- Email: [email protected]
Arsenic and lead uptake by Brassicas grown on an old orchard site
Journal of Hazardous Materials 299 (2015) 656–663
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Arsenic and lead uptake by Brassicas grown on an old orchard site Maya P. Lim, Murray B. McBride ∗ Department of Ecology & Evolutionary Biology, Cornell University, Ithaca, NY 14853, USA
h i g h l i g h t s • Arugula and collards accumulated As and Pb when grown in a long-contaminated orchard soil, with arugula exhibiting significantly higher As accumulation than collards.
• Coefficients of transfer from soil into the above-ground vegetable tissues were higher for As than for Pb. • Contamination of arugula by Pb was correlated to soil particle contamination of above-ground tissues. • Amendment of soil with compost reduced plant concentrations of As and Pb.
a r t i c l e
i n f o
Article history: Received 25 March 2015 Received in revised form 20 July 2015 Accepted 31 July 2015 Available online 3 August 2015 Keywords: Arsenic Brassicaceae Soil contamination Lead Plant uptake
a b s t r a c t Arugula (Eruca sativa) and collards (Brassica oleracea var. acephala), were grown at a former orchard where soils had been variably contaminated by lead arsenate pesticides. To test for the effect of compost on As and Pb transfer into plants, compost was added (0, 5, and 10% DW) to five plots representing a wide range of soil Pb and As. Arugula accumulated about 5 times higher As concentrations in above-ground tissues than collards, with high variability in individual plant concentrations. Soil to arugula transfer (uptake) coefficients were higher for As than for Pb, and increased with soil As. Crop concentrations of Pb varied widely within replicate samples of both arugula and collards. Arugula contamination by Pb was significantly correlated to soil total Pb, but collard contamination was not. Evidence was found using Al as an indicator of soil particle contamination of plant tissues that Pb in arugula was primarily due to soil particle deposition on foliar surfaces. Compost amendments reduced 0.01 M CaCl2 -extractable Pb but increased extractable As in the orchard soils. However, compost had the beneficial effect of reducing both As and Pb concentrations in harvested arugula grown on most of the plots. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Lead arsenates were used as insecticides in orchards in North America from the late 1800s until the 1960s [33,38], resulting in significant cumulative contamination of orchard soils with Pb and As. When these orchards are converted from agricultural to residential uses, human health concerns arise because of the possible increased exposure to toxic metals from soil ingestion and consumption of contaminated crops. The scale of this problem is uncertain, but some estimates indicate that millions of acres of orchard lands have been contaminated in North America. The state of Virginia may have 100,000–300,000 acres of old orchard land [38], Washington state is estimated to have 188,000 acres, Wisconsin about 50,000 acres, and New Jersey
∗ Corresponding author at: Section of Soil and Crop Sciences, Cornell University, Ithaca, NY 14853, USA. Fax: +1 697 255 8615. E-mail address: [email protected] (M.B. McBride). http://dx.doi.org/10.1016/j.jhazmat.2015.07.082 0304-3894/© 2015 Elsevier B.V. All rights reserved.
may have as much as 5% of its total agricultural acreage impacted by residual Pb arsenate pesticides [15]. Past research suggested minimal transfer of As and Pb into edible crops grown on historically contaminated orchard soils, especially fruits [2,10,20,22]. However, leafy vegetables and root crops have generally shown a higher degree of As and Pb contamination than fruits, with a wide range of uptake depending on crop species and the part of the plant analyzed [24,26,41]. For example, arugula accumulates substantially higher concentrations of As than collards or spinach, with lettuce being intermediate ([40] McBride, unpublished data]. Concentrations of As in some leafy vegetables grown on severely contaminated orchard soils have been found to exceed World Health Organization limits, although Pb transfer into the same vegetables has been lower [24]. In general, As shows a higher potential than Pb for uptake and translocation from roots to shoots, but this depends strongly on plant species, with some brassicaceous plants having a high As uptake potential [36]. At present, the factors controlling As and Pb transfer into vegetable crops from contaminated
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663
soils are insufficiently understood to allow reliable prediction of crop contamination. Because the total Pb and As concentrations in contaminated soils frequently correlate poorly with Pb and As contents in vegetable crops [16,32,41,45], soil testing for total metals is insufficient to establish crop uptake potential and consumer health risk. This poor correlation may reflect in part the importance of soil properties and chemical form of contamination in controlling long-term As and Pb bioavailability. For example, As in orchard soils originating from arsenate pesticide use appears to be more phytoavailable and phytotoxic than that from mining activity or smelter emissions [45]. Recent studies [3,11] conducted on industrial waste-contaminated sites showed very limited As uptake into root crops and leafy vegetables, even though soil total As concentrations ranged as high as 95–146 mg kg−1 . Since soil As contamination from arsenate pesticides may have a higher level of plant availability than other contaminant sources (e.g., certain brownfield sites), additional research on uptake of As into crops with a known tendency to accumulate this toxic element, specifically leafy vegetables, is needed. Transfer of soil Pb into food crops is difficult to predict because of the fact that for some vegetables, such as leafy greens, most of the crop contamination is due to particle deposition on plant surfaces [1,7,25,26,30,31,35]. Soil particle contamination of vegetables by dust and splash is a random and unpredictable process that may obscure any dependence of crop Pb concentration on local soil Pb concentration. Consequently, in urban vegetable gardens, soil total Pb concentration only explains a small part of the variation of Pb concentrations in leafy vegetables [25]. Soil Pb solubility is strongly affected by soil properties, particularly pH and organic matter content, but these parameters often fail to predict transfer of Pb into edible crops because of the dominance of physical contamination [16]. Proximity to sources of Pb emission is an important factor for vegetable contamination; thus, vegetables grown on Pb-contaminated soils of urban areas typically contain more Pb than market vegetables grown in largely uncontaminated regions [18,25,29,37,41]. Although, Pb contamination patterns in crops point to aerial transport and deposition of particles as being important processes for soil-to-plant transfer of Pb [1,12,30,31,43], contaminating particles may originate from both distant and local sources. In the specific case of orchard soils containing residual Pb arsenate pesticides, there is little known about the relative importance of physical (particle contamination) versus physiological (root-mediated uptake) transfer of Pb into leafy vegetables. In summary, further research is needed to understand and minimize soil-to-plant transfer mechanisms for As and Pb in vegetable crops that tend to accumulate these toxic elements from contaminated orchard soils. This study selected two brassicaceous vegetables, collards and arugula, because they are commonly grown in gardens, are prone to physical contamination of their consumed parts (leaves), and have been found in our preliminary field studies to represent moderate (collards) to high (arugula) potential for As uptake. The study was designed as an assay to measure soil-to-crop transfer coefficients for Pb and As under realistic field conditions over a range of total soil Pb and As, thus providing necessary data for risk assessments to determine human exposure. The study also included a comparison of Pb and As uptake from compost-amended and unamended soils in order to evaluate of the effectiveness of compost in reducing crop transfer coefficients.
2. Methods 2.1. Field experiment with arugula and collards Five locations, labeled A–E, within an old apple orchard on the Dilmun Hill Student Farm at Cornell University (Ithaca, NY),
657
were selected in 2010 to represent a wide range of soil Pb and As (from slightly above background for plot A to more than 800 and 150 mg kg−1 of total Pb and As, respectively, for plots D and E) in order to be used for a series of vegetable trials (see Table 1). The soils are silty clay loams (Hudson series), and were mechanically tilled with a rotavator prior to constructing wooden frames (3.7 × 0.6 m) with equally spaced board dividers in order to separate the control (no compost), 5% compost, and 10% compost treatment subplots. The compost amendments (Cornell compost from food waste and animal bedding) were added at levels calculated to initially provide 5% (1×) and 10% (2×) organic matter to the soils based on dry weight, and thoroughly mixed 15 cm into the soils manually. This compost contained low As (0.97 mg kg−1 ) and Pb (16 mg kg−1 ), with a pH in water of 7.0. The concentrations of major elements in the compost were 4.4% calcium (Ca), 1.6% magnesium (Mg), 1.5% potassium (K), 0.82% phosphorus (P), 0.45% sulfur (S), 0.88% iron (Fe), and 0.60% aluminum (Al). Concentrations of other trace elements were 24 mg kg−1 nickel (Ni), 65 mg kg−1 copper (Cu), 510 mg kg−1 manganese (Mn), 43 mg kg−1 chromium (Cr), and 300 mg kg−1 zinc (Zn). The very low Pb and As concentrations in the compost diluted the pre-existing orchard soil As and Pb concentrations upon amendment with 5% and 10% compost in 2010. Soil analyses done in 2014 (Table 1) revealed that total Pb concentrations in the 5% and 10% compost treatments averaged 11% and 12% lower than the control (0%) in all of the plots, but this reduction was not statistically significant (p > 0.05). Reductions in soil total As averaged 12% and 23% in the 5% and 10% compost treatments, respectively, with only the latter being statistically significant (p < 0.05). Composite soil samples were collected in May 2014 from all plots, oven-dried at 100 ◦ C, pulverized using a mortar and pestle, and passed through a 2 mm sieve in preparation for soil analysis. In May 2014, arugula (Eruca sativa, Burpee Seed Company, Warminster, PA, USA) and collards (Brassica oleracea var. acephala cv. “Georgia”, Burpee Seed Company, Warminster, PA, USA) were seeded directly into the subplots after thorough hand-tillage, with a large number of arugula plants established in both noncomposted and composted subplots, and a similar plant density of collards seeded only in the non-composted subplots. Replicate samples (individual plants of similar size) of the above-ground tissues of both crops were selected at random for harvest in July 2014 from the non-composted subplots (arugula N = 7–8). Arugula aboveground tissues (composites of several plants, N = 2–3) were harvested at the same time from the 5% and 10% composted subplots. The freshly harvested portions were washed thoroughly in the laboratory using tap water, placed in paper bags, and oven dried at 70 ◦ C. The tissue samples were then ground using a stainless steel mill and stored in sealed bags prior to acid digestion and measurement of tissue Pb and As. 2.2. Greenhouse experiment with collards A separate small-scale greenhouse assay was designed to test the hypothesis that low soil pH could allow high Pb uptake by collards under controlled conditions with Pb-spiked soils. This assay was intended to provide some indication of how strongly a brassica could accumulate Pb in above-ground tissue from a soil in which Pb is in a more plant-available form than that present in the orchard soil. Individual two-week-old collard seedlings were transplanted into small pots containing 45 g of a 1:1 peat to perlite mix spiked with 0, 100, and 200 mg/kg Pb using Pb(NO3 )2 dissolved in water. At each level of Pb, the pH was adjusted to 5, 6, and 7 using Ca(OH)2 , and all treatments were replicated (N = 3). Plants were maintained at 22 ◦ C with a 16:8 L:D cycle, and each pot received 10 mL of a complete soluble fertilizer (14 g L−1 solution, Peters Excel 15-5-15Cal Mag Special, Everris NA Inc., Dublin, OH, USA) weekly and addi-
658
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663
Table 1 As and Pb concentrations in soils (0–15 cm) of compost-amended plots and average concentrations (± standard deviations) of As and Pb in above-ground whole plants grown in these plots. Soil organic matter (%)
Soil As (mg kg−1 )
Plot (% compost)
Soil pH
A (0%) A (5%) A (10%)
6.04 5.97 6.05
4.44 6.74 9.03
27.2 15.6 12.4
B (0%) B (5%) B (10%)
5.78 6.57 6.61
4.69 8.09 11.3
58.5 50.7 49.1
C (0%) C (5%) C (10%)
5.37 6.14 6.38
6.76 11.8 14.8
80.1 76.2 63.1
D (0%) D (5%) D (10%)
6.15 6.01 6.39
7.26 11.5 15.9
215 239 174
E (0%) E (5%) E (10%)
5.24 6.08 6.18
8.12 10.7 13.4
191 181 148
Soil Pb (mg kg−1 )
Arugula As (mg kg−1 )
Arugula Pb (mg kg−1 )
2.46 ± 1.18 0.39 ± 0.15* 0.00 ± 0.00*
1.83 ± 1.27 0.95 ± 0.33 0.12 ± 0.08*
169 168 202
6.87 ± 4.81 No data 4.02 ± 1.69
1.32 ± 0.93 No data 0.15 ± 0.21*
350 333 312
8.59 ± 1.90 4.74 ± 4.03 3.28 ± 2.60
6.57 ± 2.58 3.28 ± 2.06* 0.69 ± 0.00*
1075 1880 1225
43.8 ± 23.4 42.5 ± 26.0 8.03 ± 2.32*
12.5 ± 5.61 9.15 ± 1.63 2.30 ± 0.06*
1036 987 853
78.8 ± 26.2 30.7 ± 1.13* 14.8 ± 10.4*
15.3 ± 3.83 7.66 ± 1.14* 5.04 ± 2.18*
67.9 37.4 32.5
Asterisks (*) denote significant differences in arugula As and Pb concentrations (p ≤ 0.05) when comparing the 5 and 10% compost treatments to the 0% compost controls. The missing arugula Pb and As data from Plot B (5% compost) resulted from mislabeling and accidentally discarding these samples.
