Spatial Variability in the Speciation of Lead (Pb) and Other Metals Across Urban Lawns Is Linked to Post-Deposition Weathering Reactions

Abstract

The historical use of lead (Pb) poses ongoing health risks via exposure to contaminated urban soils. However, there is limited information about heterogeneity in Pb speciation and distribution at the house lot scale. This study determined highly spatially resolved Pb and other metal speciation along horizontal transects and vertical soil cores from three homes in the Akron, Ohio (USA) municipal. Solid phase characterization was coupled with a sequential extraction protocol to determine operationally defined speciation (exchangeable (M_EX), reducible (M_RED), oxidizable (M_OX), and residual (M_RES)). Lead and Zn were strongly correlated across all fractions (R² = 0.92). Total extractable Pb and Zn were found in low weight percent concentrations nearest to the homes, and speciation was dominated by M_EX and M_RED. High Pb in the M_EX fraction was correlated with the presence of Pb-bearing paint chips in the soil. Lead in the M_EX fraction in soils near the homes decreased with increasing time due to exterior renovations coupled with increases in Pb and Zn in the M_RED fraction. These results suggest that homes are the dominant source of Pb and Zn due to the weathering of exterior surfaces and highlight the acute risk of exposure to more labile Pb immediately following exterior renovations and damage to home exteriors in areas of older housing stock.

1. Introduction

The industrial adoption of lead (Pb) in the United States was primarily driven by its inherent properties, particularly for use in paint production, where it was valued for its ability to enhance the brightness and durability of paints, and for use as an anti-knocking additive in automobile gasoline to improve engine performance [1]. The extensive use of Pb in the United States from 1921 to the beginning of the Pb phase-out in 1973 resulted in widespread environmental Pb exposure [2]. Exposure to Pb has been linked to significant health concerns, such as neurological abnormalities, kidney and cardiovascular damage, and anemia [3,4], with adverse health effects being pronounced in children (through age five). Although there is no established safe blood lead level (BLL) [5], the Center for Disease Control and Prevention (CDCP) uses a BLL reference value of 3.5 µg/dL to identify children with elevated BLLs. Although BLL values in children have decreased from historical highs in the late 1970s (as high as 15 ug/dL) [6] as a result of the phasing out and banning of Pb-based paint and gasoline (from the 1970s to 1990s), over half a million children in the United States were reported to have BLL values above 3.5 µg/dL in 2012 [7], and more than half of all children in the country had a BLL above 1 µg/dL between 2018 and 2020 [8]. However, the risk of exposure is not equally distributed among the population, with black children making up more than 50% of children with a BLL greater than 1 ug/dL [9], underscoring the use of high Pb levels as a socioeconomic indicator of inequality.

The use of Pb-based products, particularly in urban environments, resulted in the wide-scale contamination of soils, a primary source of Pb exposure to children, with weight percent soil concentrations well over the U.S. Environmental Protection Agency’s (EPA) actional level of 200 ppm being reported across older urban cores [10,11,12,13]. Although the Pb concentrations in soils have declined over time, there is leftover Pb in the environment due to previously deposited Pb in the soil and newer inputs from the degradation of Pb-based paint chips that continue to be added into soils. Although previous studies have typically examined the distribution of total Pb in urban soils, there has not been a comprehensive examination of speciation (i.e., the chemical and physical form(s) in which Pb is present in the environment) [14], which is crucial to establishing exposure risks [15].

Current Pb speciation and distribution in soils likely reflect a complex relationship between the input of primary sources and post-deposition transformations [16]. With respect to Pb speciation in primary sources, Pb in paint and gasoline exists in inorganic and organic forms, respectively. Lead-based paint can include Pb oxides, Pb sulfates, and carbonates [17], and as the paint weathers and breaks down on the exterior of homes, Pb(II) ions are released into the soil [18]. Despite Pb paint being banned, much of the older housing stock in urban areas is still covered with Pb paint, with at least 34.6 million homes (30% of US homes) having Pb-based paint and 18.2 million homes (15% of US homes) having deteriorated Pb-based paint. These homes were built primarily before 1978 in the US Northeast and Midwest regions. The paint can chip off through the weathering of exterior surfaces and/or renovations [15,19], releasing Pb(II) into soils. In terms of leaded gasoline, the combustion of the organo-lead compound tetraethyllead results in the release of Pb(II) into soil [20,21]. The Pb(II) released from the breakdown of paint and gasoline can then react with and form a range of labile to recalcitrant pools of soil Pb with different chemical and/or physical properties [17,22,23,24,25]. The resulting differences in reactivity impact potential exposure, with ingestion and inhalation being the primary exposure pathways, especially for children who engage in hand-to-mouth behavior [26,27]. Although the phasing out and ban of Pb-based gasoline, pipes, and paints resulted in a direct decrease in the BLL of children aged 1–5 years [8], a large pool of labile Pb still exists in urban soils. Furthermore, the risk of exposure to Pb depends on Pb speciation and its concentrations in soil, which can change over time.

