Arsenic, Cadmium, and Lead in California Cropland

TECHNICAL REPORTS: HEAVY METALS IN THE ENVIRONMENT

Arsenic, Cadmium, and Lead in California Cropland Soils: Role of Phosphate and Micronutrient Fertilizers Weiping Chen,* Natalie Krage, and Laosheng Wu University of California–Riverside Genxing Pan Nanjing Agricultural University Maryam Khosrivafard California Department of Agriculture Andrew C. Chang University of California–Riverside Phosphate and micronutrient fertilizers contain potentially harmful trace elements, such as arsenic (As), cadmium (Cd), and lead (Pb). We investigated if application of these fertilizer increases the As, Cd, and Pb concentrations of the receiving soils. More than 1000 soil samples were collected in seven major vegetable production regions across California. Benchmark soils (no or low fertilizer input) sampled in 1967 and re-sampled in 2001 served as a baseline. Soils were analyzed for total concentrations of As, Cd, Pb, P, and Zn. The P and Zn concentrations of the soils were indicators of P fertilizer and micronutrient inputs, respectively. Results showed that the concentrations of these elements in the vegetable production fields in some production areas of California had been shifted upward. Principal component analysis and cluster analysis showed that the seven production areas could be sorted into three categories: (i) enrichment of As, Cd, and Pb, which was associated with the enrichment of P and Zn in one of the seven areas surveyed; (ii) enrichment of As, which was associated with enrichment of Zn in two of the seven areas surveyed; and (iii) no remarkable correlation between enrichment of As, Cd, and Pb and enrichment of P and Zn in the other four areas surveyed.

Copyright © 2008 by the American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America. All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Published in J. Environ. Qual. 37:689–695 (2008). doi:10.2134/jeq2007.0444 Received 21 Aug. 2007. *Corresponding author ([email protected]). © ASA, CSSA, SSSA 677 S. Segoe Rd., Madison, WI 53711 USA

P

hosphate fertilizers and Fe–Mn–Zn micronutrient amendments are routinely applied on croplands to improve yields. However, the ingredients used for formulating the fertilizers and amendments may be contaminated with trace elements such as As, Cd, and Pb (McLaughlin et al., 1996; Raven and Loeppert, 1997; Bowhay, 1997). Based on a survey by the California Department of Food and Agriculture (CDFA, 1998), the trace element concentration of commercial P-fertilizers and micronutrient amendments marketed in California may vary from essentially nil to as high as 85 mg kg−1 for As, 5000 mg kg−1 for Cd, and 73,500 mg kg−1 for Pb. Although the amount being incorporated with a single application may be negligible compared with that present in the volume of receiving soil and may not be readily detectable by routine field sampling and measurement protocols, repeated applications can lead to a gradual buildup of the concentrations of trace elements in soils over time. Researchers have demonstrated that applications of P-fertilizer might inadvertently increase the trace element contents of the receiving cropland soils, especially Cd (Andrewes et al., 1996; McLaughlin et al., 1996; Richards et al., 1998; Moon et al., 2000; Abollino et al., 2002; Mann et al., 2002; de Meeữs et al., 2002). Vegetable production requires considerably higher levels of fertilizer inputs than other crops and therefore represents the worse case scenario in the accumulation of fertilizer-borne trace elements. For example, the vegetables grown in the Imperial Valley (California) typically receive 563 kg ha−1 of ammoniated phosphate (11–52–0) pre-planting (Meister et al., 2004). The climate in California often permits year-round production, and multiple crops are harvested annually. Croplands dedicated for vegetable production in California are more heavily fertilized and therefore are more likely to accumulate trace elements. The enrichments in the soils may lead to inadvertent and accelerated transfer of trace elements through the food chain. It has been reported that the concentration of an element in plant tissue increases in proportion to its concentration in soils (He and Singh, 1994; Guttormsen et al., 1995; Grant and Bailey, 1998; Huang et W. Chen, N. Krage, L. Wu, and A.C. Chang, Dep. of Environmental Sciences, Univ. of California Riverside, CA. G. Pan, Inst. of Resource, Ecosystem and Environment, Nanjing Agricultural Univ., China. M. Khosrivafard, California Dep. of Agriculture, Sacramento, CA. Abbreviations: CLS, Colusa County; COA, Coachella Valley; EF, enrichment factor; IMP, Imperial Valley; OXV, Oxnard/Ventura Area; PCA, principal components analysis; SLN, Salinas Valley; STM, Santa Maria Valley.

