussion 4.1. General description of the changes along the sequence of ponds The salt content in the water of ponds (Table 1) follows the pattern expected for progressive evaporation of seawater. During the concentration stage, salt content increases from 39x to 252x. Sediments in the concentration ponds change from black, silty fine sands with a brownish (oxidized) surface layer in the initial ponds, to bacterial mats containing grains of precipitated carbonates and gyp-
Table 4 Trace element concentration in the surface sediments from the concentration ponds (dry weight) in comparison with the sediments of the Ojo de Liebre Lagoon Sample number
Cr, Ni, V, Zn, Co, Cu, Pb, Sc, Cs, As, Sb, U, Cd, Hg, nmol/g nmol/g nmol/g nmol/g nmol/g nmol/g nmol/g nmol/g nmol/g nmol/g nmol/g nmol/g nmol/g pmol/g
Sediments of the adjacent part of the Ojo de Liebre Lagoon (n = 23) 1A 1B 1C 2A 2B 2C 3A 3B 3C 4A 4B 4C 5A 5B 5C 6A 6B 6C 7A 7B 7C 8A 8B 8C 9A 9C 10A 10C 11A
517
150
459
474
60
69
940 552 327 383 438 425 567 510 873 777 619 175 215 117 121 129 150 42 60 102 58 212 60 190 – 81 19 88 17
296 504 165 247 302 175 239 194 361 341 162 182 187 109 65 104 90 73 60 65 63 89 85 77 81 136 68 70 107
1309 1162 426 717 1078 954 901 665 1474 950 1158 563 656 347 220 132 15.7 92 26 135 77 100 499 298 544 202 11.7 67 128
1046 736 170 277 466 367 402 295 694 465 410 214 188 112 96 70 1.5 30.6 1.5 50 17 41 220 112 190 44 6 21 38
126 266 54 102 127 97 117 85 197 119 107 54 53 27 25 25 15.3 25.5 11.9 22.1 17.0 23.8 53 39 44 98 6.8 11.9 18.7
248 232 6.6 94.9 120 102 104 96 218 150 87 81 89 57 47 33 13.7 29.4 13.4 19.8 15.2 26.0 37.9 28.9 36.5 26.6 18.1 11.2 14.1
36
162
25.6
100
37 24 9.7 20.3 18.3 12.5 13.0 10.6 23.2 17.9 27.0 11.6 12.1 5.8 5.3 3.4 1.4 3.4 1.0 4.3 3.4 3.8 7.7 8.2 7.2 6.3 0.5 1.0 2.9
189 176 49 98 138 85 113 89 160 145 169 65 49 24 22 15.6 6.7 13.3 4.4 26.7 20.0 17.8 96 44 – 22 4.4 11.1 28.9
32.0 5.3 – 9.8 18.1 13.5 31.6 28.5 18.1 – 26.3 17.3 15.0 – 13.5 5.3 5.3 1.5 3.7 1.5 3.0 – 9.8 – – – 3.8 0.8 0.8
109 80 44 5.3 53 85 69 67 95 49 56 61 52 103 229 43 17 43 23 17 9.3 23 19 19 13 37 2.7 5.3 8.0
5.4
12.2
1.2
50
6.2 17.1 4.4 3.4 2.2 6.7 4.5 4.2 14.9 14.2 7.0 5.9 5.2 3.8 6.7 8.7 3.5 2.8 – 1.4 – 0.9 3.6 1.6 – 0.9 1.1 2.0 9.2
18.4 23.1 15.5 22.6 43.6 21.4 24.8 21.0 24.8 25.2 18.9 21.0 18.5 12.1 7.1 15.5 3.4 9.7 1.7 3.8 1.7 3.4 5.9 5.9 4.6 3.4 0.4 0.4 0.8
6.8 2.5 1.7 1.6 1.2 2.2 1.8 2.1 3.8 1.3 1.5 1.9 2.7 1.4 1.1 0.6 0.1 0.2 0.1 0.2 0.1 0.4 0.7 0.7 0.4 0.5 0.1 0.7 0.1
384 100 140 95 70 234 199 105 155 90 90 95 125 110 115 70 45 80 40 20 35 35 25 14 10 10 10 10 10
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
141
Fig. 2. Major components, major earth and minor element concentrations in surficial sediments of the concentration ponds versus station number: (a) carbonates and calcium; (b) organic carbon and magnesium; (c) aluminum, iron, and manganese; (d) barium and strontium; (e) nickel and vanadium; (f) chromium and zinc.
Fig. 3. Trace element concentrations in the surficial sediments of the concentration ponds versus station number: (a) scandium and uranium; (b) cobalt, copper, and lead; (c) arsenic and caesium; (d) antimony, cadmium, and mercury.
