OVERVIEW

In this study, we quantified Cd and Pb content in soils and leafy greens from five community gardens and compared our results with regional patterns of soil Cd and Pb across the Phoenix area. A previous city-wide analysis of soil metals was conducted at 200 locations in the 2000 Ecological Survey of Central Arizona (ESCA), as a part of the Central Arizona–Phoenix Long-term Ecological Research Project (Zhuo and Shock 2010, Zhuo et al. 2012). This survey showed an uneven distribution of soil Pb and Cd across the Phoenix metro area, with high concentrations in areas of former agricultural and urban use (Zhuo 2010, Zhuo et al. 2012). We expected soil metal content of community gardens to follow this larger, city-wide pattern. We also expected that soil from elevated, raised beds would contain less Pb and Cd than the non-raised beds that are located in the ground (McLean and Bledsoe 1992, Clark et al. 2008, CDPH 2014). Finally, we expected there would be no difference in metal content between shallow (0-15 cm) and deeper soils (15-30 cm) within raised beds. Growing plants in raised beds can limit root contact with potentially contaminated pre-existing soil. In addition to examining soil contamination, we investigated the metal content of leafy greens in our five study gardens to determine if it was related to the metal content of the soil in which the plants were grown. Because soil metal content in the ESCA city-wide survey was below the recommended EPA guidelines for health concerns, and because metal content in plants can reflect how much metal is found in the soil (Toth et al. 2016), we expected that the heavy metal content for leafy greens grown in community gardens would not exceed the existing guidelines for ingestion (from the European Union; 0.3 mg Pb/kg fresh weight and 0.2 mg Cd/kg fresh weight). EPA guidelines for ingestion of heavy metals in foods do not currently exist.

SITE DESCRIPTION

In September 2014, we contacted gardens in the Phoenix metropolitan area that aim to combat food insecurity and economic marginalization issues in their communities. Among these, five different community gardens agreed to participate in this study. The gardens vary in location across the metro area, in age, size, and structure; and community members utilize a variety of growing techniques, such as importing soils, building raised beds, and using compost and irrigation to enhance productivity. For privacy, garden names are not disclosed.

EXPERIMENTAL DESIGN

In each garden, we sampled from planting bed types that were used specifically to grow leafy greens. In some gardens, these sampling locations were raised beds, in which soil was elevated and contained by a wall such as wood or recycled car tires (hereafter, called ‘Raised’ beds). In other locations, leafy greens were grown using in-ground beds, where soil was not bounded or raised (hereafter called ‘In-ground’ beds; Fig. 1). In many gardens, only one bed per category (i.e. Raised or In-ground) grew leafy greens. In the case where there were multiple raised or in-ground beds growing leafy greens within a garden, we randomly chose beds and sampled soils and plants in each (i.e. some gardens have two or three replicate soil samples of plants or soil in Raised or In-ground beds).

SOIL AND PLANT SAMPLE COLLECTION

For each soil sample, we collected two separate soil cores using a slide hammer core from 0 to 15 cm depth and another two from 15 to 30 cm depth, with cores located at least 50 cm apart from one another. The two cores from each sampling location and depth were then combined into a single plastic bag and homogenized to compose one soil sample to be analyzed. From each of our five gardens, we collected and analyzed at least one homogenized soil sample (composed of two soil cores each) from each bed type that was present at that location. In sum, we analyzed 2-6 soil samples from each of five 14community gardens. Out of the 28 soil samples analyzed, 17 were collected from 0-15 cm depth, and 11 were collected from 15-30 cm depth. In addition to soil samples, we took three samples of leafy greens that were growing in each garden and bed type, where possible. These plant samples came from Cavalo nero (kale) or Spinacia oleracea (spinach), both of which were present in most gardens. At each site, one leaf from each of two-three individual plants was randomly chosen for sampling. Inner leaves were chosen by gently pulling back an outer leaf and then, using scissors, cutting the next available leaf approximately 2 cm from the stem of the plant.

To compare metal concentrations between our sampled leafy green plants and leafy green plants commonly sold in grocery stores, we purchased one bunch of conventional spinach from each of three separate grocery stores – Sprouts, Safeway, and Food City – in the Phoenix metropolitan area in March of 2015. We sampled and analyzed 3 leaves each from 3 conventional spinach bunches. Leaves were clipped above the stem, and the 3 leaves of each bunch combined in a bag prior to processing.

SOIL AND PLANT SAMPLE PREPARATION AND METAL ANALYSIS

We sieved soils to 2 mm prior to analyses. All plant leaves were rinsed thoroughly using tap water to replicate average consumer habits. We then placed both soils and plant samples in a 105°F oven to dry overnight. Once dried, we 15pulverized each sample into fine powder using a ball mill at the Goldwater Environmental Laboratory at ASU.