689

al., 2004). The possible risks derived from the increasing amount of As, Cd, and Pb in cropland soils have resulted in growing public health concerns (Krishnamurti et al., 1999; de Meeữs et al., 2002). Croplands have a long history of cultivations; yet, records on amounts and type of fertilizers applied and crops harvested are not well kept. When the trace element levels of a production field exceed the baseline, it is difficult to distinguish the contribution of trace elements from the fertilizer source and other potential sources. Trace element content of the soils may be changed through the natural weathering processes, atmospheric fallout, pesticides, and irrigation water applications. The purpose of this study was (i) to determine the As, Cd, Pb, P, and Zn contents of the benchmark soils in California; (ii) to determine the As, Cd, Pb, P, and Zn contents of the cropland soils in seven vegetable production regions in California; and (iii) to articulate the role of P fertilizer and micronutrient applications in the As, Cd, and Pb enrichment of croplands in California.

Materials and Methods Sample Collection Benchmark soils representing 50 morphologically typical soils of California were sampled by Bradford in 1967 (Bradford et al., 1967, 1996). The archived soils were included in this study. In 2001, all but one of the 50 original locations of the benchmark soils were revisited and sampled. When the benchmark soils were first identified in 1950, they were all located in relatively uninhabited areas. They were wild lands, rangelands, pastures, and low-input and lowintensity agricultural lands. The land use in several locations had been altered since the soil sampling in 1967. In California, vegetable crops have been produced in seven regions ranging from Colusa County in the north to the Imperial Valley near the US–Mexico Border for almost 100 yr. The regions from south to north are Imperial Valley (IMP), Coachella Valley (COA), Oxnard/Ventura Area (OXV), Santa Maria Valley (STM), Fresno County, Salinas Valley (SLN), and Colusa County (CLS). In each region, 65 to 170 samples were collected with the assistance of the local UC Cooperative Extension Farm Advisors. Samples were taken a minimum of 20 m away from the edge of a field to avoid influence from the road, and efforts were made to avoid potentially contaminated sites near utility poles, wooden structures, field pumping stations, etc. At each vegetable field, five surface soil samples (0–20 cm) were taken with a 5-cm bucket auger. Each of the five samples was a composite of four or five subsamples that were taken 2 m apart along a straight transect. Any organic debris at the surface was excluded. Within each agricultural region, 35 samples were taken at 150 cm deep by using a drilling rig to represent the baseline levels for As, Cd, Pb, P, and Zn. Approximately 500 g of soil were collected for each sample, including the benchmark soil samples, cropland soil samples, and baseline soil samples at a depth of 150 cm. The samples were field screened to pass a 1-mm sieve opening and stored. After air-drying, a subsample of each sample was ground with a porcelain mortar and pestle to pass a 200-mesh

690

sieve (75-μm openings) and dried in an oven overnight at 105°C before analysis.

Soil Digestion and Elemental Determinations The soil samples were microwave digested according to the USEPA Method 3052 (Wei et al., 1997; Link et al., 1998) using a CEM Mars 5 microwave system equipped with HP-500 Plus pressurized digestion vessels and perfluoro alkosy ethylene liners. An aliquot of 0.25 g soil was combined with 9.0 mL HNO3, 4.0 mL hydrofluoric acid, and 1.0 mL of deionized water in a digestion vessel and heated for 15 min (1200 W). The program was set for the temperature to rise to 180 ± 5°C within 5.5 min and to hold at that temperature for the remaining 9.5 min. Afterward, the vessels were allowed to cool in a freezer for 15 to 20 min, and then the solutions were quantitatively diluted to 25 mL with 1% v/v optima nitric acid and stored until analyzed. Graphite furnace atomic absorption spectroscopy was used to determine As, Cd, and Pb concentrations in the solutions. Total P and Zn concentrations in the solutions were determined by inductively coupled plasma optical emission spectroscopy.