142
Table 5 Correlation matrix for major components, major and trace elements in the surficial sediments of the concentration ponds
1.00
0.46 1.00
Corg
Cs
Ca
Sr
Sc
Cr
Fe
Ni
Zn
Sb
U
Al
As
Ba
Cd
Co
Cu
Pb
Mn
Hg
V
Mg
0.65 0.04 1.00
0.68 0.48 0.25 1.00
0.64 0.75 0.01 0.73 1.00
0.30 0.16 0.73 0.10 0.31 1.00
0.76 0.01 0.50 0.67 0.48 0.09 1.00
0.80 0.17 0.48 0.80 0.49 0.07 0.89 1.00
0.75 0.03 0.51 0.59 0.48 0.11 0.97 0.82 1.00
0.75 0.08 0.53 0.39 0.33 0.19 0.81 0.73 0.83 1.00
0.43 0.06 0.40 0.30 0.03 0.17 0.41 0.52 0.44 0.42 1.00
0.33 0.13 0.10 0.02 0.27 0.22 0.41 0.30 0.46 0.60 0.20 1.00
0.23 0.22 0.35 0.13 0.17 0.10 0.20 0.19 0.21 0.14 0.19 0.27 1.00
0.50 0.22 0.32 0.49 0.34 0.06 0.88 0.69 0.87 0.59 0.26 0.22 0.13 1.00
0.43 0.63 0.10 0.44 0.65 0.20 0.23 0.33 0.26 0.18 0.10 0.34 0.22 0.05 1.00
0.69 0.03 0.48 0.62 0.39 0.14 0.87 0.81 0.75 0.60 0.30 0.26 0.18 0.79 0.21 1.00
0.31 0.25 0.41 0.19 0.21 0.14 0.27 0.25 0.27 0.21 0.20 0.28 0.99 0.17 0.24 0.25 1.00
0.75 0.05 0.55 0.42 0.39 0.17 0.87 0.75 0.88 0.96 0.39 0.63 0.17 0.69 0.23 0.66 0.24 1.00
0.80 0.20 0.52 0.58 0.45 0.07 0.87 0.87 0.89 0.90 0.47 0.53 0.14 0.66 0.38 0.67 0.21 0.88 1.00
0.25 0.25 0.37 0.14 0.18 0.14 0.19 0.18 0.20 0.15 0.17 0.25 1.00 0.10 0.22 0.17 1.00 0.17 0.13 1.00
0.72 0.13 0.60 0.52 0.31 0.33 0.90 0.77 0.85 0.72 0.46 0.29 0.21 0.85 0.03 0.91 0.28 0.77 0.70 0.21 1.00
0.76 0.46 0.67 0.45 0.50 0.30 0.43 0.43 0.46 0.34 0.36 0.15 0.49 0.20 0.41 0.36 0.54 0.41 0.41 0.51 0.41 1.00
0.84 0.21 0.49 0.73 0.60 0.11 0.96 0.90 0.93 0.81 0.37 0.38 0.21 0.80 0.30 0.85 0.29 0.85 0.88 0.21 0.87 0.48 1.00
0.85 0.50 0.47 0.71 0.71 0.16 0.72 0.72 0.69 0.69 0.20 0.33 0.24 0.49 0.49 0.67 0.33 0.73 0.68 0.28 0.65 0.59 0.86 1.00
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
Parameter Salinity Ccarb Salinity Ccarb Corg Cs Ca Sr Sc Cr Fe Ni Zn Sb U Al As Ba Cd Co Cu Pb Mn Hg V Mg
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
sum in subsequent ponds, to ponds with crusts of gypsum and anhydrite (Table 1). Shell fragments were found in pond 1, traces in pond 2, but none in subsequent ponds as a consequence of the highly inhospitable conditions in hypersaline environments: high concentrations of salts, high osmotic pressure, extreme temperatures, and frequently low oxygen content (Larsen, 1980; Oren, 1999; Shilo, 1980). Terrigenous detrital sand is in low abundance in all ponds, but its mineral composition is very uniform. Quartz and feldspars dominate the light fraction, while hornblende prevails in the heavy mineral fraction, indicating, that the terrigenous material is supplied predominantly from granitic – metamorphic rocks, which are wide-spread on the surrounding land (Nikolayeva and Derkachev, pers. comm.). On many of the terrigenous mineral grains, thin crusts of authigenic fine crystalline carbonate or individual wedge-shaped
143
carbonate crystals occur. Carbonaceous lumpy precipitates are also found, consisting of aggregates of small-sized carbonaceous grains, grouped around different minerals (Nikolayeva and Derkachev, pers. comm.). The sequence of authigenic mineral precipitation coincides in general with that expected during gradual evaporation of seawater (Borchert, 1965; Garrett, 1980; Horne, 1969). Carbonate aggregates and carbonate silt were abundant in ponds 1 to 4. This is similar to Dead Sea mud samples collected from water depths of 8 –250 m, which contained more than 3.6% authigenic aragonite (Herut et al., 1997). Anhydrite and gypsum start precipitating in ponds 2 – 4, followed by massive crystallization in ponds 6 – 13. The unusual feature of the concentration ponds of the Guerrero Negro salt works is the presence of large fields of microbial mats found in ponds 3 –6 (Table 1), filling
Fig. 4. Aluminum-normalized major earth and minor element concentrations in surficial sediments of the concentration ponds versus station number: (a) inorganic carbon and organic carbon; (b) iron and manganese; (c) calcium and magnesium; (d) barium and strontium; (e) chromium and nickel; (f) vanadium and zinc.
