COPPER LEVELS IN SOILS AND TWO CROPS IN CENTRAL
CHILE.
Patricio H.
Rodríguez*, Ricardo Badilla-Ohlbaum, Andreas Birkefield, Elena Bustamante,
Andres Céspedes, (CIMM, Santiago-Chile); Rosanna Ginocchio, Gustavo E. Lagos
(Pontificia Universidad Catolica, Santiago-Chile) Juan C. Torres, (CODELCO,
Santiago-Chile)
*prodrigu@cimm.cl
Effect of Soil
Copper Content on Copper Load of Selected Crop Plants in Central Chile
LISTA
DE AUTORES Y AFILIACIONES
CORRESPONDING
AUTHOR
nota: FALTA FORMATEAR CON UN BUEN PROGRAMA GRAFICO
LAS FIGURAS Y TABLAS.
EN TODO CASO,
ESTA VERSIÓN ES ADECUADA PARA COMENTARIOS.
ABSTRACT
We have surveyed
copper levels in agricultural soils and in tissues of onion and tomato plants
grown under field conditions in Central Chile. We found no correlation between
total copper content of edible and non-edible tissues and soil copper for a
broad range of soil copper levels (50-1100 mg/Kg). However, plants grown on
high copper soils had higher levels of copper in their non-edible tissues. The
copper levels of edible tissues were normal. Also, the concentration of copper
extracted by 0.01 M CaCl2 or TCLP was not a predictor of the metal
in either plant tissues. Soils could be separated in two pools: one with copper
levels higher than 400 mg/Kg tend to be clustered in a 20 x 20 Km area
sorrounded by the other soil pool with lower copper levels. Mineralogical
characterization of selected soils indicated that total copper levels are
accounted for by mineral copper species present in these soils, along with slag
particles. Depth profiles revealed high soil copper levels associated to a
surface layer overlaying soil with lower copper levels. Therefore, soil
characterization suggest a geographically limited contamination event but it do
not pose any risk for cultivation of human-edible vegetables.
INTRODUCTION
A potential
source of risk adscribed to soil is the
exposure of humans to metals contained in crop plants growing on metal
contaminated soils. In the present study we have addressed the issue of whether
there is any relationship between copper content of agricultural soils and
copper content of edible and non-edible tissues of two crop plants, onion and
tomato, growing on soils of an important agricultural zone of Central Chile.
Copper mining is
by far the major economic activity of Chile. Given the underlying geochemistry
of the Andean range one would expect high background copper levels, at least in
the regions where rich copper ores have been found. In fact, the surveys
carried out so far have shown rather high copper levels in soils, including
soils from the Antarctic peninsula with as much as 500 mg/Kg (Carrasco &
Préndez, 1991). However, the long history of copper exploitation and
benefitiation associated to certain areas, makes it difficult to assess the
true background levels of those areas.
In this study,
we have measured the levels of copper and other metals as well as the
physicochemical characteristics of a sample of agricultural soils of central
Chile. Also, we have measured the metal content of various tissues of two crop
plants, onion and tomato, growing on the same study soils, in a attempt to
first, establish whether soil copper levels imply any risk to human consumers
of these crops and second, to explore what soil characteristics determine
copper accumulation in the plant tissues.
MATERIAL AND
METHODS
The study was
carried out in agricultural soils of the VI Administrative Region of Chile. Sites where
either of the two crops (Lycopersicon
esculentum and Allium cepa) were
grown were selected on the basis of a
sampling grid modified according to the possibility of access granted by the
farmers. Samples were
collected, prepared and analyzed under clean laboratory techniques following
procedures of the U.S. Department of Agriculture and the Environmental
Protection Agency (U.S.D.A., 1996; U.S. EPA a Document 600/R-95/077).
RESULTS AND
DISCUSSION
Table 1 shows
the statistics of metal concentration for copper, zinc and iron in plant
tissues for both tomato and onion plants. The concentration of copper in the
soils ranged from 50 to 1100 mg/Kg for tomatoes, and from 84 to 900 mg/Kg for
onions and the mean concentration of copper in tomatoes and onions were 15.2
and 7.7 mg/Kg, respectively. Copper concentrations of the corresponding edible
parts, fruits in tomato and bulbs in onions, were well within the range of
reported values for these crops (Szymczak et al.
1993; Zalewski et al., 1994; USDA b) and they did
not correlate with total soil copper.
We carried out
three copper extraction procedures: saturation extraction, simple extraction in
0.01M CaCl2, and a standard leaching procedure (Toxicity
Characteristic Leaching Procedure; U.S. EPA Method 1311), and measured the
concentration of copper in the extract and leachate solutions. Mean copper
levels extracted were very low; the largest fraction of soil copper was 3.8% by
the TCLP method, suggesting that copper in these soils is associated to highly
insoluble forms of the metal. No correlation could be established between
copper concentration in any of the three extracts and copper in the edible
tissues of the two crops. Also, we found that no combination of the measured
soil parameters allowed to make a statistically reliable prediction of the
copper concentration in edible tissues. Copper concentration was also measured
in non-edible tissues of the two crops, stems and leaves. In onions, tissue
copper concentrations were fairly homogeneous for the three tissues, with means
between 6 and 8 mg/Kg d.w. In tomatoes on the other hand, mean tissue
concentrations were 15, 22 and 46 mg /Kg d.w. for fruits, stems and leaves,
respectively.
The mean copper
concentration in the soils of the VI Region was 386 mg/Kg of soil (dry weight),
with values ranging between 50 mg/Kg and more than 1000 mg/Kg. Other soil
parameters were characteristic of agricultural soils of central Chile. The
geographical distribution of the soil copper data for the VI Region strongly
suggested that points corresponding to soils with high copper levels were
clustered. A K-means Cluster Analysis on the the soil total copper
concentration (mg/Kg), the North Universal Transverse Mercator (UTM) and East
UTM coordinates (m) detected three clusters. As shown in Fig. 1, the three
clusters correspond to two areas of soils with lower copper levels (clusters 1
and 2) separated by an area approximately 20 Km wide, where soils have have
higher copper levels (cluster 3). Clusters 1 and 2 were treated together as one
(the low copper soils). The low and high copper soils have significantly different
mean copper contents of 155 and 693 mg/Kg, respectively. Otherwise, the soil
clusters seem to belong to a same soil population. soils.
