Environ Monit Assess (2015) 187:4198 DOI 10.1007/s10661-014-4198-3
Background concentrations and reference values for heavy metals in soils of Cuba Mirelys Rodríguez Alfaro & Alfredo Montero & Olegario Muñiz Ugarte & Clístenes Williams Araújo do Nascimento & Adriana Maria de Aguiar Accioly & Caroline Miranda Biondi & Ygor Jacques Agra Bezerra da Silva
Received: 13 March 2014 / Accepted: 1 December 2014 # Springer International Publishing Switzerland 2014
Abstract The potential threat of heavy metals to human health has led to many studies on permissible levels of these elements in soils. The objective of this study was to establish quality reference values (QRVs) for Cd, Pb, Zn, Cu, Ni, Cr, Fe, Mn, As, Hg, V, Ba, Sb, Ag, Co, and Mo in soils of Cuba. Geochemical associations between trace elements and Fe were also studied, aiming to provide an index for establishing background concentrations of metals in soils. Surface samples of 33 soil profiles from areas of native forest or minimal anthropic influence were collected. Samples were digested (USEPA method 3051A), and the metals were determined by ICP-OES. The natural concentrations of metals in soils of Cuba followed the order Fe>Mn> Ni>Cr>Ba>V>Zn>Cu>Pb>Co>As>Sb>Ag>Cd> Mo>Hg. The QRVs found for Cuban soils were as follows (mg kg−1): Ag (1), Ba (111), Cd (0.6), Co (25), Cr (153), Cu (83), Fe (54,055), Mn (1947), Ni (170), Pb (50), Sb (6), V (137), Zn (86), Mo (0.1), As (19), and Hg (0.1). The average natural levels of heavy
metals are above the global average, especially for Ni and Cr. The chemical fractionation of soil samples presenting anomalous concentrations of metals showed that Cu, Ni, Cr, Sb, and As have low bioavailability. This suggests that the risk of contamination of agricultural products via plant uptake is low. However, the final decision on the establishment of soil QRVs in Cuba depends on political, economic, and social issues and in-depth risk analyses considering all routes of exposure to these elements.
M. R. Alfaro : O. M. Ugarte Instituto de Suelos, MINAG, Autopista Costa-Costa km 8½, Capdevila, Boyeros, La Habana, Cuba
C. W. A. do Nascimento (*) : C. M. Biondi : Y. J. A. B. da Silva Departamento de Agronomia, UFRPE, Rua Dom Manuel de Medeiros, S/N, Dois Irmãos, Recife 52171-900 PE, Brazil e-mail:
[email protected]
M. R. Alfaro e-mail:
[email protected]
Keywords Trace elements . Soil contamination . Soil pollution . Environmental monitoring
Introduction There has been a growing concern in several countries about the levels of potentially toxic elements in soils and the consequent risks to ecosystems and human health.