144
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
with microbial organisms of pore space between grains of terrigenous and marine biogenic sediments deposited on the floor of ponds 1 and 2, and gypsum – anhydrite crusts of ponds 7– 12. 4.2. Distribution of major and trace element concentration in the surface deposits The contents of major components, major and trace elements in the surface sediments of the concentration ponds are shown in Tables 2– 4. 4.2.1. Carbonate and Ca Previous description of mineralogical composition of sediments in the logical sequence of increasing salinity corresponds to the results of chemical analysis of Ca and carbonates in sediments (Fig. 2a). The
comparison of the curves of carbonate and Ca content in sediments plotted versus station number allows a distinction between the environments of bottom sediments in the sequence of evaporation ponds. First maximum of carbonates and Ca contents in sediments at the station 1C on longitudinal transect in first two ponds reflects the deposition of biogenic carbonates supplied from Ojo de Liebre Lagoon and/or produced inside of these, highly productive ponds, partially due the almost full absence of predators. We cannot forget also a possibility of the presence of calcium in first ponds in forms other than biogenic or chemically precipitated carbonates. They can be calcium silicates or calcium phosphates incorporated into large clay particles. High content of calcium in sediments and crusts in ponds 6– 11 are definitely due to precipitation of gypsum and anhydrite.
Fig. 5. Aluminum-normalized trace element concentrations in surficial sediments of the concentration ponds versus station number: (a) cobalt, and copper; (b) lead, and arsenic; (c) scandium, and uranium; (d) antimony, and caesium; (e) cadmium and mercury.
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
4.2.2. Organic carbon, Al, Fe and Mn Organic carbon (Corg) content of the sediments (Fig. 2b) in ponds 1 – 3 (1.5 – 8.9%) is much higher than the average for sediments of Ojo de Liebre Lagoon (mean of 0.70% in lagoon sediments of the area adjacent to the pumping station). Corg content increases at the last station 2C of the pond 2, is high in ponds 3 –5 reaching a maximum (about 12.6%) at station 5B in a fifth pond, sharply decreasing in deposits of stations 6A, 6B and 6C and gradually declining in the deposits of ponds 7– 9 until its complete disappearance in the deposits of ponds 10– 11. Relatively high contents of Corg in first two ponds presumably correspond to the gravitational settling of organic-rich biogenic particles supplied from Ojo de Liebre Lagoon or generated by photosynthesis developed in the first pond. Some portion of quite lowsized colloidal organic matter was probably scavenged by authigenic carbonates precipitating in ponds
145
2– 4, but more probably this increase in Corg content in deposits is related to the formation of bacterial mats that completely cover the floor of ponds 3 –6, where other quite high values of Corg content (up to 12.6%) were found. The thickness of bacterial mats is variable (about 2 cm at station 3A, about 5 cm at stations 4C and 5A, thin to 2 –3 cm at station 6C). The mats are inhibited by the massive gypsum precipitation in ponds 7 – 13 (Table 1). Bacterial mats are special biogeochemical environments and are very important for nutrient migration and for elements that can alter their speciation in brines and interstitial waters because of the change in the oxidation state of the elements present along the changing redox potential within the mats (Stiller and Nissenbaum, 1999). Concentration of major earth elements (Al, Fe, and Mn) was highest in surficial sediments of the first two ponds (Fig. 2c), resulting from the deposition of suspended particulate matter that enters the first pond
Fig. 6. Calcium-normalized major earth and minor element concentrations in surficial sediments of the concentration ponds versus station number: (a) inorganic carbon and organic carbon; (b) iron and manganese; (c) magnesium; (d) barium and strontium; (e) chromium and vanadium; (f) nickel and zinc.
