Progress 10/01/08 to 09/30/11
Outputs
OUTPUTS: Research in this project has focussed on the challenges presented by testing of soils for heavy metals in the urban environment, with most of the emphasis being on lead (Pb), cadmium (Cd), zinc (Zn) and arsenic (As). We have evaluated simple and inexpensive screening tests for estimating total Pb, As, Cd and Zn concentrations in contaminated urban garden soils. Because soil testing laboratories across the country are using a number of different extractants for soil Pb in particular, including Mehlich 3, 1 M nitric acid, and Modified Morgan, we compared the ability of each of these to estimate total soil Pb as measured by digestion in nitric acid (EPA Method 3050/3051). The results revealed the 1M HNO3 extraction to be the best choice as a "screening test" for Pb in urban soils, and total soil Cd and Zn can also be estimated reliably by the same test. We have further developed satisfactory semi-quantitative colorimetric test for soil Pb using the dithizone reagent, and soil copper and zinc by modifications of this colorimetric test. These visual tests has proven to provide popular and educational demonstrations for soil testing workshops held in the field and in classes with students and gardening groups. They illustrate the principle behind the soil testing process, where a toxic metal such as Pb is extracted from the soil, and the solution is then filtered and analysed by one of a number of possible methods, including the simple visual colorimetric test . An additional challenge to reliable soil testing that we have encountered in urban soils is the extreme spatial heterogeneity that frequently occurs in Pb distribution within and between garden beds. This "nugget effect" means that individual tests of composite soil samples can produce very misleading Pb values, with retesting often giving different results. We have quantitatively evaluated the potential for error by observing variability of repeated testing of a given soil sample, as well as the effect of more finely grinding samples to minimize heterogeneity. We have further mapped spatial variability in soil Pb concentrations at several urban gardening sites in Ithaca, Syracuse and New York City. The uncertainties in Pb test results are attributable to extreme variability in Pb concentration both at the field scale and within small samples; the heterogeneity affects the way we interpret our surveys of soil contaminants in urban garden soils, and how we present these results to gardeners. Our work has shown that the measurement of total Cd in urban soils by the standard EPA 3051-6010 method (acid digestion of soil and Cd measurement by ICP-emission spectroscopy) is not reliable at low to moderate soil contamination levels ( 0.2 - 5 mg/kg) because of serious matrix or elemental interferences in the most commonly used analytical emission lines for this element. A simpler and lower cost approach to measuring soil Cd using extraction by 1 M HNO3 and analysis by flame atomic absorption was investigated and found to be superior to the widely used ICP-emission method. PARTICIPANTS: The project provided the opportunity in 2010 for one Cornell graduate student as well as undergraduate students to carry out independent projects related to soil contamination by Pb, Zn, Cd and As. In addition, a visiting Fulbright scholar from Slovenia conducted experiments on soil Pb and As bioavailability measurements in our lab. TARGET AUDIENCES: Our interactions with community stakeholders (e.g., through gardening events and discussion forums in NYC and Ithaca, urban farming workshops in Buffalo, responding to information requests by email and phone) have indicated a need for comprehensive educational programs addressing diverse topics, including: 1) Training on site assessment for contaminants and soil sampling and testing protocols; 2) Information about and access to reliable, affordable, certified soil testing labs; 3) Simple guidelines for interpretation of soil test results that allow for site-specific considerations; 4) Assessment of contaminants in municipal compost and available soil/fill, and access to these materials; and 5) Possible liability issues or closure or avoidance of gardens if soil tests reveal contamination. Additional fact sheets, workshops, etc. to address these topics were developed to augment existing resources available at our CWMI website under the heading "Soil Quality". PROJECT MODIFICATIONS: Not relevant to this project.