The Pb(II) ions released following the deposition of primary phases are typically retained in soils [11,28] and sequestered by metal oxides, clays, and organic matter [29]. However, as weathering progresses, the chemical form of Pb(II) is expected to change due to its increasing degree of transformation. It is unclear how this transformation impacts its chemical form with respect to an increasing distance from the site of deposition. An increase in Pb solubility can lead to a rise in its potential bioavailability, which can vary spatially and temporally depending on the nature of transformation it undergoes. After Pb(II) compounds are introduced into the soil, the resultant particles undergo translocation across both horizontal and vertical distances from their initial point of deposition. Furthermore, these particles are subject to a range of soil formation processes that facilitate their physicochemical transformation [11] and association with different minerals in the soil. Identifying the current speciation of Pb along the vertical and horizontal transects can provide valuable insights into the spatial changes in soil Pb speciation after deposition. Indirect methods to determine metal speciation, such as sequential extractions, can be more cost-effective and efficient compared to direct methods, such as synchrotron-based X-ray Absorption Spectroscopy, in determining the distribution of metals with operationally defined fractions including exchangeable, reducible, oxidizable, and residual fractions [30,31,32,33,34,35].

Current metal speciation and the fraction of potentially bioavailable Pb and other metals are therefore the result of a complex set of interactions, including (1) the type of source(s) and rate and timing of input; (2) local mineralogical and (bio)geochemical conditions which control which metal-bearing species can be formed; and (3) the rate of metal loss through soil loss and erosion, which might result in the preferential transport of certain metal-bearing phases over others. The summation of these interactions can potentially result in heterogeneous metal speciation across an urban landscape that can change in space and time. Previous studies of Pb speciation, typically based on acid-extractable metal concentrations, provide a coarse picture of Pb speciation at the city scale. This work aimed to determine the spatial distribution of soil Pb species and other metals along vertical and horizontal transects adjacent to homes with known historical Pb paint use and to determine potential changes in speciation over time following deposition. Furthermore, the sampling density used here provided a higher resolution of changes in metal speciation across urban lawns compared to previous studies that typically take one sample per lot, allowing for the observation of previously unknown local-scale heterogeneity. This work was based in Akron, OH (USA), using a sequential extraction method combined with solid phase characterization of the soils for bulk properties and electron microscopy for targeted phase identification. This approach aimed to provide valuable insights into the environmental processes that affect the mobility and fate of Pb in the soil and identify an approach to be used for constraining the scope of targeted remediation.

2. Materials and Methods

2.1. Site Description and Sample Collection and Processing

Soil samples were collected in proximity to three residential homes in and around Akron, OH (USA) (Figure 1 and Figure S1), a metropolitan area with a population of approximately 190,000 [36]. Akron rapidly increased in population and industrial growth (dominated by the rubber and tire industries), and its population peaked at 290,000 in the 1960s. Soil samples were collected in areas dominated by older (pre-1978) housing stock [37]. Although BLL data are limited for Akron and Summit County, data from 2012 through 2014 indicate that 4.1% of tested children had elevated BLLs in comparison to the national average of approximately 2.5% around the same period [26]. The homeowners of all three homes confirmed that Pb-based paints had been used on the exterior of the houses; however, the paint was removed as part of renovations at the three Sites 2, 11, and 25 years before soil sampling occurred, respectively. From each house, three sets of samples were collected in June 2021 as follows (and summarized in Table S1): (1) along a horizontal transect collected at 1 m intervals from areas immediately adjacent to the house to the road verge, excluding areas inaccessible due to existing road pavement and sidewalks (n = 16, 10, and 13 for Sites 1, 2, and 3, respectively); (2) a horizontal transect from an area immediately next to the back of the house toward the end of the property (n = 13, 3, and 2 for Sites 1, 2, and 3, respectively); and (3) a vertical core collected from an area immediately adjacent to the front of the house at 10 cm internals (n = 5 and 4 for Sites 1 and 3; no core was collected at Site 2). For surface samples, a 2 cm deep soil sample was collected after clearing any surface vegetation that may have been present. Each sample was sieved (2 mm) to remove any debris and large rocks. Samples were dried at 40 °C for 12 h using a Binder ED-115 Oven (Binder Inc., Bohemia, NY, USA). Approximately 25 g of each sieved sample was milled using a SPEX ball mill 8000 M (Cole Palmer, Metuchen, NJ, USA) with tungsten carbide balls for 10 min to produce a homogenous powder. The milling canister was cleaned using isopropyl alcohol between runs to prevent cross-contamination. Milled soil samples were used for bulk mineralogical characterization and sequential extractions (described below), and sieved-only samples were analyzed by electron microscopy. The pH of the topsoil closest to the three homes was measured using 1 g of soil in 11 mL of DDI-H₂O (Table S2); the average soil pH values for the three sites were 5.98 ± 0.03, 5.60 ± 0.14, and 5.82 ± 0.06, respectively (Table S2).