As shown in
Table 2, when soils are separated in these two pools significant differences
are found between the extracted copper levels by the three methods, as well as
between the copper levels in non-edible tissues of crop plants. Some selective
barrier prevents the higher copper levels reaching the edible tissues. Also, it would seem that a higher extractable copper level
does reflect higher bioavailability but only when the two soil groups are
compared, i.e. no correlations exist within each soil group. This suggests that
a variable other than any of the ones that were measured determines a higher
level of bioavailabilty of copper in the high copper soils.The optical
mineralogical analysis of the study soils revealed the presence of a number of
copper minerals: chalcopyrite (CuFeS2), chalcosine (Cu2S),
coveline (CuS), enargite (Cu3AsS4) as well as slag and
carbon particles. Estimates of the percent of copper from the optical
mineralogical analysis gave figures in very close agreement with the
spectroscopic measurements of the total metal, which suggests that all the
copper in these soils is associated to these minerals. In order to
further explore the high levels of copper found in the high copper soils, we
carried out measurements of total metal content along a depth profile in a
subsample of the study soils. Figure 2 shows results of this study for copper,
zinc and lead. In the high copper soil a sharp decrease of copper concentration
is found below 25 cm, reaching a level characteristic of the low copper soils
below that depth. In a low copper soil, on the other hand, the surface level of
57 mg/Kg remains fairly constant down to the deepest level. The same pattern is
found for zinc and lead levels in both soil types. This reinforces the
hypothesis that copper levels in these soils have a characteristic bimodal
behaviour, and that the very high copper levels found in these soils are
associated to a surface layer of variable depth (20-60 cm) which is deposited
on top of a soil level with significantly lower copper concentrations.
CONCLUSIONS
Some soils of
the study area have very high levels of copper that seem to have resulted from
a major contamination event with either copper ore particles or mining waste
material, or both, at some point in time. Although these soils would be
classified as either contaminated or mineralized, a variety of crops thrive in
the study area, which is consistent with our finding that there exist no
correlation between total soil copper and concentration of the metal in fruits,
stem or leaves of the two crops, including the edible portions. However, minor
but significant differences in the mean tissue copper concentrations are found
when soils are separated in two clusters of low and high copper levels, which
is correlated with slightly higher levels of the copper that can be extracted
by three different methods from the soils of the high copper cluster.
The
mineralogical analysis shows that copper in these soils is associated to
mineral forms, which would account for the very low levels of copper extraction
by standard methods, and the low bioavailability reflected by the tissue levels
similar to those reported for plants growing on non-contaminated soils. It
remains to be shown whether other soil characteristics, such as dissolved
organic carbon, do play a role in keeping soluble copper at low levels.
Alternatively, humic acids have been shown to be able to cause mineral
dissolution (Schnitzer, 1986) Also, the results do not allow to establish
whether the copper taken up by the plants from these soil comes from the
mineralized copper particles, acting as a reservoir of the metal, or from other
sources of the metal.
REFERENCES
Carrasco, M. A.,
and Préndez, M. (1991). Element distribution of some soils of continental Chile
and the Antarctic peninsula. Projection to atmospheric pollution. Water, air, and soil Pollution 57-58, 713-722.
Szymczak, J.,
Ilow, R., and Ilow, B. R. (1993). Contents of copper and zinc in vegetables,
fruit and cereals from areas with a different degree of industrial pollution
and from green-houses. Rocz. Panstw.
Zakl. Hig. 44, 347-359.
United States Department of Agriculture, USDA a
(1996): "Soil Survey Laboratory Methods Manual"
United States Department of Agriculture, USDA b:
Nutrient Database (http://www.nal.usda.gov/fnic/foodcomp/)
U.S. EPA a Document 600/R-95/077: "Laboratory
methods for soil and foliar analysis in long-term environmental monitoring
programs"
Zalewski, W.,
Oprzadek, K., Syrocka, K., Lipinska, J., and Jaroszynska, J. (1994). Value of
harmful elements in fruit and vegetables grown in the province of Siedlce. Rocz. Panastw. Zakl.
Hig. 45, 19-26.Abstract
In this study wWe
have surveyed copper levels, along with a
number of other soil
parameters, in agricultural
soils and inplant tissues of
onion and tomato plants grown on
agricultural soils of in Central
Chile. We found no correlation between copper content of edible and non-edible
tissues and soil copper forfor plants growing on
quite a broad range of of total soil copper
levels (50-1100 mg/Kg), indicating
that the very high total soil copper contents do not represent a risk to human
consumers of these crops. The copper
levels of edible tissues were similar to
those reported by others for these foodsnormal.
Also, the concentration of copper extracted by three
different methods, saturation, extract, 0.01 M CaCl22
extract and acid leachingTCLP procedure, was not a good predictor of the
level of accumulated metal
in either plant tissues type. Statistical analysis showed that soils
can be separated in two pools, depending on
their copper level and geographic position. S:
soils with copper levels higher than 400 mg/Kg tend
to be clustered in a 20 x 20 Km area with little
overlap withsorrounded by
soils with lower copper levels, suggesting a geographically limited
contamination event.Plants grown on
high copper soils had: higher levels of copper in their non-edible tissues. High copper
soils have significatively higher
cadmium and molybdenum levels, but otherwise they show no significative differences
with low copper soils. However, the
three extraction procedures released significatively higher copper levels from the high copper soils,
though the extracted fractions of metal were still very low. Also, separation
of plants according to the soil pool where they were growing, revealed that for
non-edible tissues, plants growing on high copper soils had significatively higher
copper contents. No such difference was found for tomato fruits and onion
bulbs, the edible portions of these crops.
