Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Urban agriculture is gaining attention as a means to revitalize abandoned urban properties. The recent interest in this practice over the past decades has provided increased food security for low income families and city residents (Lovell 2010). Urban residents can either grow their own food, be part of a CSA (Community Supported Agriculture) program, or gain easier access to affordable supplies of vegetables or fresh produce from local farmers markets – reducing the food deserts in these cities. One of the major challenges of growing vegetables in formerly blighted properties in an urban environment is the possibility of soil contamination . Concerns about the perceived human health risk of gardening in urban soils due to possible or real contamination continue to deter potential gardeners from growing crops on blighted, formerly used properties. Common urban soil contaminants include lead (Pb), arsenic (As), cadmium (Cd), zinc (Zn), and polycyclic aromatic hydrocarbons (PAHs ) (Spittler and Feder 1979; Chaney et al. 1984; Alloway 2004; Roussel et al. 2010). Of these contaminants, Pb is by far the most dominant and wide-spread in urban environments. Soil remediation or managing risk posed by contaminants can be challenging as a result of poor soil quality and the presence of co-contaminants. Options such as raised-bed gardening or soil replacement can be physically and financially restrictive. Therefore, there is a great need for sharing science-based knowledge on risk management associated with Pb and other urban soil contaminants.

When growing in-situ (growing crops directly in the native soil) is selected, the soil very likely will need to be amended using compost and fertilizer because most urban soils, whether the soil is associated with a brownfield (vacant or abandoned properties with real or perceived contamination issues) or not, tend to be of poor quality. Aside from providing organic matter and improving the soil structure, the immediate and direct effect of adding compost to soil is dilution of potential soil contaminants. This will provide an immediate reduction in risk. Further, some soil amendments that are frequently used to improve soil quality may also help bind contaminants and thus reduce the bioavailability of metals (lead , and cadmium)/metalloids (arsenic). This section focuses on highlighting research findings that show the beneficial effects of soil amendments/compost additions on reducing risk from soil contaminants. Information on the mechanisms of this risk reduction will also be provided.

Researchers at Kansas State University have been evaluating the uptake of heavy metals, metalloids and other contaminants by food crops grown on urban garden sites. The goal of this research is to enhance the capabilities of garden/farming initiatives to produce crops locally without concern about adverse health effects for the grower or the end consumer; to increase confidence in urban food production quality; to provide resources for producers, urban land managers, local and state government, and extension agents to implement proposed best management practices, and to contribute to the meaningful revitalization of brownfields sites in a sustainable manner. The research is made possible by a grant from the U.S. Environmental Protection Agency (EPA).

Compost Addition Will Dilute Overall Contaminant Concentrations

Compost is one of the most effective tools for growing food in urban soils. Amending soils with compost will dilute overall soil contaminant concentrations. So that urban gardeners who add compost to soils annually will, over time, lower total contaminant concentration in soils significantly.

The above figure shows an example from a test site in Kansas City, MO (Fig. 1). The columns represent four field plots that received a single application of compost . Leaf based compost was applied at a rate of 28 kg/m2 before vegetables were planted (Attanayake et al. 2014).

Fig. 1
figure 1

Dilution effect on total contaminant concentration in soils

Full size image

Adding compost to soil may also reduce the hazards of contaminants in soil by making them less bioaccessible (see Fig. 2).

Fig. 2
figure 2

Washington Wheatley Community garden site, Kansas City, MO

Full size image

Bioaccessibility of Soil Contaminants

The total concentration of a contaminant in soil does not necessarily reflect the health risk posed by the contaminant. The health risk will vary based on the portion of the total contaminant that is available for physiological dissolution and be available for subsequent absorption by the human body. For soil contaminants like Pb and As, the primary risk comes from eating the soil (in small amounts, like a child licking a dirty finger) and having the contaminants absorbed into the body through the digestive system. The term bioaccessibility refers to the fraction of a substance (i.e., Pb or As) from a particular exposure medium (in this case soil) that is soluble in the simulated gastrointestinal environment and is available for absorption (Hettiarachchi and Pierzynski 2004). It is a term used to predict how much of a health risk is posed by contaminants in soils and other media. For example, a soil with a Pb concentrations of 1000 mg kg−1 where 40 % of the Pb is bioaccessible has an effective Pb concentration of 400 mg kg−1. Several methods have been developed and modified to estimate the bioaccessibility of soil Pb and As (Ruby et al. 1996; Medlin 1997; Rodriguez et al. 1999; Oomen et al 2002; Juhasz et al. 2007; 2009; Smith et al. 2011; USEPA 2012; Scheckel et al. 2013). In all these methods, soils are mixed with a simulated gastric (stomach) solution in an acidic environment (commonly used pH values are 1.5, 2.0, 2.3 and 2.5) at 37 °C (body temperature) to extract bioaccessible Pb and As. In addition to the pH variability, there are some variations in the composition of the enzyme mixture, soil: solution ratio, and the extraction time among these methods generating some confusion about what method is appropriate.