tional deionized water as needed. After three weeks, above-ground tissues were harvested, ground, and analyzed for Pb and other elements by the method described below for field-grown collard and arugula samples.
soils by Pb:
= 2.3. Analytical methodology Soil pH was measured by electrode in 2014 (using 1:2 parts soil to deionized water) and ranged from 5.24 to 6.15 in the subplots without added compost, and from 5.97 to 6.61 in the compostamended subplots (see Table 1). The slightly alkaline nature and high buffering capacity of the compost made the compost-amended soils generally less acidic, even four years after the amendment. The soil organic matter contents of all soils were measured in November 2012 using loss-on-ignition. This was done by measuring weight loss of 5 g soil weighed into a crucible, pre-dried at 105 ◦ C, after heating the soil in a muffle furnace at 375 ◦ C for 24 h. Soil samples were analyzed for total Pb, As, and other elements by the Cornell Nutrient Analysis Lab using an automated hot-plate nitric/perchloric acid digestion method followed by ICP-OES analysis of the digestates. Duplicate subsamples of soils from the 0%, 5%, and 10% compost treatments were chosen from four of the plots that represent a wide range of soil Pb and As, and then extracted 24 h for readily available Pb and As using 0.01 M CaCl2 as previously described [21]. These extracts, after filtration using Whatman #42 paper, were analyzed for Pb and As by ICP-OES. Detection limits for Pb and As in the 0.01 M CaCl2 solutions were 0.002 and 0.0005 mg L−1 , respectively. Vegetable tissue samples were digested on an automated hotplate using concentrated nitric/perchloric acid and analyzed for Pb, As, and Al using ICP-OES. Aluminum analysis was done because physical contamination of plant tissues by soil particles can be estimated by analyzing the tissue content of elements that are known to be immobile in the soil and not generally taken up through roots (e.g., Al, Ti, and Cr) [6,8,26]. We selected Al as the most sensitive indicator of soil particle contamination because of its very high concentration in mineral soils and greater susceptibility than Ti or Cr to dissolution by the nitric/perchoric acid tissue digestion process [8]. The mineral soils of this site without added compost averaged about 14,000 mg kg−1 Al as measured by acid digestion and analysis by ICP-OES, with compost amendment reducing soil Al to about 13,000 mg kg−1 on average. Therefore, the following equation was used to provide an estimate of the physical contamination of arugula grown in the compost-amended and unamended
Pb contamination index mg kg−1 Plant tissue Al mg kg−1
Measured soil Al mg kg
−1
× soil Pb mg kg−1
The index assumes that Pb-bearing particles deposited on plant surfaces have approximately the same elemental composition as the whole soil. This assumption could fail, for example, if finer soil particles that easily adhere to surfaces have much higher or lower Pb concentrations than larger particles. Procedures used to ensure precision and accuracy in the measurements of Pb and As in the plant tissues and soils by ICP-OES included the use of certified SRMs (NBS 1571-orchard leaves, NIST 2702-marine sediment, NIST 2709-San Joaquin soil) as well as laboratory internal standards in the sample sets, along with duplicate samples and blanks in each sample set. Elemental recoveries for Pb and As were 97% and 79%, respectively, from SRM 2702, and 76% and 77%, respectively, from SRM 2709. Recoveries for Pb and As from the orchard leaf SRM (1571) were 84% and 80%, respectively. 2.4. Statistical analysis Welch’s t-test (used because of data heteroskedasticity) for unpaired data was employed for comparisons of treatment effects with the p-value for significance set at 0.05. The statistical significance (p ≤ 0.05) of correlation coefficients (r) of linear regression were determined using Fisher’s two-tailed significance test. 3. Results and discussion 3.1. As–Pb relationships in the soil The wide range of total Pb (and associated As) as well as the variation of the As/Pb ratio in the soils of the orchard field site used in this study is illustrated in Fig. 1. The soil total As and Pb concentrations were highly correlated at this orchard site [13], a fact attributable to the same source of contamination for both elements (i.e., lead arsenate used historically as a pesticide). Previous research showed that there has likely been some long-term loss of As from the topsoil (relative to Pb) because the As/Pb ratio in the orchard soils is substantially lower than that expected for the specific Pb arsenate pesticide applied at the site [13]. Further analysis
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663
659
that the high compost addition in 2010 mobilized and facilitated leaching of a small fraction of the total soil As. Numerous studies have observed that increasing organic amendments and dissolved organic matter can have this effect in soils [5,42,44]. The 0.01 M CaCl2 -extractable As and Pb concentrations were generally low, but but tended to increase with increasing soil total As and Pb (Fig. 2). The results indicate that plant-available As was approximately linearly proportion to total soil As concentration, whereas available Pb increased substantially only when total soil Pb exceeded approximately 400 mg kg−1 . Compost amendment increased available As at all levels of soil total As (see Fig. 2a), consistent with the commonly observed phenomenon that compost amendments increase the fraction of soil As in soluble form [5,14,42]. Compost had the opposite effect on Pb, reducing its availability as measured by 0.01 M CaCl2 extraction; however, this effect was only detectable at the highest soil Pb concentrations. Previous studies have similarly shown that natural organic materials such as composts can reduce both solubility and bioavailability of Pb [19,44,48]. 3.2. As and Pb transfer into field-grown arugula and collards
Fig. 1. Relationship of soil total As:Pb ratio to soil total Pb. Unshaded squares with a dashed best-fit line indicate As:Pb ratios at 0% compost addition, partially shaded squares with a dotted best-fit line indicate As:Pb ratios at 5% compost addition, and fully shaded squares with a solid best-fit line indicate As:Pb ratios at 10% compost addition.
of these soils in the present study revealed that the ratio of As to Pb in the topsoil varies substantially as a function of total Pb as shown by the slopes of the best-fit lines in Fig. 1. This trend could be explained by greater historical As loss (relative to Pb) from those topsoils most heavily contaminated with the pesticide. Since As is known to migrate to some degree from the topsoil by leaching [34], this result implies that As solubility is disproportionately higher in the more severely contaminated soils of this site. Compost added at the highest (10%) level may have slightly reduced the soil As:Pb ratio as indicated by a small but statistically insignificant shift of the best-fit line for the 10% compost-amended soil in Fig. 1 compared to the 5% compost-amended and unamended soils. It is possible
Arugula was found to be a strong accumulator of As in the field trials, as shown by the high average plant concentrations of As (44–79 mg kg−1 ) when grown in the uncomposted high-As plots (see data for plots D and E in Table 1). Arugula had consistently higher (approximately five-fold) As content than collards at the same soil contamination level (Fig. 3a and c). This result concurs with the observations of other researchers, where arugula was found to be a stronger accumulator than other leafy green vegetables [40]. Arugula also averaged relatively high Pb concentrations (12.5–15.3 mg kg−1 ) when grown on the unamended (0% compost) high Pb (D and E) plots. The As concentrations in arugula and collards tended to increase with soil total concentrations of these toxic metals as shown by the best-fit relationships between plant tissue and soil As in Fig. 3a and c. Furthermore, the As soil-plant transfer coefficients (defined as the slopes of the trend lines in Fig. 3a and c) were close to 0.10 for the least As-contaminated plots, but were higher for the severely As-contaminated plots (0.20 and 0.41 for unamended plots D and E, respectively), suggesting that the bioavailable fraction of soil As was higher in these soils. Conversely,
Fig. 2. Relationship between soil concentrations and extractable concentrations of As (a) and Pb (b) using calcium chloride (CaCl2 ) for 0%, 5% and 10% compost addition to the soil. The error bars denote ± one standard deviation unit.