O. M. Ugarte e-mail:
[email protected]
C. W. A. Nascimento e-mail:
[email protected]
A. Montero CEADEN, Calle 30 No. 502, Miramar, Playa, Ciudad de la Habana, Cuba e-mail:
[email protected]
A. M. de Aguiar Accioly Embrapa Mandioca e Fruticultura, Rua Embrapa, S/N, Cruz das Almas, BA, Brazil e-mail:
[email protected]
4198, Page 2 of 10
This concern has given rise to specific laws of national or regional scope, which provide tools for monitoring soil quality. The Netherlands was the first country to create a national program for the assessment of soil contamination, establishing intervention levels (Swartjes et al. 2012), and being followed by countries like China (Chen et al. 1991), Austria, Poland, Germany (Kabata-Pendias and Pendias 2000), England (McGrath and Zhao 2006), and Italy (Bini et al. 2011). However, few countries in the southern hemisphere have undertaken efforts towards legislation compatible with their pedological and socioeconomic realities, which are very different to Europe and the USA. For example, the Brazilian legislation establishes three distinct guiding values: quality reference values (QRVs), prevention values (PV), and investigation values (IV) (CONAMA 2009). The QRVs are based on the analysis of metal content in soils under natural condition (little or minimal anthropic interference) while the PVs and IVs are based on risk analysis (Biondi et al. 2011). Taking into account the similarity of the tropical soils from Cuba and Brazil, the Brazilian legislation was used as a basis for comparison. Such studies become even more important in the peculiar socioeconomic situation of Cuba, with limited technical and financial resources for environmental monitoring. Due to its geochemical characteristics, this Caribbean country has soils naturally high in some heavy metals, which become a challenge to the establishment of QRVs for heavy metals. The Republic of Cuba is an archipelago formed by the island of Cuba (104,945 km2), the Isla de la Juventud (2200 km2), and about 4195 islets. The country is located in the Caribbean Sea bordered by the USA to the north, Jamaica to the south, Haiti to the east, and the Yucatan Peninsula to the west. The terrain is mostly plain (75 %) alternating with the three mountainous regions located in the west, central, and east of the island. Cuba has a humid subtropical climate with an average annual temperature of 25 °C in summer and 20 °C in winter. The average rainfall is 1059 mm in the rainy season, from May to October, and 316 mm in the dry season, from November to April (Suárez et al. 2012). Studies assessing the environmental quality of soils in Cuba are very scarce, especially involving various metals on a nationwide scale. Most studies on Cuban soils focus on a small number of metals under specific situations (Muñiz Ugarte 2008; Romero et al. 2010;
Environ Monit Assess (2015) 187:4198
Rodríguez et al. 2012). Therefore, further studies are needed to base the risk assessment of potentially contaminated areas and, consequently, protect the population. In this context, this study aimed at establishing benchmarks of quality for Cd, Pb, Zn, Cu, Ni, Cr, As, Hg, V, Ba, Sb, Ag, Co, Fe, Mn, and Mo in Cuban soils, thus providing a basis for the development of specific regulation to monitor metal levels in soils. Geochemical associations between trace elements and Fe were also studied in order to provide an index for establishing background concentrations of metals in soils (Hamon et al., 2004). To our knowledge, this work represents the first nationwide study on the levels of potentially toxic elements in soils of Cuba.