146
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
with the sea water pumped from Ojo de Liebre Lagoon, and is gradually deposited on the floor along pond 1 by simple gravitational settling of coarse, dense particles, possibly enhanced by high salinityinduced coagulation of colloidal material (e.g. Fe hydroxides), as was reported for salt works in Greece (Varnavas and Lekkas, 1996). The maximum in pond 2 is presumably related to the scavenging of colloidal particles rich in Al and Fe, as well as some Mn coprecipitation with authigenic CaCO 3 cannot be excluded (Herut et al., 1997). A clear peak of Al, Mn, and Fe in the sediments at stations 3C, 4A, 4B, and 4C can be explained by the use of fine aluminosilicates and Fe and Mn oxyhydroxides as a substrate in the formation of the bacterial mat community. This was followed by minimal amounts in the gypsum encrusted floors of ponds 6 –7 and an unanticipated increase in ponds 8 – 10 (Fig. 2c). The latter could be related to the erosion of the earthen dikes separating
the ponds or to input of airborne particles from the adjacent dune fields, as suggested by the presence of quartz and feldspar (Nikolayeva and Derkachev, pers. comm.) in the material on the floor of the last concentration ponds 8– 10, where gypsum and anhydrite precipitates completely dominate. 4.2.3. Sr and Ba Fig. 2d shows the strontium and barium concentrations in the pond sediments versus salinity of the brines. These distributions are probably reflects formation of their carbonates or co-precipitation of strontium carbonates with CaCO3 or dolomite (Fig. 2c). 4.2.4. Trace metals (As, Cr, Ni, Co, Cu, U, V, Zn, Cd, Pb, Sb, Hg, and Cs) The graphs of trace metals concentrations in surficial deposits versus station number reveal distribution patterns quite similar to those found for Fe, Al, and
Fig. 7. Calcium-normalized trace element concentrations in surficial sediments of the concentration ponds versus station number: (a) cobalt, and copper; (b) arsenic, and lead; (c) scandum and uranium; (d) antimony and caesium; (e) cadmium and mercury.
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
Mn (Figs. 2e,f and 3a– d). When compared to marine sediments of Ojo de Liebre Lagoon, the sediments in ponds 1 and 2 have higher Cu, Ni, Zn, and other trace metal concentrations. This may be a result of several processes, such as grain-size sorting of particles during gravitational deposition of fine sediments, co-precipitation with freshly formed colloidal Fe hydroxides (Varnavas and Lekkas, 1996), calcium carbonates, or co-precipitation of U and Cd with calcium phosphates (Piper, 1991, 1994; Zanin et al., 2000). Concentrations of these elements, showing high values in the pond 1, display a gradual decrease along the sequence of evaporation ponds. Some of them (Co, V, Zn, Cd, Pb, and Cs) show a slight increase in the zone of the second maxima for Al, Fe, and Mn (ponds 3 and 4), and a third maxima (pond 8) presumably associated with aeolian inputs (Fig. 3b). Cr, Ni, Cu, U, and Hg, however, did not show similar enrichment in the final concentration ponds (Fig. 3a –d).
147
Unlike Dead Sea shallow surface sediments reported by Herut et al. (1997), no enrichment of Cd was found in these deposits (Fig. 3d). In seawater at the entrance to the system, the concentration of dissolved Cd is about 107 pmol/l, but interestingly, this element disappeared completely in the brines of the ponds (Shumilin et al., in press). This could be related to co-precipitation of Cd isomorphically with calcium phosphate during early stages of evaporation, as reported for Cd-rich phosphate detritus in surface sediments of the Dead Sea (Herut et al., 1997). Such co-precipitation of Cd with calcium phosphate may have produced elevated Cd contents in the phosphatic rocks of the southeastern part of Baja California Peninsula (Piper, 1991, 1994). An interesting feature of the Guerrero Negro saltworks is the accumulation of As in the microbial mat deposits of ponds 5 at the station 5C (Fig. 3c). Bioaccumulation within the mat substrate or redox
Fig. 8. Organic carbon-normalized major earth and minor element concentrations in surficial sediments of the concentration ponds versus station number: (a) inorganic carbon; (b) aluminum, iron, and manganese; (c) calcium and magnesium; (d) barium and strontium; (e) nickel, and vanadium; (f) chromium, and zinc.
148
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
influence on As behavior in the brine/sediment pore water system could be the cause of this phenomenon that deserve further investigation. Sb behavior is completely different from the rest of the elements studied in the pond deposits (Fig. 3d). Besides typical for many of other trace metals enrichment in sediments of ponds 1, 3 and 6, this element’s special feature is its increase in the last concentration pond, presumably caused by co-precipitation of the anion Sb(OH)6 with gypsum or anhydrite. 4.3. Normalization of element contents in sediments Normalization is very helpful in minimizing the influence of the texture and mineralogy on the trace element variations in sediments (Loring, 1991; Luoma, 1990; Windom et al., 1989). Typical reference elements for such normalization for marine sediments
are Al, Sc, Li, Cs, Rb, and sometimes, Fe and Corg. Our data displayed high correlation between contents of Al, Sc, Cs, and Fe in the pond sediments (Table 5). Al was chosen as a major silicate element to normalize trace element contents in the surficial deposits, Ca to reveal the possible influence of precipitation of Ca CO3 and CaSO4 2H2O on trace elements, while Corg was used to evaluate the impact of the microbial mats. 4.3.1. Al normalization The element/aluminum ratios versus station number generally are characterized by a maximum at the ponds 5– 7 the 80x to 120x salinity range (the case for As, Ba, carbonates, Cs, Cu, Mg, organic carbon, Ca, Sr, Co, Cr, Fe, Mn, Hg, U, Sb and V), or by a gradually ascending curve followed by a maximum (the case for Cd, Mg and V) (Figs. 4 and 5). This is probably due to strong impoverishment in Al
Fig. 9. Organic carbon-normalized trace element concentrations in surficial sediments of the concentration ponds versus station number: (a) cobalt, and copper; (b) arsenic, and lead; (c) scandium, and uranium; (d) antimony, and caesium; (e) cadmium and mercury.