2.2. Sequential Extraction of Pb and Other Metals from Soils

A four-step sequential extraction method based on the BCR method [38] and modified to target a range of potentially bioavailable metals and trace elements in urban soils [31] was used to determine metal speciation for the following fractions and steps using 1 g of milled soil: (1) exchangeable: 20 mL of 0.11 mol/L acetic acid at 22 °C for eight hours; (2) reducible: 20 mL of 0.11 mol/L hydroxylammonium chloride at 22 °C, followed by centrifugation and acidification with 2 mol/L nitric acid; (3) oxidizable: 10 mL of 8.8 mol/L hydrogen peroxide added twice at 85 °C, followed by extraction with 50 mL of 1 mol/L ammonium acetate, and (4) residual: aqua regia solution containing 7.0 mL of concentrated nitric acid and 2.3 mL of concentrated hydrogen chloride (HCl). The resulting solution was extracted for 1.5 h at 85 °C. After each step, the samples were centrifuged at 3000 rpm for 20 min. The decanted solutions underwent ×10 dilution for fractions 1 and 2 and ×20 dilution for fractions 3 and 4 prior to a Thermo Scientific iCAP PRO XP inductively coupled plasma optical emission spectroscopy (ICP-OES, Thermo Fisher Scientific, Waltham, MA, USA) analysis for the following elements, with their detection limits noted in parenthesis: Al (0.1 μg/L), S (0.5 μg/L), K (0.5 μg/L), Ca (0.3 μg/L), Ti (0.3 μg/L), Cr (0.3 μg/L), Mn (0.3 μg/L), Fe (0.2 μg/L), Co (0.3 μg/L), Cu (0.2 μg/L), Zn (0.1 μg/L), Cd (0.3 μg/L), and Pb (0.3 μg/L). A total of six standards, including a blank solution, were prepared to calibrate the concentrations. An internal Y standard was used to correct for plasma instability. The ICP-OES standards prepared for the analysis of fractions 1 and 2 had concentrations of 0.01 ppm, 0.1 ppm, 1 ppm, 10 ppm, and 30 ppm, respectively. For each site, five to ten randomly selected samples were treated as duplicates; the differences between duplicate sub-samples and repeated standards were lower than 5% for all elements. All reagents used were analytical grade, and solutions were prepared with distilled deionized water (DDI-H₂O) (18.2 MΩ; Milli-Q Direct-Q 3UV-R, Millipore Sigma, Burlington, MA, USA). A Pearson correlation analysis was performed (see Excel file in the electronic annex) to investigate the correlation between extracted metals and other soil properties as described below.

2.3. Solid Phase Characterization

Mineral identification and abundance were determined by X-ray diffraction using a Rigaku Miniflex 6G (Rigaku, The Woodlands, TX, USA) X-ray diffractometer (3–90°, 0.02° step size; 2° min⁻¹) of milled powders. Phase identification was performed using Rigaku’s PDXL software v.2.9, with the Whole Pattern Powder Fitting (WPPF) method connected to the ICDD PDF-2 database. Loss on ignition (LOI) was used as a proxy for organic matter content [39,40] following a previously established procedure [41]. A weighed mass of sub-sample with a target mass of 2.5 g was placed in ceramic crucibles with lids and brought to 550 °C using a Thermo Fisher Thermolyne muffle furnace (Thermo Fisher Scientific, Waltham, MA, USA) for one hour and then allowed to cool before being re-weighed. Scanning Electron Microscopy with Energy Dispersive Spectroscopy (SEM-EDS) was used to determine the composition and morphology of Pb-bearing particles in the soils sampled near the homes using a Hitachi Benchtop Scanning Electron Microscope TM3030 equipped with Quantax70 Energy Dispersive X-ray Spectrometry using dry powders deposited on carbon sticky tape (Hitachi USA, Santa Clara, CA, USA). A MalvernTM Mastersizer 2000 laser diffraction particle size analyzer equipped with a Hydro 2000MU manual inlet system (Malvern Panalytical Inc., Westborough, MA, USA) was used to conduct a particle size analysis using sieved (<2 mm) but not milled samples. The particle size distribution was binned to determine the contribution of sand- (63 µm–2 mm), silt- (2 µm–63 µm), and clay-sized (<2 µm) particles.

3. Results

3.1. Metal Speciation in Soils in Horizontal Transects and Depth Cores

Metal speciation and distribution are discussed below for metals with the highest levels of contamination (Pb and Zn) and other industrial metals potentially related to anthropogenic activities (first-row transition metals plus Cd); changes in the speciation and distribution of elements less likely to be anthropogenically impacted (Al, S, K, Ca, and Ti) are reported in the Supporting Information document (Figures S2–S6). Metal speciation is reported in terms of the total extractable concentration ([M]_T-EX), which is the sum of metal concentrations in the exchangeable ([M]_EX), reducible ([M]_RED), oxidizable ([M]_OX), and residual ([M]_RES) fractions.