Materials and methods Soil sampling Sampling locations were defined based on the National Map of Cuban Soils, scale 1:25,000 (Instituto de Suelos 1990), representing the diversity of the country’s soils and their importance to agricultural, urban, industrial, and, especially, mining activities. Surface samples of 33 soil profiles were collected in 13 of the 15 Cuban provinces. Soil samples were collected from the surface horizon of each profile with a stainless steel auger, preferably in areas of native forest or with minimal anthropic influence (Fig. 1). Provinces, geographic coordinates, soil types, geological context, and textural classes are shown in Table 1. Extraction, fractionation, and determination of metals in the soil samples A 1000-g subsample of soil collected from a 5-cm3 sample macerated in an agate mortar and passed through a 0.3-mm stainless steel mesh (ABNT n° 50) was utilized for the extraction of metals by the USEPA method 3051A (USEPA 1998) using high purity acids (Merck PA). Samples were digested in a closed system (Mars Xpress microwave) for 8 min and 40 s in the temperature ramp, the time needed to reach 175 °C; this temperature was maintained for a further 4 min and 30 s. After cooling, the samples were transferred to 25-mL certified flasks (NBRISSO/IEC), which were filled with ultrapure water. The extracts were filtered through slow filter paper (Macherey-Nagel®). Glassware was kept in
Environ Monit Assess (2015) 187:4198
Page 3 of 10, 4198
Fig. 1 Distribution of soil sampling sites in Cuban provinces
a solution of 5 % nitric acid for 24 h and rinsed with distilled water. As some metals (Cu, Ni, Cr, Sb, and As) showed values anomalous and higher than the PV and/or IV established by CONAMA (2009), a chemical fractionation (Shuman 1979) was carried out to identify the bioavailability of these elements and the risk of contaminating the food chain via plant uptake. The elements were separated into fractions (soluble, exchangeable, organic matter, and crystalline iron oxides), as follows: Soluble fraction A total of 5 g of soil was placed in a 50-mL centrifuge tube to which 20 mL of CaCl2 0.01 mol L−1 was added. The mixture was stirred for 2 h and centrifuged for 10 min at 2000×g (Novozamsky et al. 1993). This extractor was chosen due to its ability to estimate the concentration of metals in the soil solution (Melo et al. 2006). Exchangeable fraction Five grams of TFSA and 20 mL of Mg (NO3)2 1 mol L−1 were stirred for 2 h in a 50-mL centrifuge tube. Then, the sample was centrifuged, the supernatant filtered, and 20 mL of distilled water added. The sample was stirred for a further 3 min, centrifuged, and filtered. The two supernatants were combined for analysis. Organic matter fraction A total of 10 mL of NaClO 5– 6 dag L−1, pH 8.5, was added to a centrifuge tube. The sample was heated in a water bath at 100 °C for 30 min, being occasionally stirred. Then, the sample was centrifuged and the supernatant filtered. This procedure was repeated twice, and the three filtrates were combined. After adding 10 mL of distilled
water, the sample was stirred for 3 min, centrifuged, filtered, and the filtrate added to the NaOCl extract from previous extractions. Crystalline iron oxide fraction A total of 30 mL of (NH 4 ) 2 C 2 O 4 (ammonium oxalate) 0.2 mol L −1 + H2C2O4 (oxalic acid) 0.2 mol L−1 + ascorbic acid 0.01 mol L−1, pH 3.0, were placed in contact with the soil sample in a centrifuge tube and heated to 100 °C for 30 min in a water bath, being occasionally stirred. Then, the sample was centrifuged and filtered. The determination of the levels of Cd, Pb, Zn, Cu, Ni, Cr, V, Ba, Sb, Ag, Co, Fe, Mn, and Mo was carried out by optical emission spectrometry (ICP-OES/Optima 7000, Perkin Elmer). Mercury and As were determined by an atomic absorption spectrophotometer (AAnalyst 800/Perkin Elmer) coupled to hydride generator (FIAS 100/Flow Injection System/Perkin Elmer) with electrode discharge lamps (EDL). Quality control was performed using the soil sample with certified values for metals SRM 2709 San Joaquin Soil (Baseline Trace Element Concentrations) certified by the National Institute of Standards and Technology (NIST 2002). Determination of soil QRVs and geochemical associations between Fe and trace elements The benchmark of quality for each metal was calculated based on the 75th percentile of samples, with the anomalies being previously withdrawn through a boxplot. Average, minimum and maximum values, as well as the standard deviations for all elements analyzed, were also calculated. Regression relationships between Fe