Mineralogical
characterization of selected soils indicated that total copper levels are
accounted for by mineral copper species present in these soils. Also, this analysis revealed the presence of,
along with slag and matte particles, wastes associated to smelting activities, though
these species did not contain significative amounts of copper. Depth profiles
of a few soils indicatesreveal that high soil
copper levels seem to be associated
to a surface layer 20-50 cm deep
overlaying soil with lower copper levels, which
supports the hypothesis of a relatively
recent contamination event in this area..
Introduction
Concern for the content of trace metals in soils
has increased in industrialized countries in the last decade. Both research and
regulatory approaches have focused on soil metal content as a potential source
of exposure for humans, cattle and wildlife (Manicol & Beckett, 1985; Sauvé
et al., 1996). In general, risk assessment methodologies and regulatory
developments have been so far based on total metal in the soil. However, metal
toxicity to organisms is a complex function of the physical and chemical
characteristics of the soil as well as of the physiological mechanisms of
uptake of the organism (Allen et al., 1994b; Houba et al., 1996; Plette et al.,
1999). In the absence of sound and readily applicable predictive models,
regulatory agencies rely on the “precautionary principle”, often leading to
unnecesarily restrictive or simply unattainable soil quality standards.
The function relating soil parameters to metal
bioavailability seems to be a complex one and several models have been used to
approach the problem (Allen et al., 1994a; Allen et al., 1994b; Gárate et al.,
1993; Hesterberg et al., 1993; Ma, 1999; McKenna et al., 1993). Dissolved
organic carbon, pH and calcium play a major role in determining metal sorption and
complexation reactions that decrease the concentration of soluble and
bioavailable forms of the metal, but this will vary with the metal, soil type
and the form in which the metal itself enters the soil (Boon et al., 1998;
Bowers et al., 1997). However, most of the studies on metal contaminated soils
have not addressed the latter, i.e. what are the specific copper species that
account for the contaminating levels and how heterogeneous is that population
of species.
In the case of agricultural soils, metals have
received a great deal of attention because of their potential effect on crop
yields and soil quality in the long term (Manicol & Beckett, 1985; Korthals
et al., 1996; Lexmond, 1980). In countries where sewage sludges are routinely
applied as fertilizers, there is growing pressure for more restrictive
regulation of sludge application to agricultural lands, in spite of the fact
that metals in sludges are mostly highly complexed and unavailable to organisms
(Davis & Carlton-Smith, 1984; Wang et al., ).
In agricultural soils where sewage sludge is not
used, copper and other metals may come from irrigation water, fertilizers and
pesticides, natural minerals, contamination from other sources, or a mixture of
all of them. This complexity is further compounded by the fact that historical
records of land use practices, natural events (floodings, lanmdslides...etc.)
and discontinued industrial activies are seldom available.
AA
potential source of risk adscribed to soil metals, and an
understandable source of concern for farmers and consumers,
is
the exposure of humans to metals contained in crop plants growing on metal
contaminated soils. In the present study, part of a
broader attempt to assess the potential exposure to copper of the human
population in Chile, we have
addressed the issue of whether there is any relationship between copper content
of agricultural soils and copper content of edible and non-edible tissues of
two crop plants, onion and tomato, growing on soils of an important
agricultural zone of Central Chile.
Copper
mining is the by far the
major economic activity of Chile. Both active
and inactive copper mines and their associated operations are found all
througout the North and Central regions of the country, ranging from small and
technologically primitive works to Chuquicamata, the largest open pit mine in
the world. Given the underlying geochemistry of the Andean
range one would expect high background copper levels, at least in the regions
where rich copper ores have been found. In fact, the very few surveys carried
out so far have shown rather high copper levels in soils, including soils from
the Antarctic peninsula with as much as 500 mg/Kg (Carrasco & Préndez, 1991; González, 1986). However, the
long history of copper exploitation and benefitiation associated to certain
areas, makes it difficult to assess the true background levels of those areas.
In
this study, we have measured the levels of copper and other metals as well as
the physicochemical characteristics of a sample of agricultural soils of
central Chile. Also, we have measured the metal content of various tissues of
two crop plants, onion and tomato, growing on the same study soils, in a
attempt to first, establish whether soil copper levels imply any risk to human
consumers of these crops and second, to explore what soil characteristics
determine copper accumulation in the plant tissues.
MATERIAL
AND METHODS
The
study was carried out in agricultural soils of the VI Administrative Region of
Chile, in sites located W of the city of Rancagua, along
the Cachapoal River valley. Agriculture is one of the main activities of the
area, and a variety of crops are grown in relatively small farm plots:
potatoes, wheat, corn, onions, tomatoes, cereals, vines...etc. Upstream the river, some 50 Km east of Rancagua, the El Teniente copper mine and Caletones smelter
are located in an area with a long history of copper mining.
Sample Collection.
Sites
where either of the two crops (Lycopersicon
esculentum and Allium cepa) were grown (Lycopersicon esculentum and Allium cepa)
were selected on the basis of a sampling grid modified according to the
possibility of access granted by the farmers to the types
of crops selected for the study. Samples were
collected, prepared and analyzed under clean laboratory techniques
following procedures of the U.S. Department of Agriculture and the
Environmental Protection Agency(U.S.D.A., 1996; U.S. EPA a
Document 600/R-95/077).
. Soil samples
were taken to a depth of 20 cm using a stainless steel hand auger (Eijkelkamp).
Before taking the soil samples the hand auger was conditioned with the soil of
the sampling site. The plant samples were taken by digging them out with a
stainless steel shovel. Both sample types (soil and plants) were stored in Polyethylene plastic
bags which were carried in Polyethylene plastic containers. The samples were transferred into the
laboratory after each sampling day. The plant samples were than immediately introduced into the drying process to avoid decomposition.
Sample Preparation (total metal content)
The soil samples were dried in an forced air drying
cabinet at a temperature of approximately 30ºC. After drying the soil samples
were rid of coarse particles (roots, stones, etc) an than sieved
through a 2 mm meshsize Polyethylene sieve. Aliquots of 50 mg of the soil
fractions < 2 mm were grinded in an agate ball mill an stored in Polyethylene
wide mouth sample bottles. The rest of the soil samples are stored in their Polyethylene
bags inside a plastic container. The soil sample preparation was done with
protocols of the ISO (ISO, 1994) and the USDA (USDA, 1996) and the plant sample
preparation after a document of the USEPA (USEPA, 1995).