Quality Control and Quality Assurance To ensure the accuracy and precision of the analyses, quality control and quality assurance protocols were adopted to check the consistency of outcomes from batch to batch. The accuracy of As, Cd, and Pb calibration curves was verified by using an NIST standard of Trace Elements in Water (NIST SRM 1640), which was analyzed after the initial calibration. If the recoveries of the NIST SRM 1640 standard exceeded ±10% of the certified value, the calibration curve was rejected until the criterion was met. For the accuracy of the sample digestion and analysis, aliquots of NIST standard reference material (NIST SRM 2709, San Joaquin Soil) were included in every batch of 12 soil samples digested. Unless the outcome of the NIST SRM 2709 was within 10% of the certified values, the analysis for the batch was rejected. To ensure the precision of the analysis, each sample was digested and analyzed in duplicate. If the relative percentage difference of the duplicate was greater than ±10%, the analysis of a sample was rejected, and the analysis was repeated. To check for the interferences of the background matrices, spiked samples were included. The analyses were rejected and repeated when the spike recoveries exceeded ±10%.

Statistical Analysis The statistical inference of the data was obtained by using SAS for Windows (version 9.0) and STATISTICA package for Windows (version 6.0). The difference in soil trace element concentrations between the baseline and those in the production fields was tested using a Tukey test. Principal Components Analysis and Cluster Analysis were applied to evaluate the underlying structure in the data set.

Journal of Environmental Quality • Volume 37 • March–April 2008

Table 1. Element concentrations (mg kg–1) in California benchmark soil samples collected in 1967 and 2001 (n = 250). Element Arsenic Cadmium Lead Phosphorus Zinc

Range 1.8–20.5 0.03–0.44 3.6–25.0 51–1127 22–143

1967 Mean ± SD 8.8 ± 4.3 0.18 ± 0.10 12.0 ± 5.3 520 ± 262 69.2 ± 25.8

Results and Discussion Benchmark Soils The trace element contents of the benchmark soils collected in 1967 and 2001 represented changes of the baselines over time. Bradford et al. (1967) reported that the elemental contents of the benchmark soils did not change significantly from 1950 through 1967. The archived samples of the benchmark soils, taken in 1967, were analyzed along with the samples collected in 2001 using the same chemical analysis protocols. The As, Cd, Pb, P, and Zn concentrations of benchmark soils sampled in 2001 compared with 1967 were a snapshot of the changes in the As, Cd, Pb, P, and Zn contents of the benchmark soils as California underwent rapid urbanization (Table 1). Although the average concentrations of Cd, Zn, and P were numerically higher in the 2001 samples than in the 1967 samples, the differences were not significant (p < 0.05). The average As concentration in the 2001 samples was numerically lower than the average concentration in the 1967 samples; again, the difference was not statistically significant (p < 0.05). The average Pb concentrations of the samples collected in 2001 were significantly higher (p < 0.05) than that of the 1967 samples. An increase in the soil Pb concentrations between 1967 and 2001 is not unexpected because the use of leaded gasoline was not banned in California until the mid-1970s. The respective elemental concentration ranges displayed by the samples collected in 1967 and 2001 overlapped. Increases in elemental concentrations at some locations were balanced by decreases at others. Except for Pb, the baseline levels of these elements in the benchmark soils remained the same over the 35-yr period.

Cropland Soils The As, Cd, Pb, P, and Zn concentrations of the vegetable production fields in California varied from 1.2 to 18, 0.15 to 2.4, 6 to 62, 206 to 2620, and 21 to 145 mg kg−1, respectively (Table 2). The As and Zn contents of the cropland soils were within the concentration ranges defined by the benchmark soils. The upper end of the concentration ranges of Cd, Pb, and P for the vegetable fields far exceeded those of the benchmark soils. The lower end of the range remained within the baseline of those defined by the benchmark soils. These results indicate that fertilizer practices have caused the concentrations of Cd, Pb, and P in the vegetable production fields in California to shift upward. The As and Pb concentrations of the soils in vegetable production (n = 665) follow a normal distribution (R2 = 0.96 and 0.85, respectively) (Fig. 1). The Cd concentration of the soils in