Bioaccessibility in Urban Soils Before and After Compost Addition

We measured both the total and bioaccessible concentrations of Pb and/or As in a number of soil samples collected from our study sites in Kansas City, KS, Indianapolis, IN, Tacoma and Seattle, WA, Philadelphia, PA, Toledo, OH, and Pomona, CA. The bioaccessibility of soil Pb and As in the urban soils were low compared to the bioaccessibilities recorded in literature for highly contaminated mine impacted soils (Scheckel et al. 2013; Attanayake et al. 2014; Defoe et al. 2014). From our sampling, the risk of exposure to soil Pb and As via direct ingestion of contaminated urban soils seems to be much lower compared to mine impacted soils. The absolute bioaccessibility (total concentration × % that is bioaccessible) of soil Pb and As can be further reduced by adding compost , due to significant dilution of the contaminants in the soil by compost addition. Further, some soil amendments added alone or in combination appear to change the chemical form of Pb and As resulting in even greater reductions in bioaccessibility (as a percentage of total concentrations). For some soils, reductions are unclear or insiginificant, most likely because of the inherent low soil Pb and As bioaccessibility in tested urban soils. For example, we added four different soil amendments to plots at a test site in Indianapolis: mushroom compost, leaf compost, biosolids (from municipal wastewater treatment) and composted biosolids (Fig. 3).

Fig. 3
figure 3

Test plots at Monon garden site, Indianapolis, IN at instillation and as the crops ripened

Full size image

The composted biosolids reduced the bioaccessible fraction of soil Pb by about 50 % compared to the unamended soil (Fig. 4). The other amendments did not appear such a strong effect on the bioaccessible fraction of Pb. All of the amendments reduced the total concentration of Pb in the soil by dilution (Attanayake et al., 2015).

Fig. 4
figure 4

Soils were collected from a test site located in Indianapolis, IN. Four different types of compost were added

Full size image

Compost Addition Will Also Increase Biomass Production Helping to Further Reduce Food-Chain Transfer of Contaminants

While the primary risk associated with contaminants in urban soils is from direct ingestion of the soils, there is also the potential for these contaminants to be taken up into plant tissue. For many urban growers, this risk is the one of greatest concerns. Despite this concern, it is not at all clear that this is a realistic risk. There are currently no standards for arsenic concentrations in food crops. Arsenic uptake is minimal under aerobic conditions. To date the primary concern with food arsenic has focused on rice. This is due to enhanced mobilization of arsenic under anaerobic conditions (flooded) and arsenic uptake sharing Si-transport pathway, highly expressed in rice (Zhao et al. 2010). The World Health Organization/ Food and Agriculture Organization (WHO/FAO) has set standards available for Pb concentrations in food crops. These standards are derived from the assumption that 100 % of Pb present in produce is bioavailable, so the recommendations are overly protective. Moreover, the recommended limits assume life-long consumption, and all vegetables of a particular category type (i.e., all types of leafy vegetables) to have the same level of lead . The recommended limits are 0.3 mg kg−1 Pb fresh weight for leafy vegetables, and 0.1 mg kg−1 Pb for fruiting vegetables, root, and tuber crops (FAO/WHO-CODEX, 1995; 2010 amendment). Our tests with compost addition to urban soils showed that in addition to diluting total soil concentrations of contaminants, compost will increase plant yield thereby also diluting any contaminant concentrations in plant tissue. A community garden in Tacoma , WA, instillation of our test plots in that garden and the increased growth as a result of compost or biosolids addition in a test plot in Tacoma, WA are shown in the pictures Fig. 5.

Fig. 5
figure 5figure 5

Community garden in Tacoma , WA, installing tests plots and differences in growth between amended and nonamended plots are striking. This picture shows growth difference in tomato plants growing in unamended control plots and Tagro (a biosolids based soil product) amended plots, Tacoma, WA test site

Full size image

Swiss chard yield in compost amended plots was 2.5× greater than in control plots in a test plot in Kansas City, MO (Table 1).