We measured soil Pb and Cd content on plant and soil samples using standard EPA methods for soil trace metals (EPA 1996). We first digested about 0.25 g each of dried soil and plants in a solution of HNO 3 , HF, and HBrO 3 to dissolve the soil and plant material prior to analyses. We then diluted the samples and used inductively coupled plasma optical emission spectrometry (ICP-OES) to determine metal concentrations. The detection limit for our analyses was 0.001- 0.01 mg/kg of soil or plant material for Pb and <0.0001 mg/kg for Cd. We did not complete a full spectral analysis of other metals in the soil samples, although this method would have controlled for interactions that other metals may have with metals of interest (Cd and Pb).

ANALYSIS OF SOIL CHARACTERISTICS

In addition to metal analysis, on each soil sample we measured a suite of soil properties that can affect metal solubility, including texture (particle size analysis), pH (a measure of acidity or alkalinity), and organic matter content. Soil texture analysis. Soil particle size, or texture, influences soil porosity, water holding capacity, and how metals move through soil. We determined soil texture by using a modified hydrometer method (Bouyoucos 1962), which estimates the soil content (in %) of sand (2.0-0.05 mm diameter particles), silt (0.05-0.002 mm), and clay (<0.002 mm) (Gee and Bauder 1986). We shook a solution of 40 g of oven-dried soil with 100 mL of a sodium hexametaphosphate solution to prepare the samples for analyses. After shaking, we put each sample in a 1-L suspension cylinder and filled the cylinder to a 1-L mark with deionized water. We used a mixing rod to mix the sample until it was homogenized within the cylinder, then we placed the hydrometer into the sample. After 40 additional seconds of no mixing, we recorded the hydrometer reading. We took hydrometer readings on each sample at 40 seconds to determine the combined percent silt and clay content and 7 hours to determine the percent clay content. We subtracted the clay content from the clay plus silt content to determine percent silt, and we determined percent sand content by subtracting the percent silt plus clay from 100%.

Soil pH and organic matter analysis. Soil pH was assessed using a modified EPA method (Ghose and Pettygrove 2014). We shook approximately 15 g of soil in 30 ml of deionized (DI) water for 30 minutes, and then used a calibrated pH meter to read the pH of each sample. Percent soil organic matter was determined using a modified loss on ignition (LOI) method (Schulte and Hopkins 1996). We placed 20 g of oven dried soil samples in a 550°C furnace for 6 hours and measured the loss of mass to determine organic matter percentage.

LITERATURE CITED

Bouyoucos, G. J. 1962. Hydrometer Method Improved for Making Particle Size Analysis of Soils. Agronomy Journal 54:464-465.

CDPH. 2014. Growing and Eating Fruits and Vegetables Safely. Connecticut Department of Public Health, Environmental & Occupational Health Assessment Program, Hartford, CT.

Clark, H. F., D. M. Hausladen, and D. J. Brabancer. 2008. Urban Gardens: Lead Exposure, Recontamination Mechanisms, and Implications for Remediation Design. Environmental Research 107:312-319.

EPA. 1996. Microwave Assisted Acid Digestion of Siliceous and Organically Based Matrices (Method 3052). Environmental Protection Agency Washington, D.C.

Gee, G. W., and J. W. Bauder. 1986. Methods of Soil Analysis, Part I, 2nd ed., American Society of Agronomy/Soil Science Society of America, Madison, WI.

Ghose, R., and M. Pettygrove. 2014. Urban Community Gardens as Spaces of Citizenship. Antipode 46:1092-1112.

McLean, J. E., and B. E. Bledsoe. 1992. Behavior of Metals in Soils. EPA, Washington, D.C.

Schulte, E. E., and B. G. Hopkins. 1996. Soil organic matter: analysis and interpretation. Soil Science Society of America, Madison.

Toth, G., T. Hermann, M. R. Da Silva, and L. Montanarella. 2016. Heavy metals in agricultural soils of the European Union with implications for food safety. Environment International 88:299-309.

Zhuo, X. 2010. Distributions of Toxic Elements in Urban Desert Soils. Arizona State University, Tempe, AZ.

Zhuo, X., C. G. Boone, and E. L. Shock. 2012. Soil lead distribution and environmental justice in the Phoenix metropolitan region. Environmental Justice 5:206-213.

Zhuo, X., and E. L. Shock. 2010. Ecological Survey of Central Arizona: Soil trace elements.in Central Arizona–Phoenix Long Term Ecological Research Project, editor., Arizona State University.