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153 Table 6 Varimax factor score matrix resulting from the Q-mode factor analysis of the data set on the composition of surficial sediments of the concentration ponds of Guerrero Negro saltworks Parameter
Factor 1
Factor 2
Factor 3
Factor 4
Salinity C-carbonates Organic carbon Al As Ba Ca Cd Co Cr Cs Cu Fe Hg Mg Mn Ni Pb Sb Sc Sr U V Zn Explained variance Total proportion
0.90 0.26 0.62 0.74 0.40 0.83 0.59 0.42 0.88 0.90 0.71 0.90 0.92 0.62 0.85 0.86 0.84 0.35 0.46 0.94 0.18 0.35 0.96 0.47 12.02 0.50
0.10 0.65 0.09 0.39 0.46 0.23 0.33 0.73 0.22 0.15 0.08 0.18 0.22 0.50 0.15 0.29 0.23 0.76 0.07 0.26 0.05 0.75 0.15 0.07 3.31 0.14
0.09 0.59 0.54 0.09 0.52 0.05 0.66 0.43 0.06 0.09 0.41 0.09 0.03 0.11 0.24 0.26 0.07 0.44 0.07 0.01 0.65 0.44 0.09 0.26 2.75 0.11
0.31 0.25 0.44 0.21 0.01 0.01 0.03 0.26 0.16 0.03 0.22 0.09 0.14 0.37 0.17 0.01 0.13 0.26 0.61 0.12 0.61 0.31 0.02 0.06 1.68 0.07
of sediments constituted of bacterial mats. This differs from the conservative behavior shown for Fe with respect to Al, observed in the transition from stream
149
sediments in the drainage basin of the Jordan River to the sediments of the hypersaline Dead Sea (Herut et al., 1997). 4.3.2. Ca normalization Because Ca is a major agent of chemical transformation in concentration ponds from precipitation of carbonates and gypsum/anhydrite, the graphs of [element]/[Ca] ratios versus salinity display two principal modes of distribution, with one or two maxima on the curve (Figs. 6 and 7).Two peaks of carbonates (Fig. 6a) corresponds to the participation of other component, probably Mg, in the formation of the carbonate phase in the sediments. First peak on the curve for a station 3C can be seen for the plots for many other Ca-normalized elements, such as Fe, Mn, Cr, V, Ni, Zn, Co, Cu, Pb, As, Sc, U, Sb, Cs and Cd. These elements are probably co-precipitated or absorbed (scavenged) by a newly formed solid phase of carbonates. Carbonates and some elements (As, Cd, Mg and Sr) also show second maximum for station 5B, probably reflecting the appearing the carbonate skeletons or carbonates substrate of the microbial mats or their incorporation into gypsum matrix during the initial stage of the precipitation of CaSO4 2H2O. 4.3.3. Corg normalization Strong enrichment in carbonates is clearly seen in sediments of ponds 1– 2 and ponds 8– 11 causing the increase of the [element/organic carbon ratios] for Mg, Al, Fe, Mn,Ni, V, Zn, Cr, Co, Cu, Pb, As, Sc, Sb and
Fig. 10. Cluster analysis of the set of data on element concentration in sediments and salinity of brines of the Guerrero Negro salt works.
150
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
Cd (Figs. 8 and 9). This is probably a result of the association of the majority of elements with organic matter in sediments in first ponds and an apparent effect of the supply of inblown material into last ponds, characterized by almost complete absence of organic matter in gypsum/anhydrite matrix of the deposits. 4.4. Statistical analysis A correlation matrix for the set of elements studied is presented in Table 5. High positive correlation coefficients of Fe with Al, Mn, Co, Ni, Cu, V, and Cs are typical for coastal marine sediments with high terrigenous contributions, but they are also high for Fe –Ba and Fe –Mg pairs, which is probably a result of the incorporation of Fe-rich fine terrigenous particles inside freshly precipitated minerals, jointly with a barium carbonate/sulfate and magnesium carbonate. Principal component analysis of these data allows for distinguishing five factors which explain 87% of the variance of the parameters of the surficial sediments and salinity of the brines (characterized in Table 6).