3.1.1. Pb and Zn

The distribution of Pb in topsoil along horizontal transects in front of each home exhibited similar overall trends at all three homes, with a maximum [Pb]_T-EX closest to the homes, which rapidly decreased within two to three meters (Figure 2a). The maximum [Pb]_T-EX values for the three homes were 21,720 mg/kg, 11,115 mg/kg, and 1854 mg/kg, respectively. At Site 1, the maximum [Pb]_T-EX value was located at a 1 m distance, whereas the maximum [Pb]_T-EX value at Sites 2 and 3 were located immediately adjacent to the homes (i.e., at 0 m). At all three sites, the [Pb]_T-EX values remained relatively constant beyond two to three meters from the houses, with average values of 166 ± 68 mg/kg, 1422 ± 781mg/kg, and 243 ± 198 mg/kg, respectively. A noticeable deviation to the flat trend across the lawn was observed at Site 1, where a spike in [Pb]_T-EX of 548 mg/kg was observed at the end of the transect near the street curb. Although fewer topsoil samples were collected along the horizontal transect from the back of each home, it is clear that these soils exhibited a similar overall trend compared to the front, with the highest [Pb]_T-EX values being found immediately adjacent to the homes, which decreased quickly within 2–3 m. The maximum [Pb]_T-EX values for the three homes were 381 mg/kg, 6620 mg/kg, and 1368 mg/kg, respectively (Figure 2b). The average values for [Pb]_T-EX further from the homes were 215 ± 69 mg/kg, 129.51 mg/kg at the only remaining point, and 174 ± 119 mg/kg, respectively. Within the vertical soil core (Figure 2c) at Site 1, the highest [Pb]_T-EX value (2826 mg/kg) was observed in the shallow sample (0–10 cm) and decreased steadily with depth to 454 mg/kg. At site 3, [Pb]_T-EX did not exhibit a trend with depth, and the average value was 2186 ± 225 mg/kg. Lead speciation in topsoil from the horizontal transect in the front of each house was dominated by the reducible fraction (Pb_RED) closest to the home and the residual fraction (Pb_RES) further from the homes (Figure 2d–f) with a relatively linear change in each fraction across the transect. The maximum percentages of Pb_RED were found closest to each house, with values of 74%, 69%, and 75%, respectively, which decreased to 40%, 32%, and 29%, respectively, closest to the street. The minimum percentages of Pb_RES closest to each house were 5%, 13%, and 9%, respectively, and increased to 48%, 44%, and 55%, respectively, closest to the street. Lead in the exchangeable fraction was the highest nearest the houses, with maximum percentage Pb_EX values of 25%, 14%, and 4%, respectively. No apparent trend in the contribution of Pb in the oxidizable fraction (Pb_OX) was observed, with average Pb_OX contributions of 13 ± 9%, 9 ± 3%, and 12 ± 7% being found on each respective site. Lead speciation in topsoil from the transect at the back of the houses (Figure 2g–i) did not exhibit the same clear trends compared to the front of the houses, although Pb_RED was the dominant fraction at the back of the houses. The speciation of Pb in the vertical cores from Sites 1 and 3 did not exhibit a significant trend with depth (Figure 2j,k), with average percentage values for Pb_EX, Pb_RED, Pb_OX, and Pb_RES of 8 ± 3%, 61 ± 8%, 13 ± 8%, and 18 ± 9%, respectively, for Site 1, and 17 ± 5%, 64 ± 4%, 6 ± 1%, and 14 ± 2%, respectively, for Site 3.

The distribution pattern of Zn in the soil along the horizontal transects in front of each home displayed similar trends across all three homes. Maximum [Zn]_T-EX values were observed immediately adjacent to the homes, followed by a rapid decline within two to three meters (Figure 3a). The maximum [Zn]_T-EX values for the three homes were 2525 mg/kg, 1120 mg/kg, and 609 mg/kg, respectively. At Site 1, the maximum [Zn]_T-EX value was located at a 1 m distance, whereas the maximum [Zn]_T-EX values at Sites 2 and 3 were located immediately adjacent to the homes (i.e., at 0 m). At all three sites, the [Zn]_T-EX values remained relatively constant beyond two to three meters from the houses, with average values of 624 ± 596 mg/kg, 399 ± 256 mg/kg, and 268 ± 118 mg/kg, respectively. A noticeable deviation to the uniform trend across the lawn was observed at Site 1, where a spike of 568 mg/kg in the [Zn]_T-EX value was observed at the end of the transect near the street curb. In the horizontal transects collected from the rear of each house, the highest [Zn]_T-EX values were observed immediately adjacent to the homes, which decreased quickly within 2–3 m. The maximum [Zn]_T-EX values for the three homes were 669 mg/kg, 779 mg/kg, and 1268 mg/kg, respectively (Figure 3b). The average values for [Zn]_T-EX further from the homes were 408 ± 107 mg/kg, 345 mg/kg at the only remaining point, and 645 ± 540 mg/kg, respectively. Along the vertical transect (Figure 3c), the [Zn]_T-EX values at Sites 1 and 2 did not exhibit apparent trends and averages at 654 ± 178 mg/kg and 534 ± 14 mg/kg, respectively. Along the front horizontal transect, Zn speciation in the topsoil was dominated by the exchangeable fraction closest to the houses, but once the Zn concentration dropped off within 2–3 m, the residual fraction dominated (Figure 3d–f). The minimum percentages of Zn_RES closest to each house were 20%, 56%, and 62%, respectively, and they increased to 93%, 96%, and 93%, respectively, closest to the street. Zinc in the exchangeable fraction was the highest nearest the houses, with maximum percentages of Zn_EX values of 43%, 27%, and 12%, at the three homes, respectively. The contribution of Zn_RED decreased with an increasing distance from the homes, averaging 12 ± 14%, 6 ± 3%, and 3 ± 3% at Sites 1, 2, and 3, respectively. No apparent trend in the distribution of Zn in the oxidizable fraction was observed, with average percentages of Zn_OX values of 7 ± 7%, 6 ± 5%, and 6 ± 4%. Zinc speciation in topsoil from the transect in the back of the houses (Figure 3g–i) did not exhibit the same clear trends compared to the front of the house, although Zn_RED was the dominant fraction in the back of the houses. Along the vertical core at site 1, from 0–10 cm to 40–50 cm, the percentage contribution of Zn_EX decreased from 18% to 3% with depth as the percentage contribution of Zn_RES increased from 34% to 72%, respectively. At Site 3, from 0 cm to 10 cm, 30 cm, and 40 cm, the Zn_EX value increased from 9% to 12% as the Zn_RES value decreased from 77% to 71% with depth, respectively. The percentage contributions of Zn_RED and Zn_OX do not exhibit significant variations with the depth and have averages of 15 ± 11% and 14 ± 7% in site 1 and averages of 6 ± 1% and 10 ± 2% in Site 3, respectively.