660
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663
Fig. 3. Relationships of arugula As and Pb concentration (a and b, respectively) and collard As and Pb concentrations (c and d, respectively) to soil total As and Pb. Best-fit lines and equations are shown in each case.
the Pb soil-plant transfer coefficients were much lower than those for As regardless of soil Pb level, and averaged 0.016 ± 0.007 for all of the unamended plots. Compost amendment of the soils caused both the Pb and As transfer coefficients to be reduced. The relationships between crop Pb and soil Pb concentration, shown in Fig. 3b and d, revealed a more random pattern than for As. Arugula generally averaged higher Pb when grown in soils with greater Pb levels, but collards showed no significant relationship of plant Pb to soil Pb concentration. Overall, there was a high degree of apparently random variability in arugula and collard concentrations of As and Pb within replicate samples (see Table 1, Fig. 3). Weak or non-existent relationships of vegetable Pb concentrations to soil total Pb concentrations have been observed in numerous other field studies [12,16,29,37] and point to the likelihood that, at contaminated sites, inherently random processes such as particulate deposition may have a dominant role in crop contamination by Pb [7,25]. The compost treatments reduced average As and Pb concentrations in arugula grown in all plots (Table 1). The 10% compost
amendment generally had a larger effect than the 5% amendment in reducing crop As and Pb, but not all reductions were statistically significant. All plots showed statistically significant (p ≤ 0.05) reductions in crop Pb at the 10% compost level, but only half showed significant reductions at the 5% compost.. For crop As, 3 of 5 plots showed significant reductions at the 10% compost level, and half showed significant reductions at the 5% level. The reduced toxic metal concentrations in arugula caused by compost amendment may be tentatively attributed to one or several factors, including physical dilution of As and Pb concentrations in the soils by compost amendments, and chemical stabilization of soil As and/or Pb by interaction of these metals with the added organic matter. Both effects have been observed in other field studies that used sewage sludge compost amendments in brownfield remediation, with the evidence for Pb stabilization by adsorption being more convincing than that for As [3,11]. However, other factors may also reduce measured Pb and As concentrations in crops grown on heavily composted soils, including the well-known “growth dilution” effect on trace metal concentration in plants resulting
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663
from more vigorous growth and biomass production compared to control treatments [17]. In the present study, the compost amendments markedly increased plant growth and biomass, although yields were not measured. In addition, reduced contamination of the above-ground tissues by soil particles from dust and rain splash may have occurred due to the observed effect of compost on improving soil aggregation and structure and increasing canopy cover of soil. Of these factors, the soil physical dilution effect is unlikely to be very important, since dilution by compost reduced soil total Pb concentration by only about 11–12% on average. The chemical stabilization factor appears to be important for Pb but not for As based on the results of the 0.01 M CaCl2 extractions, which indicated that compost decreased available Pb but increased available As. Nevertheless, plant As concentrations were lower in arugula grown in compost-amended soils compared to unamended soils, so that factors other than soil As availability must have limited plant tissue concentrations of As. For example, uptake of arsenate from the compost-amended soil could have been suppressed by competition for uptake with phosphate (or other anionic nutrients) present in the compost [28]. Published research studies of the effect of compost on crop uptake of As have shown inconsistent results, with not all composts increasing As solubility and higher soluble or extractable As not necessarily resulting in greater As in the crop [3,4,5,9,11,14]. The observed beneficial effect of compost in limiting arugula contamination by Pb and As may therefore be caused by multiple factors. A likely explanation for the highly variable Pb contamination of leafy crops is physical contamination of the plant surfaces in the field by soil particles arising from dust and soil splash [7,25]. Dust and aerosol deposition from the air, rather than root uptake and translocation into plant tops are considered to be the source of most of the Pb in plants [27]. The harvested arugula and collards had been washed several times with water before tissue analysis, but previous studies have shown that various methods of washing and cleaning are only moderately effective in reducing vegetable Pb contamination [6,11]. A previous greenhouse study of lettuce (Lactuca sativa) uptake of Pb and As from soils collected from the same orchard site did not show the extent of Pb contamination (lettuce Pb never exceeded 1 mg kg−1 ) observed in the present study [24], suggesting that field conditions are more conducive to particle contamination of crops than greenhouse conditions. Because of the likelihood of particle contamination of plant surfaces in the field experiment, arugula tissue concentrations of Al were measured and used to estimate soil particle contamination of the crop as described in the methods. This is a reasonable approach because the soils are not strongly acidic (soil acidity increases Al uptake in many plant species) and almost all cultivated crops, including brassicaceous species, are Al excluders [23]. Previous studies, using Al and Cr as indicators of contamination, revealed washed vegetables to contain from 0.07 to 4.88% soil by dry weight [6]. The Pb contamination index, calculated from measured total Al in the vegetables as described in the methods section, was strongly correlated to the measured Pb concentrations in arugula for the non-composted and composted plots (R2 = 0.777 and 0.808 for the control and compost-amended plots, respectively) as shown in Fig. 4. The ability of the contamination index to closely estimate the actual tissue Pb concentration implies that soil particle contamination of arugula is the main process of Pb transfer into this crop. The data in Fig. 4 also reveal that Pb contamination tended to be lower for arugula grown on the composted compared to the uncomposted plots, as also indicated by the effect of compost amendment in reducing the average Pb concentration in arugula (Table 1). The generally lower contamination indices for arugula grown on the composted plots indicate generally lower soil particle contamination (i.e., lower plant tissue Al) of the compost-grown crop (Fig. 4). Overall, the Pb contamination index, which takes into
661
Fig. 4. Relationship of predicted arugula Pb concentration (based on the contamination index) to the measured Pb concentration in crops from the control (open squares) and compost-amended (shaded squares) field plots.
account physical contamination of above-ground tissues by soil particles, explains the highly variable arugula Pb concentrations better than soil Pb concentration alone. Collards grown on the same field plots showed a very weak and barely significant (R2 = 0.075, p = 0.05) relationship of this Pb contamination index to actual collard crop Pb (data not shown). The average Pb contamination index was much lower for collards (mean = 0.79 ± 0.97) than for arugula (mean = 5.74 ± 5.54) because of lower average collard tissue Al. Thus, the highest observed collard Pb levels (data “outliers”) seen in Fig. 3d) could not be explained by the index. Nevertheless, the lower average Al content of collards compared to arugula can be attributed to the smooth large leaves of collards, which are more easily washed to remove surface contaminants than the trichome-studded, convoluted leaves of arugula. The high Pb levels in some collards are therefore unlikely to be due primarily to soil particle contamination. The data “outliers” originated in plots without added compost, which had somewhat lower pH than the composted plots, and may therefore have had higher Pb bioavailability. Poor plant vigor in the uncomposted plots may also have contributed a “growth concentration” effect for Pb, a phenomenon reported in other studies of trace metal uptake where plants are stunted by growing conditions [17,39]. Greenhouse studies using soils artificially spiked with soluble Pb salts have shown that very high Pb concentrations in the above-ground tissues of brassicaceous plants are possible [46,47], although the lability of soil Pb in experiments with artificially Pb-spiked soils was probably much greater than was the case for the historically contaminated orchard soils used here. There was a significant positive correlation between collard Pb concentration and that of other elements such as S (R2 = 0.257, p < 0.01), with the highest Pb tissues having quite elevated S concentrations. This may be related to poor plant growth and stunting, which can be associated with abnormal concentrations of nutrients in tissues [17], and could possibly have contributed to the few unusually high tissue Pb concentrations in collards. Unlike Pb, As in arugula grown on the non-composted plots was not significantly correlated to the As contamination index, calculated analogously to the Pb index (data not shown). This result is not surprising as As has a much higher soil solution concentration in this orchard soil than Pb ([26]), and most As transfer into the leafy vegetables is presumeably by root uptake from soil solution. The
662
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663
Fig. 5. Above-ground concentrations of Pb in greenhouse-grown collards at three levels of Pb added to soil (0, 100, 200 mg kg−1 ) and soil adjusted to pH 5 (gray bars), 6 (black bars), and 7 (diagonally hatched bars).
much higher As/Pb concentration ratio in the arugula plants (averaging 3.1 over all plots) compared to that in the soil (about 0.28 averaged over all plots) provides further evidence that soil particle contamination did not have an important role in As transfer into arugula. 3.3. Pb transfer into greenhouse-grown collards Spiking the greenhouse soil with 100 and 200 mg kg−1 of soluble Pb and adjusting the pH over the range of 5–7 caused no measurable difference in collard growth compared to the control as indicated by measured dry matter yields. Plant growth was vigorous for all treatments with no indication of Pb phytotoxicity. Pb concentrations in the harvested collards increased significantly for both the 100 and 200 mg kg−1 levels of soil Pb (p < 0.0001), although uptake of Pb was not significantly affected by soil pH (Fig. 5). Pb uptake coefficients were calculated to be 0.018, 0.014, and 0.013 at pH 5, 6, and 7, respectively. These are much lower Pb uptake coefficients than those reported for Chinese cabbage (Brassica pekinensis) grown on Pb-spiked soils [46,47], but collards have been shown to accumulate significant Pb in the edible tissues from freshly Pb-spiked soils [46]. Much higher Pb concentrations were occasionally measured in field-grown collards than observed in this greenhouse study. These extreme Pb values are not readily explained, but might have resulted from several field-related factors such as longer growth period and restricted growth due to abiotic (e.g., fluctuating temperature, poor soil structure, and soil moisture extremes) and biotic (e.g., herbivory and exposure to pathogens) stresses. One suggestion of plant stress in the field is the higher sulfur concentrations measured in the field-grown (9570 ± 4540 mg kg−1 ) compared to the greenhouse-grown collards (6030 ± 2540 mg kg−1 ). Plant tissue Pb and S concentrations were statistically correlated (R2 = 0.257, N = 54, p < 0.01) in field-grown collards, with S in the most Pbcontaminated collards exceeding 10,000 mg kg−1 . This very high S level could be due to poor growth of plants, which allows nutrients to reach high concentrations in tissues [17,39]. 4. Conclusions Arugula had a stronger tendency than collards to accumulate As in above-ground tissues when grown in orchard soils contaminated
historically by Pb arsenate pesticides. Arsenic uptake coefficients for arugula (ratio of plant to soil As concentration) generally were much higher than those for Pb and increased at higher soil total As concentrations. Although arugula is not an As hyperaccumulator, it does appear to be more efficient than most leafy vegetables in accumulating As in above-ground tissues, and may be a useful indicator species for assessing bioavailability of As at contaminated sites. In contrast to As, Pb transfer into arugula seemed primarily due to physical contamination of the plants by soil particles as indicated by a relationship of Pb to Al measured in the above-ground tissues. Several unusually high Pb levels measured in collards could not be explained by soil particle contamination of the tissues and may have been related to stunted growth. Compost amendments to the contaminated orchard soils substantially reduced the As and Pb concentrations in the harvested crop in most cases, but this may have been predominantly from metal dilution in the plant tissues by more vigorous growth rather than an overall reduction of metal bioavailability by adsorption on organic matter. The compost amendment actually increased As extractability in the soil even as it decreased extractable soil Pb. There are no health guidance values established in the USA for As and Pb concentrations in vegetables, but values suggested by other studies are 0.3 mg kg−1 Pb (fresh wt. basis) based on EU standards [25] and 0.1 mg kg−1 As (fresh wt. basis) based on a conservative risk assessment [11]. For arugula and collards, with measured moisture contents of 88–89% in the present study, these guidance values translate on a dry weight basis to about 2.7 mg kg−1 Pb and 0.9 mg kg−1 As. Inspection of the average Pb concentrations in arugula grown on the orchard plots (Table 1) shows that only those plots with higher than 300 mg/kg total Pb produced crops exceeding the guidance value, although compost amendment did lower crop Pb to acceptable levels in some of the high-Pb soils. On the other hand, As in arugula exceeded the guidance value by a large margin for all plots, and even the plot with lowest soil As (<30 mg kg−1 ), unless it had been amended with compost, produced arugula with As exceeding the guidance value. Collards accumulated lower As concentrations than arugula when grown in the same plots. Nevertheless, all collards grown on unamended plots with >150 mg kg−1 total As exceeded the As guidance value, more than half of collards grown on unamended soils with 50–100 mg kg−1 total As exceeded the guidance value, and only the least contaminated plot (<30 mg kg−1 ) produced collards below the guidance value. The Pb guidance value was exceeded in about in about 60% of collards grown on all the unamended plots (N = 54). Acknowledgements This research was funded by Federal Hatch project NYC-125445, the Toward Sustainability Foundation research grant, and the Atkinson Sustainable Biodiversity Fund. We thank Betsy Leonard for allowing this research to occur at the Dilmun Hill Student Organic Farm and Alice Jenkins for laboratory assistance. References [1] V. Angelova, R. Ivanova, K. Ivanov, Heavy metal accumulation and distribution in oil crops, Commun. Soil Sci. Plant Anal. 35 (2005) 2551–2566. [2] C.F. Aten, J.B. Bourke, J.H. Martini, J.C. Walton, Arsenic and lead in an orchard environment, Bull. Environ. Contam. Toxicol. 24 (1980) 108–115. [3] C.P. Attanayake, G.M. Hettiarachchi, S. Martin, G.M. Pierzynski, Potential bioavailability of lead, arsenic, and polycyclic aromatic hydrocarbons in compost-amended urban soils, J. Environ. Qual. 44 (2015) 930–944. [4] X. Cao, L.Q. Ma, Effects of compost and phosphate on plant arsenic accumulation from soils near pressure-treated wood, Environ. Pollut. 132 (2004) 435–442. [5] X. Cao, L.Q. Ma, A. Shiralipour, Effects of compost and phosphate amendments on arsenic mobility in soils and arsenic uptake by the hyperaccumulator Pteris vittata L, Environ. Pollut. 126 (2003) 157–167.