The soil
samples underwent then a microwave acid digestion to prepare them for the
determination of Cu, Pb, Zn, Cd, Fe, Ti, Mo, Ca. The digestion was done after a modified protocol
of the USEPA (USEPA Method 3052) in a
MILESTONE mls 1200 mega
microwave digestion system. 9 ml HNO3 Suprapur (Merck), 5 ml HF p.a.
(Riedel) and 2 ml H2O2 p.a. (Merck) were added to 0,25g of soil sample and
then digested for 24 min (6 min - 250 W, 6 min -
400 W, 6 min - 650 W and 6 min 250 W). After the digestion followed an evaporation phase of 15 min at 800 W. To the evaporated,
viscous sample 5 ml HNO3 Suprapur (Merck) was added and heated 5 min
at 800 W. Finally the sample was cooled down to room temperature and then
transferred to a 50 ml volumetric flask and made up to volume with
distilled and deionized water (H2O DDI, >18MW) and acidified with HNO3 Suprapur (Merck) to 0,2%.
The plant samples were separated into roots leaves
stems and fruits. The separated parts were washed with 0,01N HCl - distilled
water - 0,05M EDTA - distilled and deionized water. After the washing the
plants were dried in normal ventilated drying cabinets at a temperature of
approximately 60º C. The samples were pre-crushed after drying with an agate
mortar an pestle and than fine crushed in an agate ball mill. Only the fruits
of the tomato plants got an additional treatment with liquid nitrogen to freeze
remaining moisture before introducing the sample in the agate ball mill. The
crushed vegetal samples were stored in Polyethylene sample bottles. The vegetal samples were
digested after the same modified USEPA protocol used for soils. 6 ml HNO3 Suprapur
(Merck) and 1 ml H2O2 (Merck) were added to approx. 0,25 g
vegetal sample and the mixture was
digested in the microwave oven for 20 min (2 min - 250 W,
2 min - 0 W, 6 min - 250 W, 5 min - 400 W and
5 min - 600 W). The evaporation phase was 5 min at 800 W.
After evaporation the samples were transferred to 25 ml volumetric flask
and filled to volume with distilled and deionized water (H2O DDI, >18MW) and acidified to 0,2% HNO3 Suprapur (Merck). Every digestion batch had one blank sample, one
standard reference material sample, one duplicate sample and one quality
control sample (QCPS) for matching the QA/QC criteria.
Sample Preparation and Analysis (TOC, pH, Cu+Ca in
saturation extract, CE, CEC)
For estimating the other physical chemical
parameters the soils samples were prepared according to the requirements of the
to the different analytical methods. Preparing a saturation extract for analysis of Ca and Cu was done after a USDA (USDA, 1996) procedure.
Distilled and demineralized water is added to a soil sample until it reaches
saturation criteria, e.g. slowly flowing when the sample-container is tipped.
The saturated sampled was allowed to stand overnight and rechecked the
following day for its saturation behavior. If the check was positive a subsample
was taken to analyze the water content and the main sample was transferred to
an Büchner filter funnel. The extract was obtained through vacuum filtration.
The 10 mM
CaCl2 extract was carried out by adding 50 g of soil to 50 ml of 10 mM CaCl2,
mixing thoroughly and then letting the mixture stand overnight. Then the sample
was transferred to a Büchner funnel and filtered. Copper was measured in the
filtrate by atomic absortion spectrometry.
The Toxicity
Characteristic Leaching Procedure was performed by Lakefield Research Chile
S.A. following Procedure USEPA 1312.
The soil pH
was determined in 1:1 soil-water and 1:2 soil-CaCl2 solutions after a protocol
of the USDA.
The total
organic carbon content of the soil samples was analyzed after the Walkley
Black method, described in the USDA methods manual. The sample is wet ashed by
a mixture of 1N potassium dichromate (K2Cr2O7) and concentrated sulfuric acid
(H2SO4). After 30 min. of reaction the excess K2Cr2O7 is
potentiometrically back-titrated with ferrous (II) sulfate (FeSO4). The reduced dichromate produced
during reaction with the soil is assumed to be equivalent to the total organic
carbon content in the soil.
The soil texture was analyzed after a protocol of
Gee & Bauder (Gee & Bauder, 1986). In a pretreated soil suspension
(removal of organic matter, salts, iron oxides and carbonates) the density is
measured with an hydrometer after defined settling times.
Electrical
conductivity in the saturation extract was analyzed with an conductivimeter
after a USDA procedure.Calcium and Copper in the saturation extract was analyzed with atomic
absorption spectrometry (USDA, 1996). The cation exchange capacity (CEC) of the
soil is analyzed after a procedure of USDA. The sample is saturated with 1N ammonium acetate solution
at pH 7. The absorbed (exchanged) ammonia is determined by Kjedahl distillation
of the soil suspension and subsequent titration of the distillate with
0,2 N HCl.
Sample Analysis (total metal content)
The digested soil samples were analyzed for their
total metal contents (Cu, Zn, Pb, Cd and Mo) after method protocols SW-486 of USEPA. Cadmium,
lead and molybdenum were analyzed with graphite
furnace-atomic absorption spectrometry using a
Perkin Elmer AAnalyst 300 spectrometer with HGA 800 graphite furnace.
Copper and zinc were analyzed with flame-atomic absorption spectrometry using a Perkin Elmer
AAnalyst 300 spectrometer. In samples with low Cu and Zn levels the graphite
furnace technique was applied. Background non atomic absorption was corrected with an deuterium
continuos lamp. The atomic absorption analytical device is housed in a class
1000 clean-room laboratory and the loading of the autosampler tray was done in
a class 100 laminar flow cabinet. The individual elements were analyzed with
element specific USEPA methods.
The
calibration standards were prepared with high purity water (H2O DDI, >18MW) and acidified with HNO3 Suprapur (Merck) to
0,2 %. For performance control of the AAS spectrometer a certified
multielement standard[1] was used.