3.1.2. Cr, Mn, Fe, Co, Cu, and Cd

In the horizontal transects from all three sites, [Cr_T-EX] did not present conspicuous trends over horizontal and vertical transects (Figure 4). Across all the transects on the sites, the [Cr_T-EX] values averaged at 54 ± 12 mg/kg, 36 ± 7 mg/kg, and 35 ± 5 mg/kg. Across all the transects, Cr_RES dominated the speciation, increasing from a 78% contribution at 0 m to 92% at 18 m, while Cr_OX followed with a decrease from 20% at 0 m to 8% at 18 m at Site 1. In contrast, Cr_EX and Cr_RED contributed less than averages of 0.26 ± 1% and 0.05 ± 1% to Cr_T-EX. This trend was maintained in Sites 2 and 3 (Figure 4d–k). The average speciation contributions in Site 2 were 0%, 2.8 ± 3%, 15.5 ± 8%, and 82 ± 9%, while in Site 3, they were 0%, 1.6 ± 1%, 14 ± 8%, and 85 ± 8%, respectively.

The total extractable Mn in the front horizontal transects in Site 1 was the lowest near the home and relatively constant beyond 5 cm, whereas the [Mn]_T-EX values at Sites 2 and 3 were the highest near the homes and decreased with distance, with higher values for Site 3 being found near the end of the transect. The values of [Mn]_T-EX in the rear horizontal and vertical cores did not exhibit significant trends with distance or depth. The average [Mn]_T-EX values from all samples for the three sites were 1457 ± 315 mg/kg, 910 ± 414 mg/kg, and 1087 ± 319 mg/kg (Figure 5a–c). Along the front transect at Sites 1, 2, and 3 (Figure 5d–f), the Mn_EX proportion decreased with an increasing distance from the homes. The percentage contribution of [Mn]_EX varied from 36% to 19%, from 18% to 6%, and from 27% to 18%, respectively. In contrast, the Mn_RES and Mn_OX proportions remained relatively stable as the distance from the homes increased, with average percentage contributions of 33 ± 7% and 4 ± 1% in Site 1, 18 ± 9% and 3 ± 1% in Site 2, and 10 ± 7% and 3 ± 2% in Site 3, respectively. Across the horizontal back transect of the three sites, Mn_RES saw a slight increase in proportion with an increasing distance from the homes, with average concentrations shifting from 33% to 50%, from 55% to 57%, and from 63 to 85%, respectively. Along the back transect, no apparent trend was observed in Mn speciation (Figure 5g–i). The average percentage contribution of [Mn]_EX across Sites 1, 2, and 3 were 23 ± 7%, 13 ± 5%, and 21 ± 1%, respectively. Mn_RED contributes average percentages of 45 ± 10%, 55 ± 25%, and 47 ± 18% to the Mn_T-EX in Sites 1, 2, and 3, respectively. Along the vertical transect at Site 1, the percentage contribution of [Mn]_EX decreased from 56% at 0 cm to 8% at 50 cm, while [Mn]_RES increased from a 17% contribution at 0 cm to 66% at 50 cm with little change in the contributions of [Mn]_RED and [Mn]_OX with depth. The average percentage contributions of [Mn]_T-EX were 34 ± 12% and 3 ± 1%, respectively. In contrast, in Site 3, the speciation of Mn with depth remained relatively constant. The average concentrations of Mn_EX, Mn_RED, Mn_OX, and Mn_RES were 17 ± 4%, 31 ± 3%, 1.5 ± 1%, and 51 ± 3%, respectively (Figure 5h–j).

The average weight percentages of Fe_T-EX across the horizontal distance for the three homes were 8.9 ± 2%, 8.8 ± 2%, and 7.9 ± 2% at Sites 1, 2, and 3, respectively. In the vertical transect, the Fe_T-EX content increased with depth, ranging from 3.93% at 0–10 cm to 12.9% at 40–50 cm at Site 1 and from 9.7% at 0–10 cm to 10% at 30–40 cm at Site 3 (Figure 6a–c). The speciation of Fe did not significantly change with distance or depth. The contribution of Al_EX across all transects was generally below the detection limit, while Al_RED, AL_OX, and Al_RES averaged at 2 ± 1%, 1.5 ± 2%, and 96 ± 3% (Site 1); 2 ± 1%, 1.6 ± 2%, and 96 ± 2% (Site 2); and 1.6 ± 1%, 1 ± 1%, and 97 ± 1% (Site 3), respectively (Figure 6d–k). In the vertical profile at Site 1, Fe_T-_EX increased with depth from the topsoil, concurrent with the decrease in both Fe_OX and Fe_RED with depth.