Table 1 Effects of compost addition on Swiss chard biomass (Kansas City, MO site, 2010)
Full size table

Concentrations of Pb, As and PAHs in Vegetables Harvested at Test Sites Were Low

Lead concentrations in the soils we tested across the U.S. ranged from 100 to 2200 mg kg−1. We grew a variety of garden vegetables at each of these sites, including different root crops (crops where the edible portion grows directly in the soil). Root crops were the only crops affected by soil lead with carrots taking up more lead than radishes, beets and sweet potatoes. In more or less neutral, sandy soils (loamy sand with 72.2 % sand, 25.6 % silt and 2.2 % clay) with lead concentrations ranging from 200 to 250 mg kg−1, lead concentrations in root crops exceeded the WHO/FAO recommended maximum level (ML) of 1–1.5 mg/kg (dry weight) in both plots with and without added compost . The compost used was a locally available horse manure mix with soil. However, in non-sandy soils with more or less neutral pH and soil lead concentrations of about 250–350 mg kg−1, root crops had lead concentrations below the WHO/FAO recommended maximum level. Lead concentrations in swiss chard (a leafy green) and tomato (a fruiting vegetable from a test plots established in an urban garden, with and without compost addition, are shown in Fig. 3. Vegetables were cleaned mimicking home kitchen procedures (using running tap water) and cleaned more thoroughly using laboratory cleaning procedures (using deionized water and a surfactant). The higher Pb concentrations in the kitchen cleaned crops are likely from dirt particles that did not wash off. Addition of compost to the soil lowered the Pb concentration in both crop types (Fig. 6).

Fig. 6
figure 6

Comparison of vegetable Pb concentrations with maximum allowable levels (ML) of Pb in vegetables as recommended by WHO- CODEX/FAO. (a) Swiss chard in 2010; ML = 5.0 mg kg-1-dry weight, moisture content 94 %. (b) Tomato in 2010; ML = 1.6 mg kg-1-dry weight, moisture content 94 % (Adapted from Attanayake et al. 2014)

Full size image

Arsenic concentrations in soils from the test sites ranged from 50 to 130 mg kg−1. Arsenic uptake by all crop types was low. There are no WHO/FAO MLs for As in vegetables. Compost amendment reduced As concentrations in vegetables by about 46–80 % (Defoe et al. 2014).

We also tested for plant uptake of polycyclic aromatic hydrocarbons (PAHs ). PAH uptake by all crop types was non-detect with total PAH concentrations in soil ranging up to 23–50 mg kg−1 (Attanayake et al., 2015).

Soil pH Can Influence Soil Contaminant Transfer

Adjusting pH to around neutral will reduce the mobility of cationic metals such as Pb, Cd, Zn, Cu, etc. and lower the potential for plant uptake. Unlike Pb, As exists in the soil solution mainly as an oxyanion, and sorption of As decreases with increasing soil pH. So for soils where As is a concern, the pH should not be adjusted to values over 6.5 to avoid enhanced arsenic mobility (Pierce and Moore 1982; Raven et al. 1998; Dixit and Hering 2003). This may not be critical as our studies have shown very low As crop uptake. Maintaining a soil pH around 6.5 would be favorable for multiple contaminants.

The Nutrient Phosphorus, Needed for Healthy Root Growth and Flower Production, Will Transform Metals (More Specifically Lead ) into Their Phosphate Forms and Reduce Their Bioavailability

Lead phosphates, and in particular pyromorphites, are one of the most stable forms of Pb in soils under a wide range of environmental conditions (Lindsay 1979; Nriagu 1984). The conversion of more soluble Pb to pyromorphite when soils are amended with apatite or soluble inorganic forms of P has been documented by many researchers (Ma et al. 1993; Laperche et al. 1996; Cotter-Howells and Caporn 1996; Hettiarachchi et al. 2001). Composts can contain significant quantities of P. We have compared the soil P concentrations measured using a standard soil fertility test with the bioaccessible Pb measured using modified Ruby et al. (1996). The figure shows that higher soil P concentrations are associated with reduced Pb bioaccessibility. Studies have also shown reduced plant Pb uptake when apatite or triple superphosphate (TSP) addition to Pb contaminated soils (Laperche et al. 1997; Brown et al. 1999; Hettiarachchi and Pierzynski 2002) (Fig. 7).

Fig. 7
figure 7

Relationship between percentage of bioaccessible Pb and soil Mehlich III-P in the <2 mm size soil fraction 16 days after compost addition (Attanayake et al. 2014)

Full size image

Soils May Be Impacted by More Than One Contaminant and a Mixture of Amendments (Compost , Phosphorus, Biosolids ) Would Be Beneficial

In some cases soils are contaminated by more than one contaminant. When Pb and As are both elevated in a soil, use of phosphorous amendments to reduce the Pb availability in soils, has been found to increase the availability of As (Li et al. 2014; Manning and Goldberg 1996; Smith et al. 2002). Arsenate and phosphate (PO4 3−) are considered chemical analogues, because the respective pentavalent oxyanions (H2AsO4 −/HAsO4 2− and H2PO4 −/HPO4 2−) are similar in structure, chemical reactivity, and sorption patterns (Antelo et al. 2005; Guan et al. 2009). So, addressing urban soils co-contaminated with Pb and As presents a huge challenge. P-based treatment combinations or soil amendments yielding positive reductions for Pb, may increase the mobility and availability of As. In cases where co- contamination is a concern, using combinations of amendments, such as compost or biosolids based soil amendments , will both reduce the total concentration and may also reduce the bioaccessible fraction of multiple elements (Brown et al. 2012; Defoe et al. 2014).