Chen et al.: As, Cd, & Pb in California Cropland Soils

Median 8.5 0.17 11.4 520 72.3

Range 1.8–16.6 0.07–0.53 4.9–26.8 124–1503 34–124

2001 Mean ± SD 7.6 ± 3.9 0.22 ± 0.11 14.6 ± 5.5 616 ± 342 75.9 ± 21.8

Median 6.5 0.19 13.6 564 75.7

vegetable production (n = 665) follows a log-normal distribution (R2 = 0.95) (Fig. 1). If it is assumed that the trace element content of these soils has been altered by anthropogenic factors, the distribution patterns seem to indicate that the influences on As and Pb were symmetrical and that the influences on Cd were uneven and dominated by a few unusually high values.

Production Regions The external factors influencing the trace element contents of soils in each production region may not be the same. Trends of As, Cd, and Pb accumulation in each production region need to be assessed separately. Table 3 summarizes the As, Cd, Pb, P, and Zn concentrations of the cropland soils in the surveyed production regions in comparison with their respective baseline values. The concentration and degree of enrichment varied considerably among the production regions. The As, Cd, and Pb contents of the cropland soils in two (FSN and STM), five (FSN, IMP, SLN, OXV, and STM), and three (SLN, OXV, and STM) of the seven regions, respectively, were significantly higher than their respective baseline levels (p < 0.05). For the remainder, contents were within the baseline ranges.

Role of Fertilizers and Micronutrients The accumulation of As, Cd, and Pb in cropland soils may be caused by P-fertilizers and/or micronutrient applications (Thuy et al., 2000) and by diffuse sources such as atmospheric fallout and irrigation (Romic and Romic, 2003). The accumulation patterns may be characterized by the enrichment factor (EF), which is defined as the trace element concentration of the affected soils divided by the respective baseline value (at depth of 150 cm; see Table 3) of a region. In this manner, the data representing each region may be normalized and pooled for analysis. The CVs for the element concentrations in the benchmark soil samples (Table 1) and soil samples at 150 cm depth were around 50%. On a 95% confidential level, the trace element concentration of the unaffected soils would be within baseline mean ± 1.96 × baseline mean × CV ≈ baseline mean ± baseline mean (i.e., the EF would be >0 and 1, indicating that the Cd contents have been increased in response to anthropogenic inputs. In the case of As and Pb, more of the EFs were distributed in the upper range (1 ≤ EF ≤ 2) than the lower range (0 ≤ EF ≤ 1), and a few of them were >2, indicating that As and Pb had been slowly accumulating in California cropland soils, although

their concentrations were within the baseline. Phosphorus and zinc are normally incorporated through fertilizer and micronutrient applications and are indicators of P fertilizer and micronutrient inputs, respectively. The majority of the EFs of P and Zn were >1, indicating that their concentrations in the soil are shifting upward with the inputs. In the case of P, significant numbers of the EFs were between 2 and 3.5, indicating that the amounts of P fertilizer these soils received have been substantial. However, it was not apparent that the increasing As, Cd, and Pb contents in California croplands were correlated to the increase of P or Zn in the soils. When each production region was examined separately by linear regression, the Cd contents in croplands of OXV and FSN were found to be increasing in proportion to the phosphorus concentrations of the soils, with R2 = 0.90 (n = 65) and 0.47 (n = 65), respec-

Table 3. Element concentrations (mg kg−1) of baseline and cropland soils collected in 2001 from the studied production regions. Production No. of region† observations CLS

85

COA

80

FSN

65

IMP

80

OXV

65

SLN

120

STM

170

As Baseline Croplands 11.4 11.1 (6.1–20.5) (6.1–18.4) 4.2 3.0 (3.4–5.5) (1.2–5.4) 8.2 11.0 (6.3–9.8) (8.2–13.6) 7.6 7.7 (3.8–10.9) (4.9–11.5) 7. 7.7 (5.3–9.8) (4.7–11.0) 8.9 7.0 (4.7–11.8) (3.0–14.5) 4.0 6.6 (1.3–6.7) (3.3–13.1)