Factor 1 (50% of the variance) displays large (more than 0.6) positive loadings for Corg, Cs, Sc, Cr, Fe, Ni, Al, Ba, Co, Cu, Mn, V, and Mg. This factor has large negative loadings for salinity and Hg. We believe this represents the input of suspended particulate matter from the lagoon to the pond system and their dilution by inert (unenriched in trace metals) authigenic calcium minerals, such as aragonite, gypsum, and anhydrite. Mercury is probably solubilized from sediments to form negatively charged chloride complexes along with the increase in chloride concentration in pond waters by evaporation (Shumilin et al., in press). Factor 2 (14% of the variance) displays high positive loadings for Cd, Pb, and U and relatively high negative loadings for carbonates. This factor probably corresponds to the co-precipitation of Cd and U with calcium phosphate found for other environments (Herut et al., 1997; Piper, 1991, 1994; Shumilin et al., 2001). Factor 3 (11% of the variance) has relatively high positive loading for carbonates (0.59) and As (0.52) and relatively high negative loading for Ca and Sr. It
Fig. 11. General scheme of the geochemical processes controlling trace element behavior in the concentration ponds of the Guerrero Negro salt works.
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
can be related to the bioaccumulation or redox precipitation of As in the microbial mats. Factor 4 (7% of the variance) displays relatively high positive loading for Sr and relatively high negative loading for Sb. Results of the cluster analysis are presented in Fig. 10. Two principal suites of major components and elements can be seen on the tree diagram. Suite I is more extensive and includes many of the analyzed components and elements (Sr, Sb, Mg, Al, Mn, Ba, Cu, Ni, Cr, Ni, V, Fe, Sc, Cs, Zn, Corg, As, and carbonates). Several subgroups of elements can also be seen in this figure. Suite I and its subgroups reflect the specific associations of elements formed during the simultaneous deposition of terrigenous and biogenic suspended particulate matter supplied with seawater from Ojo de Liebre Lagoon and precipitation of major minerals in the concentration ponds. Suite II includes Ca, Cd, Pb, U, Hg, and salinity. A feature of interest is the uncommon behavior of Hg, parallel with salinity and not with Corg, as found in sediments of Ojo de Liebre Lagoon and recorded in other coastal marine embayments. 4.5. General scheme of geochemical processes in the concentration ponds Considering the lithological and geochemical information obtained in this study and the results of trace element behavior in overlying brines (Shumilin et al., in press), we propose a schematic representation of the processes controlling mineral precipitation and major earth, and minor and trace element concentrations in surficial sediments in the concentration ponds (Fig. 11).
151
two concentration ponds. Freshly deposited fine sediments in these ponds are enriched in organic matter, biogenic carbonates, Al, Fe, Mn, and most of the trace elements, if to compare their concentrations with those of surficial sediments of the adjacent lagoon. Because of gradual evaporation, intensive evaporation-induced precipitation of authigenic calcium and magnesium carbonates occurs in ponds 3 and 4 causing a scavenging of Al, Fe, Mn carried by colloidal particles and new enrichment of sediments by trace metals. The sediment in ponds 4 and 5 includes a thick microbial mat layer, which is enriched in organic carbon, where the accumulation of As and Cd is observed as well as additional incorporation of the carbonates into as a substrate of microbial mats. In ponds 7 through 11, the deposits generally are low in organic matter and trace elements (except for Sb) and are formed by intensive gypsum and anhydrite precipitation, with an admixture of terrigeneous material in ponds 8– 10 supplied from a pond dyke erosion or carried in with wind from adjacent sand dunes. The sediments in the concentration ponds are, in general, impoverished with respect to Cs, Ba, Sc, Cr, Fe, Mn, and Zn compared to sediments of the adjacent Ojo de Liebre Lagoon. This depletion is clearly seen for most elements (Al, Fe, Mg, Mn, Cr, Ni, Co, Cd, Pb, Zn, and Hg) as the pond number and salinity increases. This is interpreted as a result of dilution of terrigenous and marine biogenic sedimentary material initially supplied to the pond system from the adjacent lagoon by authigenic minerals (calcium carbonate, gypsum, and anhydrite) formed during the evaporation process.