The concentration of Co did not exhibit noticeable trends over horizontal and vertical transects. Across the transects on the sites, the Co_T-EX concentration averaged at 37 ± 21 mg/kg, 20 ± 4 mg/kg, and 18 ± 5 mg/kg in Sites 1, 2, and 3, respectively (Figure 7). Cobalt speciation across the front and back transects in the three homes remained relatively constant with an increasing distance from the homes. The average percentage contributions of the exchangeable, reducible, oxidizable, and residual fractions were 11 ± 8%, 15 ± 8%, 9 ± 6%, and 65 ± 16% in Site 1; 9 ± 4%, 7 ± 8%, 8.6 ± 2%, and 75.6 ± 9% in Site 2; and 15 ± 7%, 11 ± 4%, 10 ± 3%, 64 ± 9% in Site 3, respectively.

The concentration of total extractable Cu was the highest near the homes, measuring 186 mg/kg, 93 mg/kg, and 95 mg/kg in Sites 1, 2, and 3, respectively (Figure 8). The trend generally flattened at about 5 cm from the homes, contributing an average of 61 ± 23 mg/kg, 43 ± 3 mg/kg, and 42 ± 5 mg/kg, respectively, between 5 m and the end of the front horizontal transects. Along the back transect, [Cu]_T-EX increased from 60 mg/kg to 99 mg/kg for Site 1 and remained relatively constant for Sites 2 and 3, with average values of 56 ± 12 mg/kg and 66 ± 1 mg/kg, respectively. Compared to Sites 2 and 3, Site 1 showed more variability along the horizontal transects. In contrast, the depth profile does not reveal a substantial change in concentration with depth at Site 1, averaging 54 ± 24 mg/kg, while Site 3 increased with depth from 268 mg/kg at 0–10 cm to 1636 mg/kg at 40–50 cm (Figure 8a–c). Although Cu speciation was dominated by the residual fraction, a significant fraction of Cu was in the oxidizable fraction (Figure 8) in contrast to the other metals. The values of [Cu]_OX were typically the highest near the homes and decreased with horizontal distance ranging from 37% to 10% (Site 1), 11% to 5% (Site 2), and 44% to 9% (Site 3) along the front transect (Figure 8d–f). The contribution of Cu_EX across the front transect of the sites was also the highest within three meters from the homes, with their highest proportions measuring 4.8% (Site 1), 4.6% (Site 2), and 3.7% (Site 3). The contribution of Cu_RED decreased with horizontal distance along the front transect from 7.6% to 2.5% at Site 1, from 8.4% to 1.6% at Site 2, and from 11% to 4.5% at Site 3. The concentration of Cu_EX at the back transect was below the instrument’s detection limit across all homes. However, unlike the front transect, Cu_OX did not show a conspicuous trend, with average contributions of 13 ± 7%, 6 ± 6%, and 18 ± 6% at Sites 1, 2, and 3, respectively. The contributions of Cu_RES along the back transect also averaged at 4± 2%, 4 ± 4%, and 6.7 ± 1%, respectively. Along the vertical transect (Figure 8j,k), the percentage contribution of Cu_RES at Site 1 dominated the speciation and increased from 51% at 0–10 cm to 97% at 40–50 cm, while Cu_OX presented the inverse, decreasing from 47% at 0–10 cm to 2% at 40–50 cm. The reducible fraction contributed an average of 2.5 ± 2%. Conversely, at Site 3, the contribution of Cu_EX to Cu_T-EX increased from 7% at 0–10 cm to 44% at 30–40 cm, while that of Cu_RES decreased from 51% at 0–10 cm to 17% at 30–40 cm. The average contributions of Cu_EX and Cu_OX were 24.8 ± 4% and 15.7 ± 5%.

The average Cd_T-EX across the horizontal and vertical transects was higher in Site 1 (2 ± 2 mg/kg), followed by Site 2 (1.2 ± 1 mg/kg) and Site 3 (0.96 ± 1 mg/kg). Along the front transects, Cd_T-EX was higher between one to three meters from the three homes, gradually decreasing with an increasing horizontal distance from them (Figure 9). Specifically, the Cd_T-EX concentration decreased from 8 mg/kg, 2 mg/kg, and 2 mg/kg to 2 mg/kg, 0.9 mg/kg, and 0.9 mg/kg at Sites 1, 2, and 3, respectively. The residual fraction dominated Cd speciation across all the transects in the homes, while the Cd associated with the exchangeable fraction was only present in Site 1 and concentrated close to the home. The average percentage contributions of the exchangeable, reducible, oxidizable, and residual fractions in the soil were 15.6 ± 16%, 17 ± 19%, 0.4 ± 2%, and 66.7 ± 27% in Site 1; 0%, 3.7 ± 7%, 6.2 ± 13%, and 90 ± 13% in Site 2; and 6.5 ± 14%, 0%, and 93.5 ± 14% in Site 3. In Site 1, Cd speciation along the vertical transect decreased with depth from a 34% percentage contribution between 0 and 10 cm to 11% between 40 and 50 cm in the Cd_EX fraction, while the Cd_RED fraction decreased from an 81% percentage contribution between 0 and 10 cm to 0% between 40 and 50 cm. In contrast, the percentage contribution of the Cd_RES fraction increased from 15% between 0 and 10 cm to 89% between 40 and 50 cm. The residual fraction dominated the speciation of Cd in Site 3 and did not present any change in speciation.