Cd Baseline Croplands 0.14 0.23 (0.08–0.19) (0.15–0.35) 0.16 0.32 (0.11–0.18) (0.19–0.90) 0.22 0.28 (0.14–0.27) (0.20–0.43) 0.25 0.38 (0.17–0.32) (0.25–0.50) 0.45 1.05 (0.27–0.77) (0.45–2.38) 0.18 0.40 (0.13–0.28) (0.25–0.65) 0.45 0.73 (0.21–0.84) (0.31–1.67)

Pb Baseline Croplands 9.3 8.5 (5.6–14.6) (6.0–17.0) 17.7 12.0 (14.3–30.1) (9.2–21.7) 11.1 10.1 (7.1–14.3) (7.8–15.0) 19.1 17.0 (14.0–31.7) (13.5–24.0) 12.7 20.0 (10.6–14.2 (14.5–38.9) 15.0 21.4 (13.4–16.3) (13.6–62.2) 13.9 16.5 (5.8–18.8) (11.4–22.0)

P Baseline 419 (256–621) 876 (368–1416) 461 (231–648) 654 (472–848) 838 (585–1335) 367 (242–582) 377 (122–717)

Croplands 486 (214–829) 1196 (527–2623) 500 (356–870) 1125 (832–1519) 1377 (805–2536) 766 (442–1155) 712 (206–1206)

Zn Baseline Croplands 78 83 (36–122) (56–112) 61 83 (34–77) (54–128) 83 62 (61–143) (48–77) 58 60 (33–76) (39–81) 69 90 (45–92) (58–149) 65 67 (48–78) (43–104) 40 52 (4–103) (21–84)

† COA, Coachella Valley; CLS, Colusa County; FSN, Fresno County; IMP, Imperial Valley; Oxnard/Ventura Area (OXV); SLN, Salinas Valley; STM, Santa Maria Valley.

692


Fig. 2. Enrichment of As, Cd, and Pb as a function of phosphorus and zinc in cropland soil of production region. The enrichment factor is the trace element concentration of the affected soils divided by the respective baseline value of a region. (a) As vs. P; (b) As vs. Zn; (c) Cd vs. P; (d) Cd vs. Zn; (e) Pb vs. P; (f) Pb vs. Zn.

tively. In addition, the As contents in croplands of FSN (n = 65) were positively correlated to phosphorus (R2 = 0.57) and Zn concentrations (R2 = 0.71) through linear regressions. The underlying structure in the data set was further evaluated by principal components analysis (PCA) and cluster analysis. Principal components were derived from the sample correlation matrix to give equal emphasis to all elements regardless of their absolute variances. The correlation matrix between the principal components (As, Cd, Pb, Zn, and P) is presented in Table Table 4. Correlation matrix of the element concentrations in cropland soils (n = 665) of California. As Cd Pb Zn P

As 1.00 −0.12 −0.14 0.04 −0.42

Cd −0.12 1.00 0.37 −0.02 0.29

Pb −0.14 0.37 1.00 −0.06 0.17


Zn 0.04 −0.02 −0.06 1.00 0.37

P −0.42 0.29 0.17 0.37 1.00

4. Significant correlations were observed between Cd and Pb and soil phosphorus, indicating the application of P-fertilizers contributes significantly to the accumulation of Cd and Pb in California vegetable croplands. Under natural conditions, Cd and Pb associate with natural soil components, such as iron oxides and clay minerals, rather than with phosphate (Chen et al., 1999; Sharma et al., 2000). In addition, soil As was found to be highly negatively correlated to soil phosphorus. The eigenvalues and communalities from the first three factors of the PCA analysis are summarized in Table 5. The Table 5. Eigenvalues and communities for the first three factors of the PCA analysis. Communities Factor Eigenvalue Proportion As Cd Pb P Zn Factor 1 1.806 36.1 0.362 0.400 0.302 0.644 0.082 Factor 2 1.201 24.0 0.002 0.174 0.299 0.152 0.608 Factor 3 0.990 19.8 0.549 0.122 0.074 0.004 0.206

693

Fig. 5. Dendrogram of cluster analysis based on the average concentration of the five elements.