4198, Page 4 of 10
Environ Monit Assess (2015) 187:4198
Table 1 Province, geographic coordinates, soil classes, parent material, and textural class of the collected samples Profile
Province
Coordinates
Soil class
Parent material
Textural class
1
Pinar del Río 1
Typic Kandiustalf
Mica schist
Sandy loam
2
Pinar del Río 2
Typic Kandiustalf
Mica schist
Sandy loam
3
La Habana 1
Rhodic Eutrodox
Hard limestone
Clay
4
La Habana 2
Rhodic Eutrodox
Hard limestone
Clay
5
La Habana 3
Rhodic Xanthic Eutrodox
Hard limestone
Clay
6
Mayabeque 1
Typic Rhodudalf
Hard limestone
Clay
7
Mayabeque 2
Typic Rhodudalf
Hard limestone
Clay
8
Mayabeque 3
Typic Rhodudalf
Hard limestone
Clay
9
Artemisa
Rhodic Eutrodox
Hard limestone
Clay
10
Matanzas 1
Rhodic Eutrodox
Limestone
Clay
11
Matanzas 2
Rhodic Eutrodox
Limestone
Clay
12
Villa Clara 1
Typic Haplustoll
Soft limestone
Clay
13
Villa Clara 2
Typic Haplustept
Limestone
Clay
14
Sancti Spiritus 1
Typic Haplustert
Bedload-transported carbonates
Clay
15
Sancti Spiritus 2
Typic Haplustept
Igneous rock
Sandy loam
16
Sancti Spiritus 3
Calcic Haplustept
Sandy shale
Clay
17
Sancti Spiritus 4
Typic Haplustept
Igneous rock
Sandy loam
18
Cienfuegos
Typic Haplustept
Granodiorite
Sandy clay loam
19
Ciego de Ávila
Rhodic Eutrodox
Limestone
Clay
20
Camagüey 1
Udic Haplustert
Bedload-transported carbonates
Clay
21
Camagüey 2
Lithic Haplustept
Ultramafic rock
Clay
22
Camagüey 3
Vertic Haplustept
Igneous rock
Clay
23
Camagüey 4
Calcic Haplustept
Soft limestone
Clay
24
Camagüey 5
Rhodic Eutrodox
Ultramafic rock
Clay
25
Holguín 1
Lithic Haplustept
Ultramafic rock
Clay
26
Holguín 2
Typic Haplustert
Bedload-transported carbonates
Clay
27
Holguín 3
Lithic Haplustept
Igneous rock
Clay
28
Holguín 4
Lithic Haplustept
Soft limestone
Clay
29
Holguín 5
N 271,500 E 215,100 N 269,900 E 212,700 N. 349,700 E. 359,200 N. 346,800 E. 358,300 N. 348,100 E. 355,800 N. 353,200 E. 383,100 N. 353,400 E. 382,800 N. 352,300 E. 383,400 N. 335,400 E. 351,200 N 298,400 E 517,300 N 298,300 E 517,300 N 284,325 E 587,550 N 306.900 E 598.400 N 213.100 E 682.650 N 242,850 E 659,900 N 244,575 E 657,975 N 257,400 E 633,500 N 259,000 E 579,000 N 233,000 E 769,000 N 185,200 E 815,500 N 196,600 E 810,400 N 291,000 E 392,500 N 297,600 E 397,500 N. 185,430 E. 825,200 N. 257,000 E. 567,500 N 213,800 E 556,200 N. 222,100 E. 618,500 N. 229,200 E. 646,200 N. 207,800
Lithic Haplustept
Ultramafic rock
Clay
Environ Monit Assess (2015) 187:4198
Page 5 of 10, 4198
Table 1 (continued) Profile
Province
30
Santiago de Cuba
31
Guantánamo 1
32
Guantánamo 2
33
Guantánamo 3
Coordinates E. 613,500 N. 170,400 E. 581,300 N 658,600 E 183,300 N 176,375 E 659,975 N 172,497 E 663,529
Soil class
Parent material
Textural class
Typic Haplustept
Igneous rock
Clay
Lithic Haplustept
Soft limestone
Clay
Calcic Haplustept
Soft limestone
Clay loam
Typic Haplustept
Siliceous sandstone
Sandy clay loam
and Cd, Pb, Zn, Cu, Ni, Cr, V, Ba, Sb, Ag, Co, Mn, and Mo were determined from the linear correlations of the data, following removal of 25 % of observations with the highest positive residuals. Predictions of background concentrations were made using the regression from the line encompassing the upper 75th percentile of the data set.
Results and discussion The determination of the elements in certified soils based on pseudo-total analyses, as the US EPA method 3051A (USEPA 1998), should not be compared to total certified levels. NIST recommends a comparison with the recoveries based on leaching values (NIST 2002).
