The urban soil environment is typified by human disturbance. Some of the more visible disturbances are construction projects, cutting and filling to make building sites, and garbage dumps. Other less apparent disturbances are caused by automobile exhaust, urban dust and manufacturing or building emissions, which contribute a significant amount of contaminant material to the soil. These human inputs can greatly change the ways that soils must be managed. In an effort to meet the needs of the urban soil managers, Cornell University, in partnership with the Natural Resource Conservation Service and the New York City Soil and Water Conservation District, began a pilot study in this survey area to test the methods and procedures of predicting heavy metal distribution in urban soils.

Heavy metals are found naturally in undisturbed soils and, in fact, small amounts of many metals are required by plants to remain healthy. The heavy metals found in soils come from three major sources: (i) the rocks from which the soil is formed; (ii) the atmosphere which carries small heavy metal particles from exhaust, emissions, and other soils to later be deposited on land; and (iii) the disposal of material contaminated with heavy metals, such as occurs with garbage dumps and with polluters dumping waste on the side of a road. This last source can create patchy points of contamination which are of high concern and pose possible dangers to people in contact with the contaminated soils.

A number of chemical reactions control the behavior of heavy metals in soils. In technical terms, these reaction mechanisms include cation exchange, specific adsorption, co-precipitation, and organic chelation (Alloway, 1995). These reactions control the exchange of heavy metals between the soil particles and soil water. Most heavy metals are stored in the soil, usually attached to organic matter and clay. Once in the soil water, heavy metals may move within the soil profile, are available to plants, and can be leached into groundwater.

Large concentrations of heavy metals in soil present a number of concerns to local citizens. These concerns include: (i) toxic effects on plants resulting from uptake of heavy metals by garden and yard plants grown in contaminated soil; (ii) eating contaminated plant tissues as well as eating and breathing of soil and dust, particularly by children; (iii) the possible harmful effect of direct skin contact with metals (iv) pollution of water resources; (v) fire and explosions, and (vi) chemical attack on building materials (ICRCL, 1987).

The objective of the heavy metal component of this soil survey is to determine the presence and concentration of 18 heavy metals within the demonstration area (Fig. 21). The metals being identified are aluminum, boron, cadmium, calcium, chromium, cobalt, copper, iron, lead, magnesium, manganese, nickel, phosphorus, potassium, sodium, sulfur, titanium, and zinc. The study will determine the changing concentrations of heavy metals from the surface to a depth of 90 centimeters (cm) within individual soils as well as the concentration differences between different soils within the demonstration area. The local scope of the study will determine the ability of conventional soil mapping to provide sufficient knowledge of concentrations of heavy metals within the mapped soils.

The survey is complicated due to the nature of the human disturbance which resulted in the distribution of heavy metals. Dumping contaminated materials, such as paint, oil, and household cleaners, often results in patchy and irregularly placed heavy metal contamination in places that are hard to detect or locate. It is common for great differences to exist in heavy metal concentration even within a small area. More uniform distributions of heavy metals occurs from sources such as automobile exhaust, manufacturing and building emissions, and urban dust blown by the wind. These inputs increase heavy metal concentrations evenly on soils over a large area, so their concentrations are fairly easy to find, describe, and predict. This mixture of disturbance patterns, patchy and uniform, requires careful and extensive soil sampling. The sampling procedure will provide information about heavy metal concentrations across the landscape and throughout the soil profile, not just at one point in the soil. To meet this need, we are using linear transects, nested sampling, semi-stratified and stratified grid samples to determine the heavy metal concentrations on this changing landscape. By focusing on a local area, we hope to determine the best sampling protocol for each heavy metal which can then be applied to the larger urban area.

Figure 21. The concentration of lead in the soil surface of a section of the survey area.

To illustrate these distribution patterns, Figure 21 shows the surface soil lead contamination in the survey area. The map was developed by plotting sample concentrations at each sample site and developing concentration gradients between the sites. The points on Figure 21 represent the sampling sites, the lines show either roads around the park or trails in the park, and the different colors show relative concentrations of lead. In this case, the highest concentration of 9003 parts per million (ppm) of lead is found on the right (eastern) side of the map, while the lowest concentration of 6 ppm is shown at the lower right (north central) section of the map. This information is extremely important as it allows users to predict heavy metal concentrations in the soilAnother important piece of information available from the samples is the behavior of heavy metal concentration within an individual soil. Penwood soils are relatively undisturbed, and the majority of contaminant heavy metals in these soils have been deposited from the atmosphere. In Figure 22, the bar graph examines the ppm of manganese and lead at different depths in a Penwood soil. The bottom bars of Figure 22 show the average concentration from the top of the soil to a depth of 30 cm. The remaining rows show the concentrations found at specific depths.

Figure 22. Distribution of lead and manganese in the Penwood soil.

The results from Figure 22 become clear when one knows the likely concentrations of these heavy metals in an uncontaminated "natural" soil. Heavy metals come from parent materials, and their concentration will typically increase with soil depth due to two processes. One is soil formation from rock parent material which have higher concentrations of heavy metals; the other is deposition of materials at the soil surface, such as plant organic matter, which typically contain low heavy metal concentrations. As the soil materials from these two processes mix, a gradient of increasing concentration is created through the soil profile.

This pattern of increasing heavy metal concentration with increasing depth of the soil can be seen by the manganese concentration in Figure 22. At 0 to 5 cm the concentration is 149.9 ppm, while at 60 to 90 cm the manganese concentration is 293.8 ppm. Lead has a different distribution in urban soils, which is shown in Figure 22 where lead decreases with depth. While lead is found in undisturbed soils at concentrations of less than 20 ppm (Davies, 1995), lead in Penwood soils was found to have a concentration of 152.6 ppm at 0 to 5 cm. A depth of 60 to 90 cm had a concentration of 7.8 ppm. The decrease of lead concentration with increasing depth can be explained by its deposition from the atmosphere. Automobile exhaust, urban dust, and atmosphere deposition create concentrations of lead on the surface greater than the amount contributed by the rock parent material.

Once the distribution and concentration of heavy metals at each depth within a soil is determined, it will be possible to determine optimum plant rooting depths by park managers or soil management needs for children's playgrounds where there is a worry of soil with high levels of lead. The local focus allows concentrated analysis of soil information as well as the examination of the effectiveness of the sampling methods. This information allows us to determine if conventional soil mapping techniques can be used to determine heavy metal concentration throughout New York City. If the results from the heavy metal component can be added to the entire soil survey, they will provide added important information for development and application of urban land management strategies. Additionally, through experimentation with the sampling methods the study has used, we will be able to determine the most effective approach to find and identify these 18 heavy metals in the patchy urban soil environment. Once the distribution of heavy metal in each soil is determined locally, it may be possible to predict the overall concentration of heavy metals in each specific soil using a limited number of samples. Understanding the distribution of heavy metals in the landscape and within each specific urban soil will increase the effectiveness of sampling and analysis of that soil, and thus decrease the investment in time and money while still meeting the specific needs of urban planners, scientists, and managers. Knowledge of the heavy metal concentrations of urban soils is an important first step towards effective landuse planning.