Impacts
The results from the soil lead (Pb), cadmium (Cd) and zinc (Zn) test comparisons have produced a highly useful observation in practice - extraction of a wide range of soil types with 1 M nitric acid provides a very good approximation of total lead, cadmium and zinc in the soil without the added effort and expense of measuring total metals using an acid soil digestion process. Furthermore, the Pb, Cd and Zn in the acid extracts are easily measured by either standard flame atomic absorption or by ICP emission with equivalent results. Because many laboratories, particularly in underdeveloped countries, have greater access to atomic absorption equipment than to ICP because of lower cost, these simpler "screening" tests may be adopted by many laboratories for practical reasons. We have used the screening test for Pb to advantage in our field experiments, where it is important to get soil Pb results in 1-2 days in order to expedite the siting and design of field plots. In fact, this screening test has been introduced as a new and less expensive soil test for lead by the Cornell Nutrient Analysis Laboratory. The observations on Pb heterogeneity in soils have informed our recommendations on number of discrete soil samples needed to characterize gardens and yards, and the methods best used to prepare these samples prior to analysis. Certainly, it is dangerous to base management of contaminated yards and gardens on the results from one or two composited soil tests, and grinding of soils has the effect of reducing soil test variability. Our understanding of heterogeneity in Pb concentrations found at several spatial scales in urban soils is helping to guide future sampling designs in New York City community gardens, and is also influencing our recommendations to homeowners who wish to test their own yards for Pb and other toxins.
Publications
- McBride, MB. 2011. A comparison of reliability of soil cadmium determination by standard spectrometric methods. Journal of Environmental Quality 40, 1863-1869
- McBride, MB, RR Mathur and LL Baker. 2011. Chemical extractability of lead in field-contaminated soils: implications for estimating total lead. Communications in Soil Science and Plant Analysis 42; 1581-1593.
|
Progress 10/01/09 to 09/30/10
Outputs
OUTPUTS: Research in the past year has focussed on the challenges presented by testing of soils for heavy metals in the urban environment, with most of the emphasis being on lead (Pb), cadmium (Cd) and arsenic (As). We are evaluating simple and inexpensive screening tests for estimating total Pb, As and Cd concentrations in contaminated urban garden soils. Because soil testing laboratories across the country are using a number of different extractants for soil Pb in particular, including Mehlich 3, 1 M nitric acid, and Modified Morgan, we compared the ability of each of these to estimate total soil Pb as measured by digestion in nitric acid. We used 45 soils collected from New York City community gardens in 2009-2010. Overall, both 1 M HNO3 extraction and Mehlich 3 gave a satisfactory correlation to total Pb determined by microwave soil digestion, with regression r2 values of 0.869 and 0.831, respectively. The correlation for Modified Morgan, while strong ( r2 = 0.702), was less acceptable if the test is to be used as a proxy for total soil Pb. These results validate the 1M HNO3 extraction as a viable "screening test" for Pb in urban soils. We have further developed a satisfactory semi-quantitative colorimetric test for soil Pb using the dithizone reagent. This visual test has proven to be a popular and educational demonstration for soil testing workshops held in the field with students and gardening groups. It illustrates the principle behind the soil testing process, where a toxic metal such as Pb is extracted from the soil, and the solution is then filtered and analysed by one of a number of possible methods. An additional challenge to reliable soil testing that we have encountered in urban soils is the extreme spatial heterogeneity that frequently occurs in Pb distribution within and between garden beds. This "nugget effect" means that individual tests of composite soil samples can produce very misleading Pb values, with retesting often giving different results. We have quantitatively evaluated the potential for error by observing variability of repeated testing of a given soil sample, as well as the effect of more finely grinding samples to minimize heterogeneity. These uncertainties in Pb test results owing to the nature of Pb contamination are affecting the way we interpret our ongoing survey of soil contaminants in urban garden soils, and how we present these results to gardeners. Our work has shown that the measurement of total Cd in urban soils by the standard EPA 3051-6010 method ( ICP-emission spectroscopy) is not reliable at low to moderate soil contamination levels ( 0.2 - 5 mg/kg) because of serious matrix or elemental interferences in the most commonly used analytical emission lines for this element. Other low-cost approaches to measuring soil Cd are being investigated, but our present recommendation is that Cd is best determined in soil digests by ICP-MS, a very sensitive method for soil trace metals analysis. PARTICIPANTS: The project provided the opportunity in 2010 for one Cornell graduate student as well as undergraduate students to carry out independent projects related to soil contamination by Pb and As. In addition, a visiting Fulbright scholar from Slovenia is conducting experiments on soil Pb and As bioavailability measurements in our lab, and is obtaining training in soils analysis. We are collaborating with laboratory directors at the Cornell Nutrient Analysis Lab (CNAL) as well as other soil testing labs in Northeastern states. TARGET AUDIENCES: Our continued interactions with community stakeholders (e.g., through gardening events and discussion forums in NYC and Ithaca, urban farming workshops in Buffalo, responding to information requests by email and phone) have indicated a need for comprehensive educational programs addressing diverse topics, including: 1) Training on site assessment for contaminants and soil sampling and testing protocols; 2) Information about and access to reliable, affordable, certified soil testing labs; 3) Simple guidelines for interpretation of soil test results that allow for site-specific considerations; 4) Assessment of contaminants in municipal compost and available soil/fill, and access to these materials; and 5) Possible liability issues or closure or avoidance of gardens if soil tests reveal contamination. Additional fact sheets, workshops, etc. to address these topics are currently under development to augment existing resources available at our CWMI website under the heading "Soil Quality". PROJECT MODIFICATIONS: Not relevant to this project.