Table 2 Recovery of heavy metals in the reference soil (SRM 2709-San Joaquin) by USEPA method 3051A
ND values not determined by the NIST (2002) a
NIST: National Institute of Standards and Technology b
% Recovery (determined)= (determined value/certified value)×100
c
% Recovery (determined) leachate base=(determined recovery / recovery by leachate)×100
Metal
Determined value mg kg−1
Recovery rates of metals in the certificated soil (SRM 2709, San Joaquin soil) compared to leachate values (Table 2) were considered satisfactory for all metals, ranging from 75 to 134 %. The elements with better recoveries were Pb, Zn, Cr, and Ba (108, 100, 108, and 102 %, respectively), whereas Co showed the lowest recovery (75 %). These results indicate the reliability of the method used in the digestion of the samples and determination of metals in soils. The natural concentrations of heavy metals in soils of Cuba had the following decreasing order: Fe>Mn>Ni> Cr > Ba > V > Zn > Cu > Pb > Co > As > Sb > Ag > Cd > Mo>Hg (Table 4). These results are mainly a reflection of the source materials and pedogenic processes, together with the geomorphological and climatic conditions that make the metal levels in each soil specific.
Certified valuea
Recovery by leachateb
Recovery leaching basec
Cd
0.52
0.38±0.01
ND
ND
Pb
14.22
18.9±0.5
69
108
Zn
100.2
94
100
Cu
43.1
34±0.7
92
134
Ni
69.57
88±5
89
89
Cr
85.9
130±4
61
108
V
53.82
112±5
55
87
968±40
41
102
Ba
407.1
106±3
Sb
5.33
7.9±0.6
ND
ND
Ag
0.7
0.41±0.03
ND
ND
Co
9.16
13.4±0.7
90
75
Mo
1.09
0.38±0.01
ND
ND
As
20.32
18.9±0.5
ND
ND
Hg
95.4
106±3
ND
ND
4198, Page 6 of 10
Environ Monit Assess (2015) 187:4198
1983; Ruiz and Pérez Jiménez 1984; Cárdenas et al. 1986). For instance, the Brazilian states of Pernambuco, Rio Grande do Norte and Espírito Santo are rich in soils derived from Precambrian crystalline rocks. The scarcity of heavy metals in these source materials reflects their low natural concentrations in the soil of these states (Table 3). Compared to soils derived from volcanic rocks from the island of Fernando de Noronha, Brazil (Oliveira et al. 2011), and the French Massif Central (SoubrandColin et al. 2005), Cuban soils showed similar average natural metal concentrations, showing higher concentrations of Ni, Zn, Cu, Cr, and Co. These results highlight the importance of establishing QRVs for Cuban soils based on the country’s specific pedological and geological conditions. The use of values from other regions could halt the Cuban economy due to an erroneous
Therefore, to extrapolate soil values obtained in other countries is a misguided strategy (Biondi et al. 2011). This statement is evident when observing the high metal levels found in Cuban soils, which are generally higher than those found in soils of other countries and Brazilian states (Table 3). The average concentration of some metals in Cuban soils exceed the average concentration range for unpolluted soils (Kabata-Pendias and Pendias 1992), the quality standards of Chinese soils (Chen et al. 1991), and even PV for Cu and IV for Ni, Cr, and Sb, according to the Brazilian legislation (CONAMA 2009). This is due to the existence of soils developed on ultramafic rocks, mainly in the eastern region, as well as the fact that soils originated from sedimentary limestone (70 % of Cuban soils) received the influence of arc-shaped ultramafic rocks from the north of the island, with high Ni and Cr contents (Camacho and Paulín
Table 3 Average natural concentrations of heavy metals in Cuban soils compared with data compiled from the international literature Heavy metals (mg kg−1)
Cd
International soils
Brazilian soils
World average
Cuba
Chinaa
Irelandb
USAa
MGc
ESd
PEe
Fernando de Noronhaf
RNg
RO e MTh
Soilsi
1.2
0.07
0.2
1.6
0.5