Mineralogical Characterization
A set of eight soil-samples originated from the
metropolitan and sixth region in the central valley of Chile were characterized
using mineral optical analysis. The mineralogical
laboratory received the samples presieved (2mm) and milled in an agate ball
mill. This loose samples were
stabilized with a special optical resin and than ground to its analytical
thickness. The non transparent (opaque) minerals, which are mostly ores, were analyzed by reflected light polarization microscopy
the transparent minerals by thin section polarization microscopy.
RESULTS
AND DISCUSSION
Metals in Edible Tissues
Table 1 shows
the statistics of metal concentration
for copper, zinc
and iron in plant
tissues for both tomato and onion plants. Copper concentrations of the
corresponding edible parts, fruits in tomato and bulbs in onions, were well
within the range of reported values for these crops (Szymczak et al. 1993;
Zalewski et al., 1994; USDA b). For instance,
the USDA Nutrient Database gives a value of 0.74 mg/Kg (wet weight) with a mean
water content of 937.6 g/Kg, that is a calculated 11.8 mg/Kg of dry weight; the
value for onions is an average of 5.8 mg/Kg d.w. For onion, however, bulb zinc
and iron concentrations were significatively higher than USDA reported values. The same was
found for zinc in tomato fruits. Since this is the first report on metal contents of
these crops in Chile, we do not know whether this is
characteristic of other agricultural areas in the country.
For
the edible tissues of both crops, tomato fruits and onion bulbs, there was no
correlation between total soil copper and tissue copper content (Fig. 1): plants growing in soils with copper concentrations
rangingthe
concentration of copper in the soils ranged
from 50 to 1100 mg/Kg for tomatoes, and from 84 to 900 mg/Kg for onions, produce tomatoes and onions with the same mean
copper content. The meand
concentration of copper in tomatoes and onions were 15.2 and 7.7
mg/Kg, respectively. This is not surprising, since it has been shown for
a variety of media and organisms that total metal concentration is
not a good estimator of metal bioavailability (Allen et al., 1994; Janssen et
al., 1997; Moolenar et al., 1997).
Different soil metal extraction procedures have
been used to better estimate the pool of metal that is bioavailable to plants
(Sauvé et al., 1996; Sauvé et al., 1997;
Temminghof et al., 1997). In our case, wWe
carried out three type of metal
extraction procedures: saturation extraction, simple extraction in 0.01 MM CaCl22,
and a standard leaching procedure (Toxicity Characteristic Leaching Procedure;
U.S. EPA Method 1311) designed to
assess hazard characteristics of solid wastes, and measured the concentration
of copper in the extract and leachate solutions.
(U.S. EPA
Method 1311). Mean copper
levels in extracted
were very low:s
and leachates are shown in Table 2. Tthe
maximal largest
fraction of soil copper was 3.8%
extracted by the TCLP
method (3.8%), reflecting the stronger extraction
solution, which contains acetic acid. In all cases, though, the fraction of
extracted copper was minimal,
which suggestssuggesting
that the high copper levels in these soils isare associated to
highly insoluble forms of the metal.
Again, though, nNo
correlation could be established between copper concentration in either any
of the three extracts and
copper in the edible tissues of the two crops. Also, statistical analysis showed thatwe
found that no combination
of the measured soil parameters allowed to
make a to predictstatistically
reliable prediction of either extractable copper or the
copper concentration in edible tissues. More complex models have been used to approach this
problem (Temminghof et al., 1995; Römkens & Dolfing, 1998) but, as we will show in the next sections, it seems
that in these soils copper is found in such a form that the total concentration
bears little relationship to the bioavailable fraction.
Metals in Other Tissues
Copper
concentration was also measured in non-edible tissues of the two crops: stems
and leaves. In onions,
tissue copper concentrations were fairly homogeneous for the three tissues,
with means between 6 and 8 mg/Kg d.w. In tomatoes on
the other hand, mean tissue
concentrations were 15, 22 and 46
mg /Kg d.w. for fruits, stems and leaves, respectively: leaf > stem > fruit (p<0.001; One Way ANOVA), with leaves
containing as much as 3 fold copper concentrations than fruits (see Table 1).
Figures 2A and 2B shows plots of the
relative tissue copper level as a function of the total soil copper
concentration for leaves and stems, respectively, in both crops. Using the
entire data set in each case, it is possible
to find positive and significative correlations between these variables for both
tissue types. However, the correlations break down if you analyse separately
the tissue data of plants growing on soils with copper levels lower and higher
than 400 mg/Kg. It would seem then that the apparent correlations result simply from the fact
that there are two clouds of points around different mean soil and tissue
copper levels.
Separating the tissue data of these two pools and
comparing the mean copper concentrations gives the results shown in Table 3.
For both crops there are significative differences between the groups for stems and
leaves, but not for the edible tissues. The mean increases in tissue copper range from 32%
(tomato stems) to 59% (onion stems). In the next section we will further
discuss the possible implications of this observation.
Soil Characterization
Except for total soil copper levels, the study soil characteristics were quite homogenous
(Table 4). For instance, more than 90% of the soil pH values were in the
7.0-8.5 range. In terms of texture, the main soil types were: loam (65%), clay
loam (9.8%), silt loam (8.1%) and sandy loam (4%). In general, the physicochemical characteristics
of the sampled soils were typical of agricultural soils of Central Chile
(González, 1986; González, 1990; González, 1991).
As shown in Table 4, The mean
copper concentration in the soils of the VI Region was 386 mg/Kg of soil (dry
weight), with values ranging between 50 mg/Kg and more than 1000 mg/Kg.
Other soil parameters were characteristic of agricultural soils of central
Chile. This mean was significatively higher than the 182 mg/Kg found in a previous
study on agricultural soils of the Metropolitan Region (MR) located north of
the VI Region. Also, in the MR, very few soil samples with more than 200 mg/Kg
of total copper were found.