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663 [6] E.E. Cary, D.L. Grunes, S.L. Dallyn, G.A. Pearson, N.H. Peck, R.S. Hulme, Plant Fe, Al and Cr concentrations in vegetables as influenced by soil inclusion, J. Food Qual. 17 (1994) 467–476. [7] A.C. Chamberlain, Fallout of lead and uptake by crops, Atmos. Environ. 17 (1983) 693–706. [8] J.H. Cherney, D.L. Robinson, L.C. Kappel, F.G. Hembry, R.H. Ingraham, Soil contamination and elemental concentrations of forages in relation to grass tetany, Agron. J. 75 (1983) 447–451. [9] R. Clemente, D.J. Walker, M. Pilar Bernal, Uptake of heavy metals and As by Brassica juncea grown in a contaminated soil in Aznalcollar (Spain): the effect of soil amendments, Environ. Pollut. 138 (2005) 46–58. [10] T.L. Creger, F.J. Peryea, Lead and arsenic in two apricot cultivars and in Gala apples grown on lead arsenate-contaminated soils, Hortic. Sci. 27 (1992) 1277–1278. [11] P.P. Defoe, G.M. Hettiarachchi, C. Benedict, S. Martin, Safety of gardening on lead- and arsenic-contaminated urban brownfields, J. Environ. Qual. 43 (2014) 2064–2078. [12] F. Douay, H. Roussel, C. Pruvot, A. Loriette, H. Fourrier, Assessment of a remediation technique using the replacement of contaminated soils in kitchen gardens nearby a former lead smelter in Northern France, Sci. Total Environ. 401 (2008) 29–38. [13] M. Fleming, Y. Tai, P. Zhuang, M.B. McBride, Extractability and bioavailability of Pb and As in historically contaminated orchard soil: effects of compost amendments, Environ. Pollut. 177 (2013) 90–97. [14] W. Hartley, N.M. Dickinson, P. Riby, N.W. Lepp, Arsenic mobility in brownfield soils amended with green waste compost or biochar and planted with Miscanthus, Environ. Pollut. 157 (2009) 2654–2662. [15] E. Hood, The apple bites back: claiming old orchards for residential development, Environ. Health Perspect. 114 (2006) A471–A476. [16] R.L. Hough, N. Breward, S.D. Young, N.M.J. Crout, A.M. Tye, A.M. Moir, et al., Assessing potential risk of heavy metal exposure from consumption of home-produced vegetables by urban populations, Environ. Health Perspect. 112 (2004) 215–221. [17] W.M. Jarrell, R.B. Beverly, The dilution effect in plant nutrition studies, Adv. Agron. 34 (1981) 197–224. [18] L. Jorhem, J. Engman, L. Lindeström, T. Schröder, Applications in food quality and environmental contamination uptake of lead by vegetables grown in contaminated soil, Commun. Soil Sci. Plant Anal. 31 (2000) 2403–2411. [19] N. Karami, R. Clemente, E. Moreno-Jimenez, N.W. Lepp, L. Beesley, Efficiency of green waste compost and biochar soil amendments for reducing lead and copper mobility and uptake to ryegrass, J. Hazard. Mater. 191 (2011) 41–48. [20] D.J. Kenyon, D.C. Elfving, I.S. Pakkala, C.A. Bache, D.J. Lisk, Residues of lead and arsenic in crops cultured on old orchard soils, Bull. Environ. Contam. Toxicol. 22 (1979) 221–223. [21] B. Kim, M.B. McBride, Phytotoxic effects of Cu and Zn on soybeans grown in field-aged soils: their additive and interactive actions, J. Environ. Qual. 38 (2009) 2253–2259. [22] K.S. MacLean, W.M. Langille, Arsenic in orchard and potato soils and plant tissue, Plant Soil 61 (1981) 413–418. [23] H. Marschner, Mineral Nutrition of Higher Plants, 2nd ed., Academic Press, San Diego, 1995. [24] M.B. McBride, Arsenic and lead uptake by vegetable crops grown on historically contaminated orchard soils, Appl. Environ. Soil Sci. 2013 (2013) 1–8. [25] M.B. McBride, H.A. Shayler, H.M. Spliethoff, R.G. Mitchell, L.G. Marquez-Bravo, G.S. Ferenz, et al., Concentrations of lead, cadmium and barium in urban garden-grown vegetables: the impact of soil variables, Environ. Pollut. 194 (2014) 254–261. [26] M.B. McBride, T. Simon, G. Tam, S. Wharton, Lead and arsenic uptake by leafy vegetables grown on contaminated soils: effects of mineral and organic amendments, Water Air Soil Pollut. 224 (1) (2013) 1–10. [27] M.J. McLaughlin, D.R. Parker, J.M. Clarke, Metals and micronutrients—food safety issues, Field Crops Res. 60 (1999) 143–163.
663
[28] A.A. Meharg, J. Naylor, M.R. Macnair, Phosphorus nutrition of arsenate-tolerant and nontolerant phenotypes of velvetgrass, J. Environ. Qual. 23 (1994) 234–238. [29] A.M. Moir, I. Thornton, Lead and cadmium in urban allotment and garden soils and vegetables in the United Kingdom, Environ. Geochem. Health 11 (1989) 113–119. [30] H. Mosbaek, J.C. Tjell, M.F. Hovmand, Atmospheric lead input to agricultural crops in Denmark, Chemosphere 19 (1989) 1787–1799. [31] G. Nabulo, H. Oryem-Origa, M. Diamond, Assessment of lead, cadmium, and zinc contamination of roadside soils, surface films, and vegetables in Kampala city, Uganda, Environ. Res. 101 (2006) 42–52. [32] P. Nathanail, C. McCaffrey, R. Ogden, N. Foster, A. Gillett, D. Haynes, Uptake of arsenic by vegetables for human consumption: a study of Wellingborough allotment plots, Land Contam. Reclam. 12 (2004) 219–238. [33] F.J. Peryea, Historical use of lead arsenate insecticides, resulting soil contamination and implications for soil remediation, In: 16th World Congr. Soil Sci. Montpellier, Fr. (1998) 20–26. [34] F.J. Peryea, T.L. Creger, Vertical distribution of lead and arsenic in soils contaminated with lead arsenate pesticide residues, Water Air Soil Pollut. 78 (1994) 297–306. [35] L.R. Prasad, B. Nazareth, Contamination of allotment soil with lead: managing potential risks to health, J. Public Health (Bangkok) 22 (2000) 525–530. [36] A. Raab, P.N. Williams, A. Meharg, J. Feldmann, Uptake and translocation of inorganic and methylated arsenic species by plants, Environ. Chem. 4 (2007) 197. [37] L. Samsøe-Petersen, E.H. Larsen, P.B. Larsen, P. Bruun, Uptake of trace elements and PAHs by fruit and vegetables from contaminated soils, Environ. Sci. Technol. 36 (2002) 3057–3063. [38] T. Schooley, M. Weaver, D. Mullins, M. Eick, The history of lead arsenate use in apple production: comparison of its impact in Virginia with other states, J. Pestic. Saf. Educ. 10 (2009) 22–53. [39] T. Shinano, T. Watanabe, Q. Chu, M. Osaki, D. Kobayashi, T. Okouchi, et al., Varietal difference in radiocesium uptake and transfer from radiocesium deposited soils in the genus Amaranthus, Soil Sci. Plant Nutr. 60 (2014) 809–817. [40] D.E. Stilwell, C.L. Musante, B.L. Sawhney, Copper, chromium and arsenic in soil and plants near coated and uncoated CCA wood, In: Proc. Annu. Int. Conf. Soils, Sediments Water Energy (2010) 10. [41] D.E. Stilwell, T.M. Rathier, C.L. Musante, Comparison of heavy metals in community garden produce versus store-bought produce, Conn. Agric. Exp. Station Bull. 1020 (2008) 1–9. [42] M. Udovic, M.B. McBride, Influence of compost addition on lead and arsenic bioavailability in reclaimed orchard soil assessed using Porcellio scaber bioaccumulation test, J. Hazard. Mater. 205–206 (2012) 144–149. [43] D. Voutsa, A. Grimanis, C. Samara, Trace elements in vegetables grown in an industrial area in relation to soil and air particulate matter, Environ. Pollut. 94 (1996) 325–335. [44] S. Wang, C.N. Mulligan, Effect of natural organic matter on arsenic release from soils and sediments into groundwater, Environ. Geochem. Health 28 (2006) 197–214. [45] G.P. Warren, B.J. Alloway, N.W. Lepp, B. Singh, F.J.M. Bochereau, C. Penny, Field trials to assess the uptake of arsenic by vegetables from contaminated soils and soil remediation with iron oxides, Sci. Total Environ. 311 (2003) 19–33. [46] Z. Xiong, F. Zhao, M. Li, Lead toxicity in Brassica pekinensis Rupr.: effect on nitrate assimilation and growth, Environ. Toxicol. 21 (2006) 147–153. [47] Z.-T. Xiong, Lead uptake and effects on seed germination and plant growth in a Pb hyperaccumulator Brassica pekinensis Rupr, Bull. Environ. Contam. Toxicol. 60 (1998) 285–291. [48] Y.-F. Zhou, R.J. Haynes, R. Naidu, Use of inorganic and organic wastes for in situ immobilization of Pb and Zn in a contaminated alkaline soil, Environ. Sci. Pollut. Res. 19 (2012) 1260–1270.