3.1.3. Element Correlation Relationships

Strong significant positive correlations were observed between Pb and Zn (for M_T-EX and across all fractions) and with other metals (Cr, Fe, Co, Cu, and Cd) (electronic annex and Table S2). In contrast, these metals exhibited a strong negative correlation with Mn. Other notable observed correlations included the following: (1) the soil organic content had a strong positive correlation with a range of elements (Mg, S, Ti, Mn, Fe, Co, Cu, Zn, Cd, and Pb) and sand-sized particles and (2) the total extractable sulfur in soils correlated strongly with the oxidizable fraction of a range of metals (Cr, Mn, Fe, Co, Cu, Zn, Cd, and Pb).

3.1.4. Changes in Speciation in Soils as a Function of Time Since Exterior Renovations

The sequential extraction results reveal that in soils immediately adjacent to the homes, the total extractable Pb and Zn (and the percentages of Pb and Zn in the exchangeable fraction) decreased as a function of time since exterior renovations occurred (i.e., when exterior Pb-based paints were removed) (Figure 10a,b). With an increasing time since renovation, Pb_T-EX was 21,700 mg/kg at Site 1, 11,115 mg/kg at Site 2, and 1865 mg/kg at Site 3, and Zn_T-EX was 2525 mg/kg at Site 1, 1120 mg/kg at Site 2, and 608 mg/kg at Site 3. The percentage of Pb_EX was 25% at Site 1, 14% at Site 2, and 6% at Site 3, and the percentage of Zn_EX was 44% at Site 1, 27% at Site 2, and 12% at Site 3. This trend was not observed for other metals, for example, Mn and Cr (Figure 10c,d).

3.2. Soil Solid Properties

3.2.1. Soil Organic Content

At Sites 1 and 3, the LOI values were the highest near the homes and decreased with distance. Their organic content decreased from 29% and 21% closest to the homes to 11% and 6% at the end of the transect, respectively (Figure 11a,c). In contrast, LOI at Site 2 was lower near the home and at the end of the transect, with a spike in LOI to 40% at 6 cm. The LOI values across the transect averaged at 16 ± 10% (Figure 11b). In the back transect, at Site 1, the organic content demonstrated a similar spatial variation. Starting at 4.5% at 0 m, LOI peaked at 5 cm (28%) and decreased with distance to 6% at 15 m. Further away, at 18 m from the home, the organic content measured 12% (Figure 11d). For both Site 2 and Site 3, which had fewer samples, the average LOI measurements were 5.5 ± 2% and 11 ± 2%, respectively (Figure 11e,f). Along the vertical transect, the LOI value was the highest at the topsoil at 43% and 7.2% at Sites 1 and 3, and it decreased with depth to 4.5% and 5%, respectively (Figure 11g,h). In general, higher LOI values correlated with higher concentrations and percentage contributions of Pb_OX. (Figure S7).

3.2.2. Particle Size Distribution and Soil Mineralogy

Across the horizontal transects, the percentage contribution was invariant. Soils from the three sites were dominated by the silt fraction across all the transects, followed by the sand fraction and the clay fraction (Figure 12). The particle size distribution in the front and back transects were similar. The average amounts of sand, silt, and clay in the front and back transects were 42 ± 9%, 54 ± 8%, and 4 ± 1%, for Site 1; 52 ± 5%, 45 ± 5%, and 2.6 ± 1%, for Site 2; and 45 ± 7%, 52 ± 7%, and 3 ± 1%, for Site 3. The only profile that showed a change in distribution was the vertical profile in Site 1, where the sand percentage decreased from 58% (0–10 cm) to 40% (40–50 cm) along the vertical transect, while the clay fraction increased from 2% in topsoil to 6% (40–50 cm). Despite the fact that the sites were in three different parts of Akron, their bulk mineralogy contents were very similar (Figure 13). The soils across the transects in the sites were composed of quartz, feldspars (albite, orthoclase, and anorthite), micas (muscovite, biotite, and paragonite), and clays (illite, kaolinite, montmorillonite, and chlorite). Across the sites, the average compositions of the soils were 57 ± 9% of quartz, 19 ± 10% of feldspars, 14 ± 6% of micas, and 9 ± 9% of clays at Site 1; 69 ± 9% of quartz, 16 ± 10% of feldspar, 16 ± 10% of mica, and 8 ± 3% of clays at Site 2; and 69 ± 7% of quartz, 18 ± 6% of feldspar, 9 ± 5% of mica, and 4 ± 3% of clays at Site 3 (Figure 13). In the topsoil near the Site 1 house, sub-mm white flakes were visually observed. These particles were found to be aggregates that contained irregularly shaped Pb-O-rich grains ranging from 1 to 25 μm (Figure 14). Notably, these particles were absent in deeper soils near the Site 1 house and along the horizontal transect. These grains are compositionally similar to Pb-bearing paint chips collected from the Site 1 house (Figure S8). Furthermore, no such particles were observed at Sites 2 or 3. The Pb-bearing aggregates were observed in a Ti-bearing background. Other Si-Al-O-rich particles, characteristic of common soil (alumino)silicate minerals, were observed forming aggregates with the Pb-bearing phases as well.