Impacts
The results from the soil lead (Pb) test comparisons have produced a highly useful observation in practice - extraction of a wide range of soil types with 1 M nitric acid provides a very good approximation of total lead in the soil without the added effort and expense of measuring total lead using an acid soil digestion process. The lead in the acid extracts is easily measured by standard flame atomic absorption or by ICP emission with equivalent results. We have used this quicker test to advantage in our field experiments, where it is important to get soil Pb results in 1-2 days in order to expedite the siting and design of field plots. In fact, this "screening test" has been introduced as a new and less expensive soil test for lead by the Cornell Nutrient Analysis Laboratory. The observations on Pb heterogeneity in soils have changed our recommendations on number of discrete soil samples needed to characterize gardens and yards, and the methods best used to prepare these samples prior to analysis. Certainly, it is dangerous to base management of contaminated yards and gardens on the results from one or two composited soil tests, and grinding of soils has the effect of reducing soil test variability. Our understanding of heterogeneity in Pb concentrations found at several spatial scales in urban soils is helping to guide future sampling designs in New York City community gardens, and is also influencing our recommendations to homeowners who wish to test their own yards for Pb and other toxins.
Publications
- No publications reported this period
|
Progress 10/01/08 to 09/30/09
Outputs
OUTPUTS: To compare various soil tests in use for toxic metals, we compared the extractability of copper and zinc in contaminated soils using the Morgan's , Modified Morgan's, DTPA and Mehlich 3 extractants. These methods were compared with 0.01 M CaCl2 extraction to determine their relative abilities to estimate phytoavailability of Cu and Zn. In aged, metal-spiked soils, all soil tests evaluated showed a linear relationship of extractable to total Cu and Zn for both soil types studied. The fraction of total Cu and Zn extracted by aggressive tests (Mehlich 3, DTPA) was much higher than the fraction extracted by CaCl2 , with the Morgan tests being intermediate. The study suggested that non-aggressive soil extracts such as 0.01 M CaCl2 are more responsive to soil properties affecting chemical lability of the metals. Therefore, chemically non-aggressive neutral salts are the most appropriate extractants where phytotoxicity is the concern in metal-contaminated soils. The second study specifically addressed methods of soil testing for a toxic metal where phytoavailability and phytotoxicity are not the primary concern. In this study, three extractants were compared with the EPA 3051 method to evaluate their ability to estimate total Pb in soils collected from field sites with histories of contamination. The extractants tested were Na citrate (0.1 M, pH 6.0), Modified Morgan (1 M ammonium acetate, pH 4.8) and 1 M nitric acid. Nitric acid not only extracted a much greater fraction of total soil Pb (on average, about 80% vs. only 6-7% for both Modified Morgan and citrate), but also had the strongest correlation to total Pb (r2 = 0.976 compared to 0.837 for Modified Morgan and 0.582 for Na citrate). Morgan, 0.597 for Na citrate). It was concluded that 1 M nitric acid is an acceptable extractant for a soil screening test to estimate total Pb ; however, Modified Morgan's and Na citrate are not reliable extractants for this purpose. A number of colorimetric methods were evaluated for their ability to estimate total or extractable copper, zinc and lead. Satisfactory methods were developed for soil copper and zinc based on cuprizone and dithizone, respectively. None of the colorimetric methods tested for soil lead was found to be completely satisfactory, as all suffered from interferences caused by other metals present in soil. The best colorimetric method employed dithizone, but had a severe positive interference from soil zinc unless cyanide was used as a chemical suppressing agent. To meet the objectives of developing web-based educational materials and conducting outreach on soil testing for toxic metals, we have prepared and posted three fact sheets on the Cornell Waste