Acknowledgements 5. Summary Lithological data and changes in elemental composition in sediments along with a sequence of the evaporation ponds and the increase of the to salinity of the overlying brines suggest that suspended particulate matter supplied to the system, mainly by seawater pumped in from the adjacent Ojo de Liebre Lagoon, and to a smaller extent, carried into the ponds by wind from adjacent dunes and by wave-induced erosion of retaining dikes, settles mostly in the first
We are grateful to Herbert Windom and Ralph Smith of the Skidaway Institute of Oceanography, University of Georgia, for the assistance with the analyses of elements. We thank the personnel of the Research and Development Department of Exportadora de Sal, especially Juan Antonio Flores, Felipa Quin˜ones Marquez, Fernando Heredia Uribe, and Sonia Gutie´rrez for their help in the planning of this work, implementing the sampling, and suggestions. Thanks to Natalya Nikolayeva and Aleksander Derkachev of
152
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153
the Laboratory of Lithology, Pacific Oceanological Institute of Russian Academy of Sciences for general lithological descriptions of sediment samples. Funding for this project was provided by Exportadora de Sal.
References Alexander, C.R., Windom, H.L., 1999. Quantification of natural backgrounds and anthropogenic contaminants in a pristine arctic environment: the Anadyr River Basin, Chukotka Peninsula, Russia. Mar. Pollut. Bull. 38, 276 – 284. Anschutz, P., Blanc, G., 1996. Heat and salt fluxes in the Atlantis II Deep (Red Sea). Earth Planet. Sci. Lett. 142, 147 – 159. Borchert, H., 1965. Principles of oceanic salt deposition and metamorphism. In: Riley, J.P., Skirrow, G. (Eds.), Chemical Oceanography, vol. 2. Academic Press, New York, pp. 205 – 276. Degens, E.T., Ross, D.A. (Eds.), 1969. Hot Brines and Recent Heavy Metal Deposits in the Red Sea. Springer Verlag, Berlin, 600 pp. Eugster, H.P., Harvie, C.E., Weare, J.H., 1980. Mineral equilibria in a six-component seawater system, Na – K – Mg – Ca – SO4 – Cl – H2O, at 25 jC. Geochim. Cosmochim. Acta 44, 1335 – 1347. Garcia-Pichel, F., Mechling, E.M., Castenholz, R.W., 1994. Dial migration of microorganisms within a benthic, hypersaline mat community. Appl. Environ. Microbiol. 60, 1500 – 1511. Garrett, D.E., 1980. By-products recovery from solar salt operations. In: Coogan, A.H., Hauber, L. (Eds.), Fifth Symposium on Salt, vol. 2. The Northern Ohio Geological Society, Department of Geology Case Western Reserve University, Cleveland, OH, pp. 281 – 293. Folk, R.L., 1974. Petrology of Sedimentary Rocks Hemphill Publishing, Houston, TX, 182 pp. Hartmann, M., Scholten, J.C., Stoffers, P., Wehner, F., 1998. Hydrographic structure of brine-filled deeps in the Red Sea—new results from the Shaban, Kebrit, Atlantis II, and Discovery Deep. Mar. Geol. 144, 311 – 330. Herut, B., Gavrieli, I., Halicz, L., 1997. Sources and distribution of trace and minor elements in the western Dead Sea surface sediments. Appl. Geochem. 12, 497 – 505. Herut, B., Gavrieli, I., Halicz, L., 1998. Coprecipitation of trace and minor elements in modern authigenic halites from the hypersaline Dead Sea brine. Geochim. Cosmochim. Acta 62, 1587 – 1598. Horne, R.A., 1969. The Structure of Water and the Chemistry of the Hydrosphere. Wiley Interscience, New York, 398 pp. Horvat, M., Mee, L.D., Oregioni, B., 1994. World-wide and regional intercomparison for the determination of trace elements in polluted marine sediment IAEA-356, International Atomic Energy Agency. Marine Environment Laboratory-MESL Report No. 56, September 1994, p. 10. Larsen, H., 1980. Ecology of hypersaline environments. In: Nissenbaum, A. (Ed.), Hypersaline Brines and Evaporitic Environments. Elsevier, Amsterdam, pp. 28 – 39. Lopez Cortes, A., 1998. Ecologı´a y biotecnologı´a de las comunidades microbianas. Cienc. Desarr. 24, 11 – 17 (in Spanish).