Fig. 3. Scores of the field samples in the two-dimensional space formed by the first factor vs. the second factor and the first factor vs. the third factor.

first three factors explain 80% of the variance in the data. The enrichments of P, Cd, Pb, and As are the dominating features

Fig. 4. Dendrogram of the seven production region in terms of PC 1 allocation.

694

in the first factor. The enrichment of Zn and As are the dominating feature in the second and third factors, respectively. Three patterns can be identified based on the factor loadings of the field samples in the 2D space (Fig. 3): (i) Enrichment of As, Cd, and Pb was associated with enrichment of P (OXV); (ii) enrichment of As was associated with the enrichment of Zn (FSN, CLS); and (iii) there was no correlation between enrichment of As, Cd, and Pb and enrichment of P and Zn (STM, SLN, IMP, COA). The role of P fertilizer on As, Cd, and Pb enrichment is further illustrated by a dendrogram of the seven production regions in terms of factor 1 allocation (Fig. 4). Because factor 1 extracted from the PCA was explained by the enrichment of P, Cd, Pb, and As, the greater the distance, the higher the enrichment. The soils of CLS and FSN recorded the lowest two mean P concentrations (see Table 3), and the means were only slightly higher than their background. Thus, the role of P fertilizer on enrichment of As, Cd, and Pb in these two regions is quite weak (the two left-hand side bars in Fig. 4). On the other hand, the OXV region recorded the highest P content, and P was significantly accumulated in this region (see Table 3). Consequentially, the role of P-fertilizer on the enrichment of As, Cd, and Pb is quite distinct (right-hand bar in Fig. 4). The other four regions (STM, SLN, IMP, and COA) were positioned between these two groups. The dendrogram based on the average linkage of different sampling fields in the seven production regions is shown in Fig. 5. The results were similar to those obtained by the PCA analysis (Fig. 3 and Fig. 4). The OXV region was distinguished from all others due to high accumulation of P and other elements. The FSN and CLS regions had similar patterns and were combined into a single cluster. Similarly, the STM and SLN regions and the IMP and COA regions were combined into separate single clusters, respectively. Given that the distance between these two middle clusters was small, they may be combined into a single category as indicated by the PCA analysis.


Conclusions The baseline ranges of As and Cd in California soils have remained unchanged for 35 yr, but Pb concentrations at the benchmark sites have increased. Correlation analysis showed that there was a tendency for Cd and Pb to accumulate in California cropland soils in proportion to levels of phosphorus fertilizer applied. Three enrichment patterns of As, Cd, and Pb were observed in the production regions. These enrichments can be statistically linked to the application of fertilizers and micronutrient amendments to soils in three of the main vegetable-producing regions.

Acknowledgments This research was supported by the California Department of Food and Agriculture. We thank three anonymous reviewers for their thoughtful comments, Dr. Albert L. Page, emeritus professor of University of California, and David Birkle, staff research associate at UC center for water resources, for their help with the English.

References Abollino, O., M. Aceto, M. Malandrino, E. Mentasti, C. Sarzanini, and F. Petrella. 2002. Heavy metals in agricultural soils from Piedmont, Italy: Distribution, speciation, and chemometric data treatment. Chemosphere 49:545–557. Andrewes, P., R.M. Town, M.J. Hedley, and P. Loganathan. 1996. Measurement of plant available cadmium in New Zealand soils. Aust. J. Soil Res. 34:441–452. Bowhay, D. 1997. Screening survey for metals in fertilizers, industrial byproduct fertilizers in Washington State. Dep. of Ecology Publ. 97-341, Washington State. Bradford, G.R., R.J. Arkley, P.F. Pratt, and F.L. Bair. 1967. Total content of nine mineral elements in 50 selected benchmark soil profiles of California. Hilgardia 38:541–556. Bradford, G.R., A.C. Chang, A.L. Page, D. Bakhtar, J.A. Frampton, and H. Wright. 1996. Background concentrations of trace and major elements in California soils. Kearney Foundation Special Rep., University of California, Riverside. California Department of Food and Agriculture. 1998. Development of risk-based concentrations for arsenic, cadmium, and lead in inorganic commercial fertilizers. Prepared by Foster Wheeler Environmental Corp., Sacramento, CA. Chen, M., L.Q. Ma, and W.G. Harris. 1999. Baseline concentrations of 15 trace elements in Florida surface soils. J. Environ. Qual. 28:1173–1181. de Meeữs, C, G.H. Eduljee, and M.Hutton. 2002. Assessment and