Inspection The
geographical distribution of the soil
copper data on a map of
the stusy areafor
the VI Region strongly
suggested that points
corresponding to soils with high copper levelsthat higher soil copper points
were clustered in a defined
area. In order to
check this, we performed aA
K-means Cluster Analysis on the three
following variables: the soil total
copper concentration (mg/Kg), the North Universal Transverse Mercator (UTM) and
East UTM coordinates (m)in meters. This method
of multivariate cluster analysis allows to identify groupings of cases without a priori knowledge of
group membership or an underlying theoretical model.
The analysis detected three
clusters. As shown in the map of
Figure 3A, and the cross section of Fig. 3B1, , the
three clusters correspond to two areas of soils with lower copper levels
(clusters 1 and 2) separated by an
area approximately 20 Km wide, where soils have have higher copper levels
(cluster 3). For the purposes of the following analysis, cClusters
1 and 2 were treated together as one pool (the
low copper soils). The low and high copper soils have significativesignificantly different ly different mean
copper contentss
of 155 and 693 mg/Kg, respectively (p<0.001;
ANOVA). Notice that the distribution of soil copper
levels is not continuous: only three sites have copper levels in the 350-500
mg/Kg range. This pattern strongly suggests that a major contamination event
occurred in this area at some point in time, which introduced a 20 Km wide
wedge of material with high copper content on top of soils with lower copper levels.
If other soil
parameter data are separated in the two clusters defined by the soil copper
level, and compared, few significative differences are found. High copper
soils have slightly but significatively higher total cadmium (0.41 vs. 0.2 mg/Kg;
p<0.005) and molybdenum (8.6 vs. 4.4 mg/Kg; p<0.005) levels..
Otherwise, the soil clusters seem to belong to a same soil population. This suggests either that the contaminating
material was mainly composed of copper in some form or that it was mixed with
the same type of soil found in the areas sorrounding the high copper soils.
As
shown in Table 2, when
soils are separated in these two pools significant differences are found
between the extracted copper levels by the
three methods, as well as
between the copper levels in non-edible tissues of crop plants. Some selective
barrier prevents the higher copper levels reaching the edible tissues. Also, iRegarding the extracted copper level by the three
extraction methods used (see above), there were also significative differences
when these data were separated in low and high total copper soils, as shown in
Table 5. In the case of the saturation and CaCl2 extracts, these differences do not reflect a
correlation between copper in the extract and total soil copper. As shown in
Figure 4 for the CaCl2 extract, higher levels of copper are extracted
from soils with higher total copper, but within each pool there are no significative correlations
between the variables. A similar situation was observed for the saturation
extracts. On the other hand, only within the low copper soils we found a significative positive
correlation between TCLP leached copper and total soil copper (r2 = 0.76;
p<0.01).
As discussed in the previous section, leaves and stems, but not edible tissues, of plants
growing on high copper soils do accumulate higher copper levels. It would seem then that
a higher extractable copper level does reflect higher bioavailability but only
when the two soil groups are compared, i.e. no correlations exist within each soil
group regardless of the rather wide range of total
copper levels represented in each..
This suggests that a variable other than any of the ones that were measured
determines a higher level of bioavailabilty of copper in the high copper soils.
Mineral Content of Soils
The
optical mineralogical analysis of the study soils revealed the presence of a
number of copper minerals: chalcopyrite (CuFeS2), chalcosine (Cu2S), coveline
(CuS), enargite (Cu3AsS4) and molibdenite
(MoS2) as well as . These minerals can originate from both natural
ore deposits and mine waste material. In general, all major copper outcrops in
Chile are of the porphyric type. Besides the
copper minerals listed, other minerals are found in this type of outcrops:
pyrite (FeS2), magnetite (Fe3O4) and hematite (Fe2O3).
Also, we found slag
and carbon particles. that can only come from mining activities. In fact, further upriver there is a major copper
smelter.
As sown in Table 6 e
Estimates of the percent of copper from the optical
mineralogical analysis gave figures in very close agreement with the
spectroscopic measurements of the total metal. This
indicates, which suggests
that almost all the copper
in these soils is associated to these
minerals forms listed above.
Also, the
Table shows the estimated concentration of slag and matte particles, indicators of anthropogenic contamination from
smelting.
In
order to further explore the nature of thehigh
levels of metal contaminationcopper
found in the hHigh copperCu sSoils, we carried
out measurements of total metal content along a depth profile at a few selected sitesin
a subsample of the study soils. Figure 25 shows some representative results of this
study for copper, zinc and lead. The panels on
the left show the copper and zinc profiles for a high copper soil while those on the left show the copper and lead profiles of a low copper soil. In
the high copper soil a sharp decrease of copper concentration is found below 25
cm, reaching a level characteristic of the low copper soils below that depth.
In a low copper soil, on the other hand, the surface level of 57 mg/Kg remains
fairly constant down to the deepest level. The same pattern is found for zinc
and lead levels in both soil typesoth soils. The 20-30cm depth in agricultural soils is related
to the influence of the plow depth. So
every contamination in this area is mechanically incorporated in the soil.This
result reinforces theour hypothesis that
copper levels in these soils have a characteristic bimodal behaviour.
The pattern found in the depth profiles strongly suggest,
and that the very high copper levels found in these
soils are associated to a surface layer of variable depth (20-60 cm) which is
deposited on top of a soil level with significativesignificantly
lower copper concentrations. This supports
the hypothesis that these high copper levels might be associated to a flooding or
landslide event which carried copper containing particles and other materials
down the valley. It remains to be shown whether the mineral copper in these
soils comes mainly from natural mineral deposits exposed by mining or from
mineral containing wastes.
Conclusions
Some
soils of the study area have very high levels of copper that seem to have
resulted from a major contamination event with either copper ore particles or
mining waste material, or both, at some point in time. Although these soils
would be classified as either contaminated or mineralized, a variety of crops
thrive in the study area, which is consistent with our finding that there exist
no correlation between total soil copper and concentration of the metal in
fruits, stem or leaves of the two crops, including the edible portions.
However, minor but significativesignificant
differences in the mean tissue copper concentrations are found when soils are
separated in two clusters of low and high copper levels, which is correlated
with slightly higher levels of the copper that can be extracted by three
different methods from the soils of the high copper cluster.