Conclusions

Research indicates that the potential exposure pathway of concern is direct exposure of humans to contaminated soils. The pathway from contaminated soil to plant to human is insignificant. Research has also shown that, in general, concentrations of lead , arsenic and PAHs in vegetables harvested at test sites were low and contaminants can be diluted by the addition of compost . Compost additions help reduce contaminant concentration in vegetables and also bioaccessible Pb and As. Roots tend to accumulate Pb. If in doubt regarding potential soil contaminants and their respective concentrations growing root crops should be avoided. We believe common sense measures, such as washing crops thoroughly prior to consumption to get rid of adhering soil particles, washing hands thoroughly after gardening, using a mulch to cover bare soil, keeping soil moist during dry and windy conditions to prevent dust generation, making sure no soil gets tracked into the house on shoes and/or clothing, and supervising children in the garden, are all effective, preventative measures to ensure safe gardening/growing.

Polycyclic Aromatic Hydrocarbons (PAHs) are a group of carcinogenic organic pollutants. Urban soils may contain high concentrations of PAHs originated from partial burning of fossil fuels and industrial emissions. They are also significant components in creosote which has been used to treat railroad ties (USEPA 2011). A field experiment conducted in a garden site located in Indianapolis, IN with elevated concentrations of Pb, As and PAHs and total of EPA’s 16 priority PAHS were ranged from 23 to 50 mg kg−1 in 2011 and 2012. The vegetables grown were collard greens, tomatoes, and carrots. Soil treatments used were non-amended control, composted biosolids , non-composted biosolids, mushroom compost , and leaf compost at the rate of compost added was 44 kg m−2. Fresh samples of lab cleaned carrots and tomatoes from the 2011 test plots and lab cleaned collard greens from the 2012 test plots were used for this analysis. The QuEChERS (“Quick, Easy, Cheap, Effective, Rugged, and Safe”) PAHs extraction method proposed by Anastassiades et al. (2003) was used as modified by Slizovskiy et al. (2010). Extracted PAHs were analyzed by GC-MS (Varian Inc., Foster City, CA) for all the 16 priority PAHs . Regardless of the treatments, the concentrations of PAHs measured in the vegetables were low or non-detectable (Attanayake et al. 2013) (Fig. 8).

Fig. 8
figure 8

Monon garden site, Indianapolis, IN

Full size image

In the context of gardening, it can be hypothesized that dermal absorption (skin contact with contaminated soil) could be a significant pathway of transferring soil PAHs to humans. Attanayake et al. (2013) conducted a study to evaluate potential for transfer of soil PAHs to humans via skin; specifically, the effect of soil matrix, the effect of soil moisture content, and the effect of aging of PAHs in soil on dermal transfer of PAHs . The treatments were soil contaminated with PAHs at two moisture contents (20 % and 40 % (v/v)), soil contaminated with PAHs  + biosolids , control soil, control soil spiked with PAHs (0, 1, 3 and 12 days of incubation after spiking), and PAHs dissolved in methanol. The concentrations of the 16 priority PAHs were the same for all the treatments, except for the biosolids and the control soil treatments. Two approaches were taken: (1) an in-vitro steady fluid experiment to evaluate the potential for transfer of PAHs from soil to blood through skin, and (2) a fluorescent microscopy study to determine the penetration depths of PAHs in skin. Only low (152–178 g mol−1) and medium (202 g mol−1) molecular weight PAHs in methanol transferred through skin. The PAHs in any of the contaminated soil treatments did not transfer through skin. In the PAHs spiked control soil at 0th or 1st or 3rd day of spiking, several low molecular weight PAHs transferred through the skin. In contrast, spiked soil incubated for 12 days after spiking, did not show detectable levels of PAHs transferring through the skin. Fluorescence microscopy study revealed that the PAHs in those treatments penetrated to a certain degree in the epidermis. Most of the PAHs that entered the skin accumulated in the stratum corneum. It appears that soil matrix and aging of PAHs in soil restricted transfer of soil PAHs from soil to humans via skin.

figure a

Light microscope image of human skin (abdomen) showing cross section up to depth of dermis.