management of risks arising from exposure to cadmium in fertilizers. Sci. Total Environ. 291:167–187 Grant, C.A., and L.D. Bailey. 1998. Nitrogen, phosphorus, and zinc fertilizer management effects on grain and cadmium concentration in two cultivars of durum wheat. Can. J. Plant Sci. 78:63–70. Guttormsen, G., B.R. Singh, and A.S. Jeng. 1995. Cadmium concentration in vegetable crops grown in a sandy soil as affected by Cd levels in fertilizer and soil pH. Fert. Res. 41:27–32. He, Q.B., and B.R. Singh. 1994. Crop uptake of cadmium from phosphorus fertilizers: II. Relationship with extractable soil cadmium. Water Air Soil Pollut. 74:251–265. Huang, B., S. Kuo, and R. Bembenek. 2004. Availability of cadmium in some phosphorus fertilizers to field-grown lettuce. Water Air Soil Pollut. 158:37–51. Krishnamurti, G.S.R., P.M. Huang, and L.M. Kozak. 1999. Desorption kinetics of cadmium from soils using M ammonium nitrate and M ammonium chloride. Commun. Soil Sci. Plant Anal. 30:2785–2800. Link, D.D., P.J. Walter, and H.M. Kingston. 1998. Development and validation of the new EPA microwave-assisted leach Method 3051A. Environ. Sci. Technol. 32:3628–3632. Mann, S.S., A.W. Rate, and A.J. Gilkes. 2002. Cadmium accumulation in agricultural soils in Western Australia. Water Air Soil Pollut. 141:281–297. McLaughlin, M.J., K.G. Tiller, R. Naidu, and D.P. Stevens. 1996. Review: The behavior and environmental impact of contaminants in fertilizers. Aust. J. Soil Res. 34:1–54. Meister, H.S., K.S. Mayberry, E.T. Natwick, T.A. Turini, J.L. Aguiar, K.M. Bali, and J.N. Guerrero. 2004. Guidelines to production costs and practices for Imperial County 2004–2005 vegetable crops. UC Cooperative Ext., Imperial County Circ., 104-V. Moon, J.W., H.S. Moon, N.C. Woo, J.S. Hahn, J.S. Won, Y. Song, X. Lin, and Y. Zhao. 2000. Evaluation of heavy metal contamination and implication of multiple sources from Hunchun basin, northeastern China. Environ. Geol. 39:1039–1052. Raven, K.P., and R.H. Loeppert. 1997. Heavy metals in the environment: Trace element composition of fertilizers and soil amendments. J. Environ. Qual. 26:551–557. Richards, I.R., C.J. Clayton, and A.J.K. Reeve. 1998. Effects of long-term fertilizer phosphorus application on soil and crop phosphorus and cadmium contents. J. Agric. Sci. 131:187–195. Romic, M., and D. Romic. 2003. Heavy metals distribution in agricultural topsoils in urban area. Environ. Geol. 43:795–805. Sharma, V.K., K.B. Rhudy, J.C. Cargill, M.E. Tacker, and F.G. Vazquez. 2000. Metals and grain size distributions in soil of the middle Rio Grande basin, Texas USA. Environ. Geol. 39:698–704. Thuy, H.T.T., H.J. Tobschall, and P.V. An. 2000. Distribution of heavy metals in urban soils: A case study of Danang-Hoian Area (Vietnam). Environ. Geol. 39:603–610. Wei, Y.L., H.M. Shyu, and K.L. Joehuang. 1997. Comparison of microwave vs. hot-plate digestion for nine real-world sediments. J. Environ. Qual. 26:764–768.

695