The
mineralogical analysis shows that copper in these soils is associated to
mineral forms, which would account for the very low levels of copper extraction
by standard methods, and the low bioavailability reflected by the tissue levels
similar to those reported for plants growing on non-contaminated soils. It
remains to be shown whether other soil characteristics, such as dissolved
organic carbon, do play a role in keeping soluble copper at low levels.
Alternatively, humic acids have been shown to be able to cause mineral
dissolution (Schnitzer,
1986) Also, the
results do not allow to establish whether the copper taken up by the plants
from these soil comes from the mineralized copper particles, acting as a
reservoir of the metal, or from other sources of the metal.
References
Carrasco,
M. A., and Préndez, M. (1991). Element distribution of some soils of
continental Chile and the Antarctic peninsula. Projection to atmospheric
pollution. Water, air, and soil Pollution 57-58, 713-722.
Szymczak,
J., Ilow, R., and Ilow, B. R. (1993). Contents of copper and zinc in
vegetables, fruit and cereals from areas with a different degree of industrial
pollution and from green-houses. Rocz. Panstw. Zakl. Hig. 44, 347-359.
United
States Department of Agriculture, USDA a (1996): "Soil Survey Laboratory
Methods Manual"
United
States Department of Agriculture, USDA b: Nutrient Database (http://www.nal.usda.gov/fnic/foodcomp/)
U.S.
EPA a Document 600/R-95/077: "Laboratory methods for soil and foliar analysis
in long-term environmental monitoring programs"
Zalewski,
W., Oprzadek, K., Syrocka, K., Lipinska, J., and Jaroszynska, J. (1994). Value
of harmful elements in fruit and vegetables grown in the province of Siedlce.
Rocz. Panastw. Zakl. Hig. 45, 19-26.
REFERENCES
Allen, H. E.,
Chen, Y.-T., Y. Li, S.-Z. L., Huang, C. P., and Sanders, P. F. (1994). Soil
partition coefficients for Cd by desorption and comparison to adsorption
measurements. American Chemical Society, Division of
Environmental Chemistry Preprints 34, 608-611.
Allen, H. E.,
Huang, C. P., Bailey, G. W., and Bowers, A. R. (1994). “Metal Speciation and
Contamination of Soil,” Lewis Publishers, Boca Raton, FL.
Boon, G. T.,
Bouwman, L. A., Bloem, J., and Römkens, P. F. A. M. (1998). Effects of a
copper-tolerant grass (Agrostis capillaris) on the ecosystem of a coppe-contaminated soil. Environmental
Toxicology and Chemistry 17, 1964-1971.
Bowers, N.,
Pratt, J. R., Beeson, D., and Lewis, M. (1997). Comparative evaluation of soil
toxicity using lettuce seeds and soil ciliates. Environmental
Toxicology and Chemistry 16, 207-213.
Carrasco, M.
A., and Préndez, M. (1991). Element distribution of some soils of continental
Chile and the Antarctic peninsula. Projection to atmospheric pollution. Water, air,
and soil Pollution 57-58, 713-722.
Davis, R. D.,
and Carlton-Smith (1984). An investigation into the phytotoxicity of zinc,
copper and nickel using sewage sludge of controlled metal content. Environmental
Pollution Series B8, 163-185.
Garate, A., Ramos,
I., Manzanares, M., and Lucena, J. J. (1993). Cadmium uptake and distribution
in three cultivars of Lactuca sp. Bulletin of
Environmental Contamination and Toxicology 50, 709-716.
Gonzáles, S.
(1986). Contenido natural de metales pesados extraíbles con EDTA en suelos del
Valle Aconcagua. Agric. Tecn. 46, 323-327.
González, S.
(1990). Prospección de elementos traza en recursos agrícolas de Chile. Aguas y
suelos de los valles Elquí y Limarí, IV Región. Agric. Tecn. 50, 60-70.
González, S.
(1991). Prospección de elementos traza en recursos agrícolas de Chile. II:
Aguas de los valles Maipo y Mapocho, R.M. Agric. Tecn. 51, 22-29.
Hesterberg,
D., Bril, J., and Castilho, P. d. (1993). Thermodynamic modeling of zinc,
cadmium, and copper solubilities in a manured, acidic, loamy-sand topsoil. Journal of
Environmental Quality 22, 681-688.
International
Organization for Standardization , ISO 11464: "Soil quality - Pretreatment
of samples for physico-chemical analysis" first edition 1994
Houba, V. J.
G., Lexmond, T. M., Novozamsky, I., and Lee, J. J. v. d. (1996). State of the
art and future developments in soil analysis for bioavailability assessment. Science of the
Total Environment 178, 21-28.
Janssen, R. P.
T., Posthuma, L., Baerselman, R., Hollander, H. A. D., Veen, R. P. M. V., and
Peijnenburg, W. j. G. m. (1997). Equilibrium partitioning of heavy metals in
Dutch field soils. II. Prediction of metal accumulation in earthworms. Environmental
Toxicology and Chemistry 16, 2479-2488.
Korthals, G.
W., Alexiev, A. D., Lexmond, T. M., Kammenga, J. E., and Bongers, T. (1996).
Long term effects of copper and pH on the nematode community in an
agroecosystem. Environmental Toxicology and Chemistry 15, 979-985.
Lexmond, T. M.
(1980). The effect of soil pH on copper toxicity for forage maize grown under
field conditions. Netherlands Journal of agricultural Science 28, 164-184.
Ma, H. (1999).
Kinetics of copper interactions with dissolved organic matter and suspended
particles and its effect on copper toxicity to Ceriodaphnia dubia. Ph.D.
Thesis., University of Delaware.
Manicol, R.
D., and Beckett, P. H. T. (1985). Critical tissue concentrations of potentially
toxic elements. Plant and Soil 85, 107-129.
McKenna, I.
M., Chaney, R. L., and Williams, F. M. (1993). The effects of cadmium and zinc
interactions on the accumulation and tissue distribution of zinc and cadmium in
lettuce and spinach. Environmental Pollution 79, 113-120.
Moolenar, S.