Loring, D.H., 1991. Normalization of heavy-metal data from estuarine and coastal sediments. ICES J. Mar. Sci. 48, 101 – 115. Luoma, S.N., 1990. Processes affecting metal concentrations in estuarine and coastal marine sediments. In: Furness, R.W., Rainbow, P.S. (Eds.), Heavy Metals in the Marine Environment. CRC Press, Boca Raton, FL, pp. 51 – 66. McCaffrey, M.A., Lazar, B., Holland, H.D., 1987. The evaporation path of the seawater and the co-precipitation of Br and K + with halite. J. Sediment. Petrol. 57, 928 – 937. Mihelcˇic´, G., Sˇurija, B., Juracˇic´, M., Barisˇic´, D., Branica, M., 1996. History of the accumulation of trace elements in sediments of the saline Rogoznica Lake (Croatia). Sci. Total Environ. 182, 105 – 115. Millero, F.R., 1996. Chemical Oceanography, 2nd ed. CRC Press, Boca Raton, FL, 469 pp. Nissenbaum, A., 1974. Trace elements in Dead Sea sediments, Israel. J. Earth Sci. 23, 111 – 116. Nissenbaum, A. (Ed.), 1980. Hypersaline Brines and Evaporitic Environments. Elsevier, Amsterdam, 278 pp. Nu¨bel, U., Garcia-Pichel, F., Ku¨hl, M., Muyzer, G., 1999. Spatial scale and the diversity of benthic cyanobacteria and diatoms in a salina. Hydrobiologia 401, 199 – 206. Oren, A., 1999. Microbiology and Biogeochemistry of Hypersaline Environments. CRC Press, Boca Raton, FL, 359 pp. Phleger, F.B., 1962. Sedimentology of Guerrero Negro Lagoon, Baja California, Mexico. Contribution No. 48, Marine Foraminifera Laboratory. Contribution from the Scripps Institute of Oceanography, San Diego, CA, 205 – 237. Piper, D.Z., 1991. Geochemistry of a tertiary sedimentary phosphate deposit: Baja California Sur, Me´xico. Chem. Geol. 92, 283 – 316. Piper, D.Z., 1994. Seawater as the source of minor elements in black shales, phosphorites and the other sedimentary rocks. Chem. Geol. 114, 95 – 114. Rauret, G., Rubio, R., Lo´pez-Sa´nchez, J.F., Casassas, E., 1988. Determination and speciation of copper of a mediterranean river (River Tenes, Catalonia, Spain). Water Res. 22, 449 – 455. Schoell, M., Faber, E., 1978. New isotopic evidence for the origin of Red Sea brines. Nature 275, 436 – 438. Shilo, M., 1980. Strategies of adaptation to extreme conditions in aquatic microorganisms. Naturwissenschaften 67, 384 – 389. Shumilin, E., Kalmykov, St., Sapozhnikov, D., Nava-Sa´nchez, E., Gorsline, D., Godı´nez-Orta, L., Sapozhnikov, Yu., Holguin-Quin˜onez, O., Rodrı´guez-Castan˜eda, A., 2000. Major and trace element accumulation in coastal sediments along the southeastern Baja California studied by instrument neutron activation analysis and Pb-210 dating. J. Radioanal. Nucl. Chem. 246, 533 – 541. Shumilin, E., Pa´ez-Osuna, F., Green-Ruiz, C., Sapozhnikov, D., Rodrı´guez-Meza, G., Godı´nez-Orta, L., 2001. Arsenic, antimony, selenium and other trace elements in sediments of the La Paz Lagoon, Peninsula of Baja California, Mexico. Mar. Pollut. Bull. 42, 174 – 178. Shumilin, E.N., Mironenko M.V., Ryzhenko, B.N., Grajeda-Mun˜oz, M. Speciation of trace elements in the brines of the evaporation ponds of the Exportadora de Sal table salt producing plant, Guerrero Negro, Baja California Sur, Mexico. Geochem. Int., in press.
E. Shumilin et al. / Marine Chemistry 79 (2002) 133–153 Smith, R.G., 1993. Determination of mercury in environmental samples by isotope dilution/ICPMS. Analyt. Chem. 65, 2485 – 2488. Stiller, M., Nissenbaum, A., 1999. Geochemical investigation of phosphorus and nitrogen in the hypersaline Dead Sea. Geochim. Cosmochim. Acta 63, 3467 – 3475. Varnavas, S.P., Lekkas, T., 1996. Factors controlling the geochemical and crystallization processes in marine brine ponds. Mar. Georesour. Geotechnol. 14, 111 – 141. Windom, H.L., Schropp, S.J., Calder, F.D., Ryan, J.D., Smith Jr., R.G., Burney, F.G., Lewis, L.C., Rawlinson, C.H., 1989. Natural trace metal concentrations in estuarine and coastal marine
153
sediments of the southeastern United States. Environ. Sci. Technol. 23, 314 – 320. Zanin, Yu.A., Zamirailova, A.G., Gilinskaya, L.G., Fomin, A.N., Kireev, A.D., 2000. Uranium of sedimentary apatite in catagenesis. Geochem. Int. 38, 452 – 458. Zdanovskii, A.B., 1979. Calculation of the solubility products of the salts in the Mg2 + , Na + , Cl , SO42 , H2O system at 25 jC. Russ. J. Inorg. Chem. 24, 275 – 278. Zherebtsova, I.K., Volkova, N.N., 1966. Experimental study of behavoir of trace elements in the process of natural solar evaporation of Black Sea water and Sasyk – Sivash brine. Geochem. Int. 3, 656 – 670.