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Arsenic and lead uptake by Brassicas grown on an old orchard site Maya P. Lim, Murray B. McBride ∗ Department of Ecology & Evolutionary Biology, Cornell University, Ithaca, NY 14853, USA
h i g h l i g h t s • Arugula and collards accumulated As and Pb when grown in a long-contaminated orchard soil, with arugula exhibiting significantly higher As accumulation than collards.
• Coefficients of transfer from soil into the above-ground vegetable tissues were higher for As than for Pb. • Contamination of arugula by Pb was correlated to soil particle contamination of above-ground tissues. • Amendment of soil with compost reduced plant concentrations of As and Pb.
a r t i c l e
i n f o
Article history: Received 25 March 2015 Received in revised form 20 July 2015 Accepted 31 July 2015 Available online 3 August 2015 Keywords: Arsenic Brassicaceae Soil contamination Lead Plant uptake
a b s t r a c t Arugula (Eruca sativa) and collards (Brassica oleracea var. acephala), were grown at a former orchard where soils had been variably contaminated by lead arsenate pesticides. To test for the effect of compost on As and Pb transfer into plants, compost was added (0, 5, and 10% DW) to five plots representing a wide range of soil Pb and As. Arugula accumulated about 5 times higher As concentrations in above-ground tissues than collards, with high variability in individual plant concentrations. Soil to arugula transfer (uptake) coefficients were higher for As than for Pb, and increased with soil As. Crop concentrations of Pb varied widely within replicate samples of both arugula and collards. Arugula contamination by Pb was significantly correlated to soil total Pb, but collard contamination was not. Evidence was found using Al as an indicator of soil particle contamination of plant tissues that Pb in arugula was primarily due to soil particle deposition on foliar surfaces. Compost amendments reduced 0.01 M CaCl2 -extractable Pb but increased extractable As in the orchard soils. However, compost had the beneficial effect of reducing both As and Pb concentrations in harvested arugula grown on most of the plots. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Lead arsenates were used as insecticides in orchards in North America from the late 1800s until the 1960s [33,38], resulting in significant cumulative contamination of orchard soils with Pb and As. When these orchards are converted from agricultural to residential uses, human health concerns arise because of the possible increased exposure to toxic metals from soil ingestion and consumption of contaminated crops. The scale of this problem is uncertain, but some estimates indicate that millions of acres of orchard lands have been contaminated in North America. The state of Virginia may have 100,000–300,000 acres of old orchard land [38], Washington state is estimated to have 188,000 acres, Wisconsin about 50,000 acres, and New Jersey
∗ Corresponding author at: Section of Soil and Crop Sciences, Cornell University, Ithaca, NY 14853, USA. Fax: +1 697 255 8615. E-mail address: [email protected] (M.B. McBride). http://dx.doi.org/10.1016/j.jhazmat.2015.07.082 0304-3894/© 2015 Elsevier B.V. All rights reserved.
may have as much as 5% of its total agricultural acreage impacted by residual Pb arsenate pesticides [15]. Past research suggested minimal transfer of As and Pb into edible crops grown on historically contaminated orchard soils, especially fruits [2,10,20,22]. However, leafy vegetables and root crops have generally shown a higher degree of As and Pb contamination than fruits, with a wide range of uptake depending on crop species and the part of the plant analyzed [24,26,41]. For example, arugula accumulates substantially higher concentrations of As than collards or spinach, with lettuce being intermediate ([40] McBride, unpublished data]. Concentrations of As in some leafy vegetables grown on severely contaminated orchard soils have been found to exceed World Health Organization limits, although Pb transfer into the same vegetables has been lower [24]. In general, As shows a higher potential than Pb for uptake and translocation from roots to shoots, but this depends strongly on plant species, with some brassicaceous plants having a high As uptake potential [36]. At present, the factors controlling As and Pb transfer into vegetable crops from contaminated
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663
soils are insufficiently understood to allow reliable prediction of crop contamination. Because the total Pb and As concentrations in contaminated soils frequently correlate poorly with Pb and As contents in vegetable crops [16,32,41,45], soil testing for total metals is insufficient to establish crop uptake potential and consumer health risk. This poor correlation may reflect in part the importance of soil properties and chemical form of contamination in controlling long-term As and Pb bioavailability. For example, As in orchard soils originating from arsenate pesticide use appears to be more phytoavailable and phytotoxic than that from mining activity or smelter emissions [45]. Recent studies [3,11] conducted on industrial waste-contaminated sites showed very limited As uptake into root crops and leafy vegetables, even though soil total As concentrations ranged as high as 95–146 mg kg−1 . Since soil As contamination from arsenate pesticides may have a higher level of plant availability than other contaminant sources (e.g., certain brownfield sites), additional research on uptake of As into crops with a known tendency to accumulate this toxic element, specifically leafy vegetables, is needed. Transfer of soil Pb into food crops is difficult to predict because of the fact that for some vegetables, such as leafy greens, most of the crop contamination is due to particle deposition on plant surfaces [1,7,25,26,30,31,35]. Soil particle contamination of vegetables by dust and splash is a random and unpredictable process that may obscure any dependence of crop Pb concentration on local soil Pb concentration. Consequently, in urban vegetable gardens, soil total Pb concentration only explains a small part of the variation of Pb concentrations in leafy vegetables [25]. Soil Pb solubility is strongly affected by soil properties, particularly pH and organic matter content, but these parameters often fail to predict transfer of Pb into edible crops because of the dominance of physical contamination [16]. Proximity to sources of Pb emission is an important factor for vegetable contamination; thus, vegetables grown on Pb-contaminated soils of urban areas typically contain more Pb than market vegetables grown in largely uncontaminated regions [18,25,29,37,41]. Although, Pb contamination patterns in crops point to aerial transport and deposition of particles as being important processes for soil-to-plant transfer of Pb [1,12,30,31,43], contaminating particles may originate from both distant and local sources. In the specific case of orchard soils containing residual Pb arsenate pesticides, there is little known about the relative importance of physical (particle contamination) versus physiological (root-mediated uptake) transfer of Pb into leafy vegetables. In summary, further research is needed to understand and minimize soil-to-plant transfer mechanisms for As and Pb in vegetable crops that tend to accumulate these toxic elements from contaminated orchard soils. This study selected two brassicaceous vegetables, collards and arugula, because they are commonly grown in gardens, are prone to physical contamination of their consumed parts (leaves), and have been found in our preliminary field studies to represent moderate (collards) to high (arugula) potential for As uptake. The study was designed as an assay to measure soil-to-crop transfer coefficients for Pb and As under realistic field conditions over a range of total soil Pb and As, thus providing necessary data for risk assessments to determine human exposure. The study also included a comparison of Pb and As uptake from compost-amended and unamended soils in order to evaluate of the effectiveness of compost in reducing crop transfer coefficients.
2. Methods 2.1. Field experiment with arugula and collards Five locations, labeled A–E, within an old apple orchard on the Dilmun Hill Student Farm at Cornell University (Ithaca, NY),
657
were selected in 2010 to represent a wide range of soil Pb and As (from slightly above background for plot A to more than 800 and 150 mg kg−1 of total Pb and As, respectively, for plots D and E) in order to be used for a series of vegetable trials (see Table 1). The soils are silty clay loams (Hudson series), and were mechanically tilled with a rotavator prior to constructing wooden frames (3.7 × 0.6 m) with equally spaced board dividers in order to separate the control (no compost), 5% compost, and 10% compost treatment subplots. The compost amendments (Cornell compost from food waste and animal bedding) were added at levels calculated to initially provide 5% (1×) and 10% (2×) organic matter to the soils based on dry weight, and thoroughly mixed 15 cm into the soils manually. This compost contained low As (0.97 mg kg−1 ) and Pb (16 mg kg−1 ), with a pH in water of 7.0. The concentrations of major elements in the compost were 4.4% calcium (Ca), 1.6% magnesium (Mg), 1.5% potassium (K), 0.82% phosphorus (P), 0.45% sulfur (S), 0.88% iron (Fe), and 0.60% aluminum (Al). Concentrations of other trace elements were 24 mg kg−1 nickel (Ni), 65 mg kg−1 copper (Cu), 510 mg kg−1 manganese (Mn), 43 mg kg−1 chromium (Cr), and 300 mg kg−1 zinc (Zn). The very low Pb and As concentrations in the compost diluted the pre-existing orchard soil As and Pb concentrations upon amendment with 5% and 10% compost in 2010. Soil analyses done in 2014 (Table 1) revealed that total Pb concentrations in the 5% and 10% compost treatments averaged 11% and 12% lower than the control (0%) in all of the plots, but this reduction was not statistically significant (p > 0.05). Reductions in soil total As averaged 12% and 23% in the 5% and 10% compost treatments, respectively, with only the latter being statistically significant (p < 0.05). Composite soil samples were collected in May 2014 from all plots, oven-dried at 100 ◦ C, pulverized using a mortar and pestle, and passed through a 2 mm sieve in preparation for soil analysis. In May 2014, arugula (Eruca sativa, Burpee Seed Company, Warminster, PA, USA) and collards (Brassica oleracea var. acephala cv. “Georgia”, Burpee Seed Company, Warminster, PA, USA) were seeded directly into the subplots after thorough hand-tillage, with a large number of arugula plants established in both noncomposted and composted subplots, and a similar plant density of collards seeded only in the non-composted subplots. Replicate samples (individual plants of similar size) of the above-ground tissues of both crops were selected at random for harvest in July 2014 from the non-composted subplots (arugula N = 7–8). Arugula aboveground tissues (composites of several plants, N = 2–3) were harvested at the same time from the 5% and 10% composted subplots. The freshly harvested portions were washed thoroughly in the laboratory using tap water, placed in paper bags, and oven dried at 70 ◦ C. The tissue samples were then ground using a stainless steel mill and stored in sealed bags prior to acid digestion and measurement of tissue Pb and As. 2.2. Greenhouse experiment with collards A separate small-scale greenhouse assay was designed to test the hypothesis that low soil pH could allow high Pb uptake by collards under controlled conditions with Pb-spiked soils. This assay was intended to provide some indication of how strongly a brassica could accumulate Pb in above-ground tissue from a soil in which Pb is in a more plant-available form than that present in the orchard soil. Individual two-week-old collard seedlings were transplanted into small pots containing 45 g of a 1:1 peat to perlite mix spiked with 0, 100, and 200 mg/kg Pb using Pb(NO3 )2 dissolved in water. At each level of Pb, the pH was adjusted to 5, 6, and 7 using Ca(OH)2 , and all treatments were replicated (N = 3). Plants were maintained at 22 ◦ C with a 16:8 L:D cycle, and each pot received 10 mL of a complete soluble fertilizer (14 g L−1 solution, Peters Excel 15-5-15Cal Mag Special, Everris NA Inc., Dublin, OH, USA) weekly and addi-
658
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663
Table 1 As and Pb concentrations in soils (0–15 cm) of compost-amended plots and average concentrations (± standard deviations) of As and Pb in above-ground whole plants grown in these plots. Soil organic matter (%)
Soil As (mg kg−1 )
Plot (% compost)
Soil pH
A (0%) A (5%) A (10%)
6.04 5.97 6.05
4.44 6.74 9.03
27.2 15.6 12.4
B (0%) B (5%) B (10%)
5.78 6.57 6.61
4.69 8.09 11.3
58.5 50.7 49.1
C (0%) C (5%) C (10%)
5.37 6.14 6.38
6.76 11.8 14.8
80.1 76.2 63.1
D (0%) D (5%) D (10%)
6.15 6.01 6.39
7.26 11.5 15.9
215 239 174
E (0%) E (5%) E (10%)
5.24 6.08 6.18
8.12 10.7 13.4
191 181 148
Soil Pb (mg kg−1 )
Arugula As (mg kg−1 )
Arugula Pb (mg kg−1 )
2.46 ± 1.18 0.39 ± 0.15* 0.00 ± 0.00*
1.83 ± 1.27 0.95 ± 0.33 0.12 ± 0.08*
169 168 202
6.87 ± 4.81 No data 4.02 ± 1.69
1.32 ± 0.93 No data 0.15 ± 0.21*
350 333 312
8.59 ± 1.90 4.74 ± 4.03 3.28 ± 2.60
6.57 ± 2.58 3.28 ± 2.06* 0.69 ± 0.00*
1075 1880 1225
43.8 ± 23.4 42.5 ± 26.0 8.03 ± 2.32*
12.5 ± 5.61 9.15 ± 1.63 2.30 ± 0.06*
1036 987 853
78.8 ± 26.2 30.7 ± 1.13* 14.8 ± 10.4*
15.3 ± 3.83 7.66 ± 1.14* 5.04 ± 2.18*
67.9 37.4 32.5
Asterisks (*) denote significant differences in arugula As and Pb concentrations (p ≤ 0.05) when comparing the 5 and 10% compost treatments to the 0% compost controls. The missing arugula Pb and As data from Plot B (5% compost) resulted from mislabeling and accidentally discarding these samples.