W., Lexmond, T. M., and Zee, S. E. A. T. M. v. d. (1997). Calculating heavy metal
accumulation in soil: acomparison of methods illustrated by a case-study on
compost application. Agriculture, Ecosystems and Environment 66, 71-82.
Peijnenburg,
W., Baerselman, R., Groot, A. d., Jager, T., Leenders, D., Posthuma, L., and
Veen, R. V. Quantification of metal bioavailability for lettuce (Lactuca sativa L.) in field soils. .
Plette, A. C.
C., Nederlof, M. M., Temminghof, E. J. M., and Riemsdijk, W. H. V. (1999).
Bioavailability of heavy metals in terrestrial and aquatic systems: a quantitative
approach. Environmental Toxicology and Chemistry 18, 1882-1890.
Römkens, P.,
and Dolfing, J. (1998). Effect of Ca on solubility and molecular size
distribution of DOC and Cu binding in soil solution samples. Environmental
Science and Technology 32, 363-369.
Sauvé, S.,
Cook, N., Hendershot, W. H., and Bride, M. B. M. (1996). Linking plant tissue
concentrations and soil copper pools in urban contaminated soils. Environmental
Pollution 94, 151-157.
Sauvé, S.,
McBride, M. B., Norvell, W. A., and Hendershot, W. H. (1997). Copper solubility
and speciation of in situ contaminated
soils: effect of copper level, pH and organic matter. Water, Air,
and Soil Pollution 100, 133-149.
Sparks, D. L.
(1995). “Environmental soil chemistry,” Academic Press, New York.
Szymczak, J.,
Ilow, R., and Ilow, B. R. (1993). Contents of copper and zinc in vegetables,
fruit and cereals from areas with a different degree of industrial pollution
and from green-houses. Rocz. Panstw. Zakl. Hig. 44, 347-359.
Temminghof, E.
J. M., Plette, A. C. C., Zee, S. E. A. T. M. V. d., and Riemsdijk, W. H. V.
(1997). Speciation of heavy metals in soil in relation to availability and
mobility. Recent Res. Devel. in Soil Science 1, 55-65.
Temminghoff,
E. J. M., Zee, S. E. A. T. M. V. d., and Haan, F. A. M. d. (1995). Speciation
and calcium competition effects on cadmium sorption by sandy soil at various
pHs. European
Journal of Soil Science 46, 649-655.
U. S. EPA
(1991). “Methods for the determination of metals in environmental samples,”
Rep. No. EPA 600/4-91/010, Washington, D.C.
United States Environmental Protection Agency, USEPA Document 600/R-95/077: "Laboratory
methods for soil and foliar analysis in long-term environmental monitoring
programs"
United States
Environmental Protection Agency, USEPA Method
3052:"Microwave Assisted Acid
Digestion of Silicious and Organically based Matrices"
United States Department of Agriculture, USDA:
"Soil Survey Laboratory Methods Manual" 1996
Wang, J.,
Huang, C. P., Allen, H. E., Poesponegoro, I., Poesponegoro, H., and Takiyama,
L. R. Effects of Dissolved Organic Matter (DOM) and pH on Heavy Metal Uptake by
Sludge Particulates Exemplified by Cu(II) and Ni(II): Three-Variable Model.
Water Environment Research. (Accepted for Publication).
Zalewski, W., Oprzadek,
K., Syrocka, K., Lipinska, J., and Jaroszynska, J. (1994). Value of harmful
elements in fruit and vegetables grown in the province of Siedlce. Rocz. Panastw.
Zakl. Hig. 45, 19-26.
Table 1 Copper
Concentrations in Crop Tissues
The mean copper concentrations
measured in plant tissues are shown with the corresponding standard deviation.
Sample sizes ranged from 18 to 22.
|
|
Mean Cu Content ± SDM (mg/Kg d.w.) |
|
||
|
|
Edible Tissue |
Stem
|
Leaf |
USDA mean content
of edible part (mg/Kg d.w.) |
|
ONION |
7.7 ± 1.3 |
6.4 ± 2.0 |
8.4 ± 3.0 |
5.8 |
|
TOMATO |
15.2 ± 3.7 |
22.6 ± 5.4 |
45.8 ± 17.2 |
11.9 |
Table 2 Cu Concentrations in Plants and Extracts for Low and High
Copper Soils.
The mean and standard deviations are shown. Samples
sizes ranged between 12 and 15 (tissues) and 9-26 (soil extracts).
|
Crop |
Tissue |
Low Cu Soils |
High Cu Soils |
ANOVA |
||
|
|
|
|
|
|
||
|
Bulb |
7.1 ± 1.2 |
8.2 ± 3.0 |
N.S. |
|||
|
|
|
|
|
|||
|
Stem |
6.4 ± 2.1 |
10.2 ± 2.6 |
p<0.005 |
|||
|
|
|
|
|
|||
|
Leaf |
6.6 ± 2.2 |
10.9 ± 2.1 |
p<0.001 |
|||
|
|
||||||
|
Tomato |
|
|
|
|
||
|
Fruit |
14.7 ± 4.1 |
15.8 ± 3.4 |
N.S. |
|||
|
|
|
|
|
|||
|
Stem |
19.5 ± 4.5 |
26.2 ± 4.2 |
p<0.005 |
|||
|
|
|
|
|
|||
|
Leaf |
35.7 ± 9.3 |
64.4 ± 11.8 |
p<0.001 |
|||
|
|
||||||
|
|
|
|
|
|||
|
|
Copper in extract (mg/Kg soil) |
|
|
|||
|
|
|
|
|
|||
|
Saturation Extract |
0.04 ± 0.016 n=26 |
0.14 ± 0.084 n=20 |
p<0.001 |
|||
|
0.01 M CaCl2 |
0.10 ± 0.067 n=14 |
0.22 ± 0.057 n=9 |
p<0.001 |
|||
|
TCLP |
0.17 ± 0.21 n=14 |
0.94 ± 0.42 n=9 |
p<0.001 |
|||
FIGURAS
Figures
FIGURA
1 --- EX FIGURA 3B CURVA CAMPANA CU VS NUTM
FIGURA
2 --- EX FIGURA 5
DEPTH PROFILES