Management Institute web site, titled "Sources and Impacts of Contaminants in Soils" (April 2009), "Guide to Soil Testing and Interpreting Results" (April 2009), and "Soil Contaminants and Best Practices for Healthy Gardens" (October 2009). More specific fact sheets dealing with arsenic and lead are being prepared. We also presented our research results on soil testing for lead to the Northeastern ASA Meeting in Portland, Maine in 2009. PARTICIPANTS: The research involved the training of several undergraduate ( Masha Pitiranggon, Diane Wu, Anna Goff ) and one graduate student ( Bojeong Kim) in conducting the work in Dr. McBride's laboratory, learning various techniques of soil and water analysis. We have partnered with the Cornell Nutrient Analysis laboratory in both evaluating the potential screening test for soil lead, and finally in providing this test for customers. We have also been working closely with Dilmun Hill Farm, a student-operated organic farm at Cornell which has serious lead and arsenic contamination from past use of lead arsenate pesticides on the Cornell orchards. This has afforded the opportunity for undergraduate students, including a Presidential Research Scholar, Bonnie Cherner, to be involved in some of our research on metal-contaminated soils. TARGET AUDIENCES: For the copper and zinc soil tests, the target audience is primarily extension agents assisting dairy and other livestock farmers, where copper and zinc are still being used in feed rations as well as for other purposes, with the result that livestock manures may be contaminated with these metals. Questions are raised by farmers about the test levels of copper and zinc in their soils, and how these tests should be interpreted. For soil lead testing, gardeners, particularly urban gardeners, are a primary audience because of widespread lead contamination in urban areas. However, soil testing laboratories in the Northeastern states and elsewhere are also a target audience, as they may be able to offer more reliable tests for soil lead based on our research. Farmers with old orchard land also will find our research results useful to their management of these contaminated soils. PROJECT MODIFICATIONS: Not relevant to this project.
Impacts
The results from the copper and soil test method comparisons have allowed quantitative relationships to be defined between the different soil tests that are being used by different states or regions. For example, a soil test value for Zn or Cu determined by Mehlich 3 in Pennsylvania could be converted to its equivalent test value using the Modified Morgan extractant in Maine. These tests could in turn be interpreted in terms of a 0.01 M CaCl2 extraction equivalent, although such test conversions introduce errors because of differences among soil types in the ease of metal extraction. The results from the soil lead (Pb) test comparisons have produced a highly useful observation in practice - extraction of a wide range of soil types with 1 M nitric acid provides a very good approximation of total lead in the soil without the added effort and expense of measuring total lead using an acid soil digestion process. The lead in the acid extracts is easily measured by standard flame atomic absorption or by ICP emission with equivalent results. In fact, this "screening test" has now been introduced as a new and less expensive soil test for lead by the Cornell Nutrient Analysis Laboratory. Quite reliable colorimetric tests for soil copper and zinc are now available from the research, although there may be limited demand for such tests despite their ease and low cost. It seems unlikely that the colorimetric test for soil lead, although potentially of great interest to homeowners and gardeners, could be made highly reliable ( not subject to false positive tests) without the dangerous cyanide reagent. At this point, therefore, this colorimetric test appears to be impractical.
Publications
- Kim, B. and M. B. McBride 2009. Phytotoxic effects of Cu and Zn on soybeans grown in field-aged soils : their additive and interactive actions. Journal of Environmental Quality, 38, 2253-2259.
- McBride, M.B., Pitiranggon, M. and Kim, B. 2009. A comparison of tests for extractable copper and zinc in metal-spiked and field-contaminated soil. Soil Science 174, 439-444 .
|
|