tional deionized water as needed. After three weeks, above-ground tissues were harvested, ground, and analyzed for Pb and other elements by the method described below for field-grown collard and arugula samples.
soils by Pb:
= 2.3. Analytical methodology Soil pH was measured by electrode in 2014 (using 1:2 parts soil to deionized water) and ranged from 5.24 to 6.15 in the subplots without added compost, and from 5.97 to 6.61 in the compostamended subplots (see Table 1). The slightly alkaline nature and high buffering capacity of the compost made the compost-amended soils generally less acidic, even four years after the amendment. The soil organic matter contents of all soils were measured in November 2012 using loss-on-ignition. This was done by measuring weight loss of 5 g soil weighed into a crucible, pre-dried at 105 ◦ C, after heating the soil in a muffle furnace at 375 ◦ C for 24 h. Soil samples were analyzed for total Pb, As, and other elements by the Cornell Nutrient Analysis Lab using an automated hot-plate nitric/perchloric acid digestion method followed by ICP-OES analysis of the digestates. Duplicate subsamples of soils from the 0%, 5%, and 10% compost treatments were chosen from four of the plots that represent a wide range of soil Pb and As, and then extracted 24 h for readily available Pb and As using 0.01 M CaCl2 as previously described [21]. These extracts, after filtration using Whatman #42 paper, were analyzed for Pb and As by ICP-OES. Detection limits for Pb and As in the 0.01 M CaCl2 solutions were 0.002 and 0.0005 mg L−1 , respectively. Vegetable tissue samples were digested on an automated hotplate using concentrated nitric/perchloric acid and analyzed for Pb, As, and Al using ICP-OES. Aluminum analysis was done because physical contamination of plant tissues by soil particles can be estimated by analyzing the tissue content of elements that are known to be immobile in the soil and not generally taken up through roots (e.g., Al, Ti, and Cr) [6,8,26]. We selected Al as the most sensitive indicator of soil particle contamination because of its very high concentration in mineral soils and greater susceptibility than Ti or Cr to dissolution by the nitric/perchoric acid tissue digestion process [8]. The mineral soils of this site without added compost averaged about 14,000 mg kg−1 Al as measured by acid digestion and analysis by ICP-OES, with compost amendment reducing soil Al to about 13,000 mg kg−1 on average. Therefore, the following equation was used to provide an estimate of the physical contamination of arugula grown in the compost-amended and unamended
Pb contamination index mg kg−1 Plant tissue Al mg kg−1
Measured soil Al mg kg
−1
× soil Pb mg kg−1
The index assumes that Pb-bearing particles deposited on plant surfaces have approximately the same elemental composition as the whole soil. This assumption could fail, for example, if finer soil particles that easily adhere to surfaces have much higher or lower Pb concentrations than larger particles. Procedures used to ensure precision and accuracy in the measurements of Pb and As in the plant tissues and soils by ICP-OES included the use of certified SRMs (NBS 1571-orchard leaves, NIST 2702-marine sediment, NIST 2709-San Joaquin soil) as well as laboratory internal standards in the sample sets, along with duplicate samples and blanks in each sample set. Elemental recoveries for Pb and As were 97% and 79%, respectively, from SRM 2702, and 76% and 77%, respectively, from SRM 2709. Recoveries for Pb and As from the orchard leaf SRM (1571) were 84% and 80%, respectively. 2.4. Statistical analysis Welch’s t-test (used because of data heteroskedasticity) for unpaired data was employed for comparisons of treatment effects with the p-value for significance set at 0.05. The statistical significance (p ≤ 0.05) of correlation coefficients (r) of linear regression were determined using Fisher’s two-tailed significance test. 3. Results and discussion 3.1. As–Pb relationships in the soil The wide range of total Pb (and associated As) as well as the variation of the As/Pb ratio in the soils of the orchard field site used in this study is illustrated in Fig. 1. The soil total As and Pb concentrations were highly correlated at this orchard site [13], a fact attributable to the same source of contamination for both elements (i.e., lead arsenate used historically as a pesticide). Previous research showed that there has likely been some long-term loss of As from the topsoil (relative to Pb) because the As/Pb ratio in the orchard soils is substantially lower than that expected for the specific Pb arsenate pesticide applied at the site [13]. Further analysis
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663
659
that the high compost addition in 2010 mobilized and facilitated leaching of a small fraction of the total soil As. Numerous studies have observed that increasing organic amendments and dissolved organic matter can have this effect in soils [5,42,44]. The 0.01 M CaCl2 -extractable As and Pb concentrations were generally low, but but tended to increase with increasing soil total As and Pb (Fig. 2). The results indicate that plant-available As was approximately linearly proportion to total soil As concentration, whereas available Pb increased substantially only when total soil Pb exceeded approximately 400 mg kg−1 . Compost amendment increased available As at all levels of soil total As (see Fig. 2a), consistent with the commonly observed phenomenon that compost amendments increase the fraction of soil As in soluble form [5,14,42]. Compost had the opposite effect on Pb, reducing its availability as measured by 0.01 M CaCl2 extraction; however, this effect was only detectable at the highest soil Pb concentrations. Previous studies have similarly shown that natural organic materials such as composts can reduce both solubility and bioavailability of Pb [19,44,48]. 3.2. As and Pb transfer into field-grown arugula and collards
Fig. 1. Relationship of soil total As:Pb ratio to soil total Pb. Unshaded squares with a dashed best-fit line indicate As:Pb ratios at 0% compost addition, partially shaded squares with a dotted best-fit line indicate As:Pb ratios at 5% compost addition, and fully shaded squares with a solid best-fit line indicate As:Pb ratios at 10% compost addition.
of these soils in the present study revealed that the ratio of As to Pb in the topsoil varies substantially as a function of total Pb as shown by the slopes of the best-fit lines in Fig. 1. This trend could be explained by greater historical As loss (relative to Pb) from those topsoils most heavily contaminated with the pesticide. Since As is known to migrate to some degree from the topsoil by leaching [34], this result implies that As solubility is disproportionately higher in the more severely contaminated soils of this site. Compost added at the highest (10%) level may have slightly reduced the soil As:Pb ratio as indicated by a small but statistically insignificant shift of the best-fit line for the 10% compost-amended soil in Fig. 1 compared to the 5% compost-amended and unamended soils. It is possible
Arugula was found to be a strong accumulator of As in the field trials, as shown by the high average plant concentrations of As (44–79 mg kg−1 ) when grown in the uncomposted high-As plots (see data for plots D and E in Table 1). Arugula had consistently higher (approximately five-fold) As content than collards at the same soil contamination level (Fig. 3a and c). This result concurs with the observations of other researchers, where arugula was found to be a stronger accumulator than other leafy green vegetables [40]. Arugula also averaged relatively high Pb concentrations (12.5–15.3 mg kg−1 ) when grown on the unamended (0% compost) high Pb (D and E) plots. The As concentrations in arugula and collards tended to increase with soil total concentrations of these toxic metals as shown by the best-fit relationships between plant tissue and soil As in Fig. 3a and c. Furthermore, the As soil-plant transfer coefficients (defined as the slopes of the trend lines in Fig. 3a and c) were close to 0.10 for the least As-contaminated plots, but were higher for the severely As-contaminated plots (0.20 and 0.41 for unamended plots D and E, respectively), suggesting that the bioavailable fraction of soil As was higher in these soils. Conversely,
Fig. 2. Relationship between soil concentrations and extractable concentrations of As (a) and Pb (b) using calcium chloride (CaCl2 ) for 0%, 5% and 10% compost addition to the soil. The error bars denote ± one standard deviation unit.
660
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663
Fig. 3. Relationships of arugula As and Pb concentration (a and b, respectively) and collard As and Pb concentrations (c and d, respectively) to soil total As and Pb. Best-fit lines and equations are shown in each case.
the Pb soil-plant transfer coefficients were much lower than those for As regardless of soil Pb level, and averaged 0.016 ± 0.007 for all of the unamended plots. Compost amendment of the soils caused both the Pb and As transfer coefficients to be reduced. The relationships between crop Pb and soil Pb concentration, shown in Fig. 3b and d, revealed a more random pattern than for As. Arugula generally averaged higher Pb when grown in soils with greater Pb levels, but collards showed no significant relationship of plant Pb to soil Pb concentration. Overall, there was a high degree of apparently random variability in arugula and collard concentrations of As and Pb within replicate samples (see Table 1, Fig. 3). Weak or non-existent relationships of vegetable Pb concentrations to soil total Pb concentrations have been observed in numerous other field studies [12,16,29,37] and point to the likelihood that, at contaminated sites, inherently random processes such as particulate deposition may have a dominant role in crop contamination by Pb [7,25]. The compost treatments reduced average As and Pb concentrations in arugula grown in all plots (Table 1). The 10% compost
amendment generally had a larger effect than the 5% amendment in reducing crop As and Pb, but not all reductions were statistically significant. All plots showed statistically significant (p ≤ 0.05) reductions in crop Pb at the 10% compost level, but only half showed significant reductions at the 5% compost.. For crop As, 3 of 5 plots showed significant reductions at the 10% compost level, and half showed significant reductions at the 5% level. The reduced toxic metal concentrations in arugula caused by compost amendment may be tentatively attributed to one or several factors, including physical dilution of As and Pb concentrations in the soils by compost amendments, and chemical stabilization of soil As and/or Pb by interaction of these metals with the added organic matter. Both effects have been observed in other field studies that used sewage sludge compost amendments in brownfield remediation, with the evidence for Pb stabilization by adsorption being more convincing than that for As [3,11]. However, other factors may also reduce measured Pb and As concentrations in crops grown on heavily composted soils, including the well-known “growth dilution” effect on trace metal concentration in plants resulting
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663
from more vigorous growth and biomass production compared to control treatments [17]. In the present study, the compost amendments markedly increased plant growth and biomass, although yields were not measured. In addition, reduced contamination of the above-ground tissues by soil particles from dust and rain splash may have occurred due to the observed effect of compost on improving soil aggregation and structure and increasing canopy cover of soil. Of these factors, the soil physical dilution effect is unlikely to be very important, since dilution by compost reduced soil total Pb concentration by only about 11–12% on average. The chemical stabilization factor appears to be important for Pb but not for As based on the results of the 0.01 M CaCl2 extractions, which indicated that compost decreased available Pb but increased available As. Nevertheless, plant As concentrations were lower in arugula grown in compost-amended soils compared to unamended soils, so that factors other than soil As availability must have limited plant tissue concentrations of As. For example, uptake of arsenate from the compost-amended soil could have been suppressed by competition for uptake with phosphate (or other anionic nutrients) present in the compost [28]. Published research studies of the effect of compost on crop uptake of As have shown inconsistent results, with not all composts increasing As solubility and higher soluble or extractable As not necessarily resulting in greater As in the crop [3,4,5,9,11,14]. The observed beneficial effect of compost in limiting arugula contamination by Pb and As may therefore be caused by multiple factors. A likely explanation for the highly variable Pb contamination of leafy crops is physical contamination of the plant surfaces in the field by soil particles arising from dust and soil splash [7,25]. Dust and aerosol deposition from the air, rather than root uptake and translocation into plant tops are considered to be the source of most of the Pb in plants [27]. The harvested arugula and collards had been washed several times with water before tissue analysis, but previous studies have shown that various methods of washing and cleaning are only moderately effective in reducing vegetable Pb contamination [6,11]. A previous greenhouse study of lettuce (Lactuca sativa) uptake of Pb and As from soils collected from the same orchard site did not show the extent of Pb contamination (lettuce Pb never exceeded 1 mg kg−1 ) observed in the present study [24], suggesting that field conditions are more conducive to particle contamination of crops than greenhouse conditions. Because of the likelihood of particle contamination of plant surfaces in the field experiment, arugula tissue concentrations of Al were measured and used to estimate soil particle contamination of the crop as described in the methods. This is a reasonable approach because the soils are not strongly acidic (soil acidity increases Al uptake in many plant species) and almost all cultivated crops, including brassicaceous species, are Al excluders [23]. Previous studies, using Al and Cr as indicators of contamination, revealed washed vegetables to contain from 0.07 to 4.88% soil by dry weight [6]. The Pb contamination index, calculated from measured total Al in the vegetables as described in the methods section, was strongly correlated to the measured Pb concentrations in arugula for the non-composted and composted plots (R2 = 0.777 and 0.808 for the control and compost-amended plots, respectively) as shown in Fig. 4. The ability of the contamination index to closely estimate the actual tissue Pb concentration implies that soil particle contamination of arugula is the main process of Pb transfer into this crop. The data in Fig. 4 also reveal that Pb contamination tended to be lower for arugula grown on the composted compared to the uncomposted plots, as also indicated by the effect of compost amendment in reducing the average Pb concentration in arugula (Table 1). The generally lower contamination indices for arugula grown on the composted plots indicate generally lower soil particle contamination (i.e., lower plant tissue Al) of the compost-grown crop (Fig. 4). Overall, the Pb contamination index, which takes into
661
Fig. 4. Relationship of predicted arugula Pb concentration (based on the contamination index) to the measured Pb concentration in crops from the control (open squares) and compost-amended (shaded squares) field plots.
account physical contamination of above-ground tissues by soil particles, explains the highly variable arugula Pb concentrations better than soil Pb concentration alone. Collards grown on the same field plots showed a very weak and barely significant (R2 = 0.075, p = 0.05) relationship of this Pb contamination index to actual collard crop Pb (data not shown). The average Pb contamination index was much lower for collards (mean = 0.79 ± 0.97) than for arugula (mean = 5.74 ± 5.54) because of lower average collard tissue Al. Thus, the highest observed collard Pb levels (data “outliers”) seen in Fig. 3d) could not be explained by the index. Nevertheless, the lower average Al content of collards compared to arugula can be attributed to the smooth large leaves of collards, which are more easily washed to remove surface contaminants than the trichome-studded, convoluted leaves of arugula. The high Pb levels in some collards are therefore unlikely to be due primarily to soil particle contamination. The data “outliers” originated in plots without added compost, which had somewhat lower pH than the composted plots, and may therefore have had higher Pb bioavailability. Poor plant vigor in the uncomposted plots may also have contributed a “growth concentration” effect for Pb, a phenomenon reported in other studies of trace metal uptake where plants are stunted by growing conditions [17,39]. Greenhouse studies using soils artificially spiked with soluble Pb salts have shown that very high Pb concentrations in the above-ground tissues of brassicaceous plants are possible [46,47], although the lability of soil Pb in experiments with artificially Pb-spiked soils was probably much greater than was the case for the historically contaminated orchard soils used here. There was a significant positive correlation between collard Pb concentration and that of other elements such as S (R2 = 0.257, p < 0.01), with the highest Pb tissues having quite elevated S concentrations. This may be related to poor plant growth and stunting, which can be associated with abnormal concentrations of nutrients in tissues [17], and could possibly have contributed to the few unusually high tissue Pb concentrations in collards. Unlike Pb, As in arugula grown on the non-composted plots was not significantly correlated to the As contamination index, calculated analogously to the Pb index (data not shown). This result is not surprising as As has a much higher soil solution concentration in this orchard soil than Pb ([26]), and most As transfer into the leafy vegetables is presumeably by root uptake from soil solution. The
662
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663
Fig. 5. Above-ground concentrations of Pb in greenhouse-grown collards at three levels of Pb added to soil (0, 100, 200 mg kg−1 ) and soil adjusted to pH 5 (gray bars), 6 (black bars), and 7 (diagonally hatched bars).
much higher As/Pb concentration ratio in the arugula plants (averaging 3.1 over all plots) compared to that in the soil (about 0.28 averaged over all plots) provides further evidence that soil particle contamination did not have an important role in As transfer into arugula. 3.3. Pb transfer into greenhouse-grown collards Spiking the greenhouse soil with 100 and 200 mg kg−1 of soluble Pb and adjusting the pH over the range of 5–7 caused no measurable difference in collard growth compared to the control as indicated by measured dry matter yields. Plant growth was vigorous for all treatments with no indication of Pb phytotoxicity. Pb concentrations in the harvested collards increased significantly for both the 100 and 200 mg kg−1 levels of soil Pb (p < 0.0001), although uptake of Pb was not significantly affected by soil pH (Fig. 5). Pb uptake coefficients were calculated to be 0.018, 0.014, and 0.013 at pH 5, 6, and 7, respectively. These are much lower Pb uptake coefficients than those reported for Chinese cabbage (Brassica pekinensis) grown on Pb-spiked soils [46,47], but collards have been shown to accumulate significant Pb in the edible tissues from freshly Pb-spiked soils [46]. Much higher Pb concentrations were occasionally measured in field-grown collards than observed in this greenhouse study. These extreme Pb values are not readily explained, but might have resulted from several field-related factors such as longer growth period and restricted growth due to abiotic (e.g., fluctuating temperature, poor soil structure, and soil moisture extremes) and biotic (e.g., herbivory and exposure to pathogens) stresses. One suggestion of plant stress in the field is the higher sulfur concentrations measured in the field-grown (9570 ± 4540 mg kg−1 ) compared to the greenhouse-grown collards (6030 ± 2540 mg kg−1 ). Plant tissue Pb and S concentrations were statistically correlated (R2 = 0.257, N = 54, p < 0.01) in field-grown collards, with S in the most Pbcontaminated collards exceeding 10,000 mg kg−1 . This very high S level could be due to poor growth of plants, which allows nutrients to reach high concentrations in tissues [17,39]. 4. Conclusions Arugula had a stronger tendency than collards to accumulate As in above-ground tissues when grown in orchard soils contaminated
historically by Pb arsenate pesticides. Arsenic uptake coefficients for arugula (ratio of plant to soil As concentration) generally were much higher than those for Pb and increased at higher soil total As concentrations. Although arugula is not an As hyperaccumulator, it does appear to be more efficient than most leafy vegetables in accumulating As in above-ground tissues, and may be a useful indicator species for assessing bioavailability of As at contaminated sites. In contrast to As, Pb transfer into arugula seemed primarily due to physical contamination of the plants by soil particles as indicated by a relationship of Pb to Al measured in the above-ground tissues. Several unusually high Pb levels measured in collards could not be explained by soil particle contamination of the tissues and may have been related to stunted growth. Compost amendments to the contaminated orchard soils substantially reduced the As and Pb concentrations in the harvested crop in most cases, but this may have been predominantly from metal dilution in the plant tissues by more vigorous growth rather than an overall reduction of metal bioavailability by adsorption on organic matter. The compost amendment actually increased As extractability in the soil even as it decreased extractable soil Pb. There are no health guidance values established in the USA for As and Pb concentrations in vegetables, but values suggested by other studies are 0.3 mg kg−1 Pb (fresh wt. basis) based on EU standards [25] and 0.1 mg kg−1 As (fresh wt. basis) based on a conservative risk assessment [11]. For arugula and collards, with measured moisture contents of 88–89% in the present study, these guidance values translate on a dry weight basis to about 2.7 mg kg−1 Pb and 0.9 mg kg−1 As. Inspection of the average Pb concentrations in arugula grown on the orchard plots (Table 1) shows that only those plots with higher than 300 mg/kg total Pb produced crops exceeding the guidance value, although compost amendment did lower crop Pb to acceptable levels in some of the high-Pb soils. On the other hand, As in arugula exceeded the guidance value by a large margin for all plots, and even the plot with lowest soil As (<30 mg kg−1 ), unless it had been amended with compost, produced arugula with As exceeding the guidance value. Collards accumulated lower As concentrations than arugula when grown in the same plots. Nevertheless, all collards grown on unamended plots with >150 mg kg−1 total As exceeded the As guidance value, more than half of collards grown on unamended soils with 50–100 mg kg−1 total As exceeded the guidance value, and only the least contaminated plot (<30 mg kg−1 ) produced collards below the guidance value. The Pb guidance value was exceeded in about in about 60% of collards grown on all the unamended plots (N = 54). Acknowledgements This research was funded by Federal Hatch project NYC-125445, the Toward Sustainability Foundation research grant, and the Atkinson Sustainable Biodiversity Fund. We thank Betsy Leonard for allowing this research to occur at the Dilmun Hill Student Organic Farm and Alice Jenkins for laboratory assistance. References [1] V. Angelova, R. Ivanova, K. Ivanov, Heavy metal accumulation and distribution in oil crops, Commun. Soil Sci. Plant Anal. 35 (2005) 2551–2566. [2] C.F. Aten, J.B. Bourke, J.H. Martini, J.C. Walton, Arsenic and lead in an orchard environment, Bull. Environ. Contam. Toxicol. 24 (1980) 108–115. [3] C.P. Attanayake, G.M. Hettiarachchi, S. Martin, G.M. Pierzynski, Potential bioavailability of lead, arsenic, and polycyclic aromatic hydrocarbons in compost-amended urban soils, J. Environ. Qual. 44 (2015) 930–944. [4] X. Cao, L.Q. Ma, Effects of compost and phosphate on plant arsenic accumulation from soils near pressure-treated wood, Environ. Pollut. 132 (2004) 435–442. [5] X. Cao, L.Q. Ma, A. Shiralipour, Effects of compost and phosphate amendments on arsenic mobility in soils and arsenic uptake by the hyperaccumulator Pteris vittata L, Environ. Pollut. 126 (2003) 157–167.
M.P. Lim, M.B. McBride / Journal of Hazardous Materials 299 (2015) 656–663 [6] E.E. Cary, D.L. Grunes, S.L. Dallyn, G.A. Pearson, N.H. Peck, R.S. Hulme, Plant Fe, Al and Cr concentrations in vegetables as influenced by soil inclusion, J. Food Qual. 17 (1994) 467–476. [7] A.C. Chamberlain, Fallout of lead and uptake by crops, Atmos. Environ. 17 (1983) 693–706. [8] J.H. Cherney, D.L. Robinson, L.C. Kappel, F.G. Hembry, R.H. Ingraham, Soil contamination and elemental concentrations of forages in relation to grass tetany, Agron. J. 75 (1983) 447–451. [9] R. Clemente, D.J. Walker, M. Pilar Bernal, Uptake of heavy metals and As by Brassica juncea grown in a contaminated soil in Aznalcollar (Spain): the effect of soil amendments, Environ. Pollut. 138 (2005) 46–58. [10] T.L. Creger, F.J. Peryea, Lead and arsenic in two apricot cultivars and in Gala apples grown on lead arsenate-contaminated soils, Hortic. Sci. 27 (1992) 1277–1278. [11] P.P. Defoe, G.M. Hettiarachchi, C. Benedict, S. Martin, Safety of gardening on lead- and arsenic-contaminated urban brownfields, J. Environ. Qual. 43 (2014) 2064–2078. [12] F. Douay, H. Roussel, C. Pruvot, A. Loriette, H. Fourrier, Assessment of a remediation technique using the replacement of contaminated soils in kitchen gardens nearby a former lead smelter in Northern France, Sci. Total Environ. 401 (2008) 29–38. [13] M. Fleming, Y. Tai, P. Zhuang, M.B. McBride, Extractability and bioavailability of Pb and As in historically contaminated orchard soil: effects of compost amendments, Environ. Pollut. 177 (2013) 90–97. [14] W. Hartley, N.M. Dickinson, P. Riby, N.W. Lepp, Arsenic mobility in brownfield soils amended with green waste compost or biochar and planted with Miscanthus, Environ. Pollut. 157 (2009) 2654–2662. [15] E. Hood, The apple bites back: claiming old orchards for residential development, Environ. Health Perspect. 114 (2006) A471–A476. [16] R.L. Hough, N. Breward, S.D. Young, N.M.J. Crout, A.M. Tye, A.M. Moir, et al., Assessing potential risk of heavy metal exposure from consumption of home-produced vegetables by urban populations, Environ. Health Perspect. 112 (2004) 215–221. [17] W.M. Jarrell, R.B. Beverly, The dilution effect in plant nutrition studies, Adv. Agron. 34 (1981) 197–224. [18] L. Jorhem, J. Engman, L. Lindeström, T. Schröder, Applications in food quality and environmental contamination uptake of lead by vegetables grown in contaminated soil, Commun. Soil Sci. Plant Anal. 31 (2000) 2403–2411. [19] N. Karami, R. Clemente, E. Moreno-Jimenez, N.W. Lepp, L. Beesley, Efficiency of green waste compost and biochar soil amendments for reducing lead and copper mobility and uptake to ryegrass, J. Hazard. Mater. 191 (2011) 41–48. [20] D.J. Kenyon, D.C. Elfving, I.S. Pakkala, C.A. Bache, D.J. Lisk, Residues of lead and arsenic in crops cultured on old orchard soils, Bull. Environ. Contam. Toxicol. 22 (1979) 221–223. [21] B. Kim, M.B. McBride, Phytotoxic effects of Cu and Zn on soybeans grown in field-aged soils: their additive and interactive actions, J. Environ. Qual. 38 (2009) 2253–2259. [22] K.S. MacLean, W.M. Langille, Arsenic in orchard and potato soils and plant tissue, Plant Soil 61 (1981) 413–418. [23] H. Marschner, Mineral Nutrition of Higher Plants, 2nd ed., Academic Press, San Diego, 1995. [24] M.B. McBride, Arsenic and lead uptake by vegetable crops grown on historically contaminated orchard soils, Appl. Environ. Soil Sci. 2013 (2013) 1–8. [25] M.B. McBride, H.A. Shayler, H.M. Spliethoff, R.G. Mitchell, L.G. Marquez-Bravo, G.S. Ferenz, et al., Concentrations of lead, cadmium and barium in urban garden-grown vegetables: the impact of soil variables, Environ. Pollut. 194 (2014) 254–261. [26] M.B. McBride, T. Simon, G. Tam, S. Wharton, Lead and arsenic uptake by leafy vegetables grown on contaminated soils: effects of mineral and organic amendments, Water Air Soil Pollut. 224 (1) (2013) 1–10. [27] M.J. McLaughlin, D.R. Parker, J.M. Clarke, Metals and micronutrients—food safety issues, Field Crops Res. 60 (1999) 143–163.
663
[28] A.A. Meharg, J. Naylor, M.R. Macnair, Phosphorus nutrition of arsenate-tolerant and nontolerant phenotypes of velvetgrass, J. Environ. Qual. 23 (1994) 234–238. [29] A.M. Moir, I. Thornton, Lead and cadmium in urban allotment and garden soils and vegetables in the United Kingdom, Environ. Geochem. Health 11 (1989) 113–119. [30] H. Mosbaek, J.C. Tjell, M.F. Hovmand, Atmospheric lead input to agricultural crops in Denmark, Chemosphere 19 (1989) 1787–1799. [31] G. Nabulo, H. Oryem-Origa, M. Diamond, Assessment of lead, cadmium, and zinc contamination of roadside soils, surface films, and vegetables in Kampala city, Uganda, Environ. Res. 101 (2006) 42–52. [32] P. Nathanail, C. McCaffrey, R. Ogden, N. Foster, A. Gillett, D. Haynes, Uptake of arsenic by vegetables for human consumption: a study of Wellingborough allotment plots, Land Contam. Reclam. 12 (2004) 219–238. [33] F.J. Peryea, Historical use of lead arsenate insecticides, resulting soil contamination and implications for soil remediation, In: 16th World Congr. Soil Sci. Montpellier, Fr. (1998) 20–26. [34] F.J. Peryea, T.L. Creger, Vertical distribution of lead and arsenic in soils contaminated with lead arsenate pesticide residues, Water Air Soil Pollut. 78 (1994) 297–306. [35] L.R. Prasad, B. Nazareth, Contamination of allotment soil with lead: managing potential risks to health, J. Public Health (Bangkok) 22 (2000) 525–530. [36] A. Raab, P.N. Williams, A. Meharg, J. Feldmann, Uptake and translocation of inorganic and methylated arsenic species by plants, Environ. Chem. 4 (2007) 197. [37] L. Samsøe-Petersen, E.H. Larsen, P.B. Larsen, P. Bruun, Uptake of trace elements and PAHs by fruit and vegetables from contaminated soils, Environ. Sci. Technol. 36 (2002) 3057–3063. [38] T. Schooley, M. Weaver, D. Mullins, M. Eick, The history of lead arsenate use in apple production: comparison of its impact in Virginia with other states, J. Pestic. Saf. Educ. 10 (2009) 22–53. [39] T. Shinano, T. Watanabe, Q. Chu, M. Osaki, D. Kobayashi, T. Okouchi, et al., Varietal difference in radiocesium uptake and transfer from radiocesium deposited soils in the genus Amaranthus, Soil Sci. Plant Nutr. 60 (2014) 809–817. [40] D.E. Stilwell, C.L. Musante, B.L. Sawhney, Copper, chromium and arsenic in soil and plants near coated and uncoated CCA wood, In: Proc. Annu. Int. Conf. Soils, Sediments Water Energy (2010) 10. [41] D.E. Stilwell, T.M. Rathier, C.L. Musante, Comparison of heavy metals in community garden produce versus store-bought produce, Conn. Agric. Exp. Station Bull. 1020 (2008) 1–9. [42] M. Udovic, M.B. McBride, Influence of compost addition on lead and arsenic bioavailability in reclaimed orchard soil assessed using Porcellio scaber bioaccumulation test, J. Hazard. Mater. 205–206 (2012) 144–149. [43] D. Voutsa, A. Grimanis, C. Samara, Trace elements in vegetables grown in an industrial area in relation to soil and air particulate matter, Environ. Pollut. 94 (1996) 325–335. [44] S. Wang, C.N. Mulligan, Effect of natural organic matter on arsenic release from soils and sediments into groundwater, Environ. Geochem. Health 28 (2006) 197–214. [45] G.P. Warren, B.J. Alloway, N.W. Lepp, B. Singh, F.J.M. Bochereau, C. Penny, Field trials to assess the uptake of arsenic by vegetables from contaminated soils and soil remediation with iron oxides, Sci. Total Environ. 311 (2003) 19–33. [46] Z. Xiong, F. Zhao, M. Li, Lead toxicity in Brassica pekinensis Rupr.: effect on nitrate assimilation and growth, Environ. Toxicol. 21 (2006) 147–153. [47] Z.-T. Xiong, Lead uptake and effects on seed germination and plant growth in a Pb hyperaccumulator Brassica pekinensis Rupr, Bull. Environ. Contam. Toxicol. 60 (1998) 285–291. [48] Y.-F. Zhou, R.J. Haynes, R. Naidu, Use of inorganic and organic wastes for in situ immobilization of Pb and Zn in a contaminated alkaline soil, Environ. Sci. Pollut. Res. 19 (2012) 1260–1270.