4. Discussion

4.1. Dominant Controls on Pb and Zn Speciation and Distribution Around Urban Homes

Lead and Zn were above local background concentrations (52 mg/kg and 91 mg/kg, respectively) [42] in all samples based on the total extractable Pb and Zn concentrations, indicating that Pb and Zn are more impacted by anthropogenic activities compared to other trace metals. With respect to current metal speciation present in the soils (when the soils were sampled in 2021), Pb and Zn were detected in all three labile fractions (exchangeable, reducible, and oxidizable) of the soil samples (Figure 2 and Figure 3), indicating that Pb and Zn have undergone complex (bio)geochemical reactions subsequent to the initial deposition given that typical Pb and Zn sources would not be dominated by the same fractions and/or be in the same proportions within those fractions. With respect to current speciation, the exchangeable fractions of Pb and Zn were shown to be dominated by metals weakly bound to clay minerals and other soil solids and soluble metal-bearing phases [43]. Clay minerals are effective sinks from metals such as Pb and Zn over a wide range of soil conditions [44,45] via the formation of Pb- and Zn-bearing surface sorption complexes [46]. Lead and Zn in the reducible fraction are commonly dominated by Fe-(oxy)hydroxides [47], consistent with our observation of Pb and Zn correlated with Fe (electronic annex). Lead and Zn sorption to Fe-(oxy) hydroxides has commonly been observed in contaminated soils [16]. Manganese oxides are also known to effectively sequester Pb and Zn [48,49]; however, in Akron soils, there is no evidence that the Mn oxides are sequestering Pb and Zn. Manganese was negatively correlated with Pb and Zn (electronic annex), indicating that Mn-bearing phases are likely not a dominant sink for Pb and Zn. Although Mn oxides are recognized as effective sorbents of Pb [48], the results suggest that Fe oxides are the dominant sink for Pb and Zn in this fraction and are outcompeting Mn oxides. It is likely that the sorption of Pb was dominated by Fe oxides in these soils given their higher abundance based on the higher total extractable concentration of Fe compared to Mn oxides based on the average values of 0.1 wt% for Mn_T-EX and 0.9 wt% for Fe_T-EX across all the sites (Figure 5 and Figure 6). Lead and Zn in the oxidizable fraction are typically dominated by solid organic matter [50]. The proportions of Pb and Zn associated with organic matter remained relatively constant throughout the horizontal transect. Flower beds near homes and trees along the horizontal transects correlate with the higher LOI (Figure 7), supporting the observations that increased organic input in the soil offers more binding sites for Pb in the oxidizable fraction [51]. Lastly, Pb and Zn in the residual fraction are likely bound within phosphates and other recalcitrant phases [43,52].

The total extractable Pb and Zn also varied with distance and depth, and the Pb_T-EX and Zn_T-EX concentration distributions align with previous findings showing that the concentrations of Pb and Zn in the soil decreases with an increasing distance from the homes due to the deposition of exterior Pb-based paint from homes into the soils [53]. Furthermore, at Site 1, soil particles consistent with paint chips observed near the home (Figure 14) suggested that the homes are the primary source of Pb. Similarly, zinc roofing sheets, identified as a prominent source of Zn deposition in residential areas [54,55], were used to finish the Akron homes, likely serving as the main source of Zn in the soils as the roofing sheets weather. In contrast, higher concentrations of Pb and Zn near structures have been attributed to the role structures play in trapping fine metal particles during wind transportation from more distant sources [10,56]. However, the presence of paint chips in the soil close to the Site 1 home suggests that the homes are more likely the dominant metal source. Increases in Pb and Zn near the road at Site 1 suggests the potential trapping of Pb from legacy automobile emissions in the roadside soils. The deposition of Pb and Zn here is more likely the result of historic leaded gasoline use and the wearing of Zn-bearing brake pads [57]. The higher Pb and Zn concentrations near the road at Site 1 and not at Sites 2 and 3 are also consistent with the higher traffic density at Site 1, which is closer to Akron’s urban core. With respect to vertical speciation and distribution, the concentrations of Pb_T-EX and Zn_T-EX decreased with depth. This decline is likely due to metal leaching, where a small fraction of the Pb and Zn migrate from the topsoil [28]. Physical processes, such as erosion and metal translocation, along depth profiles, have also been proposed to account for variations in metal concentration with depth [11].

In addition to changes in concentration, Pb and Zn exhibited changes in speciation, most notably that the percentages of Pb and Zn in the exchangeable and reducible fractions were the highest near the homes and decreased with distance, with greater percentages of Pb and Zn being found in the residual fraction furthest from the homes. These results are consistent with a more labile source of Pb and Zn sourced from the homes and less labile Pb and Zn further from the homes that would be more consistent with background concentrations and speciation [58]. The percentages of Pb and Zn associated with the oxidizable fraction varied much less than the other fractions, and in general, higher LOI values corresponded to slightly higher percentages of Pb and Zn in the oxidizable fraction (Figure 2 and Figure 3), which is consistent with a high affinity of soil organic matter for Pb and Zn [59]. Ultimately, the current speciation and distribution of Pb and Zn in the soils is likely controlled by Pb and Zn being sourced primarily from the homes, although speciation changes with the time since deposition.

4.2. Factors That Control Changes in Pb and Zn Speciation over Time

The three Akron homes underwent major external renovations at 2, 11, and 25 years prior to the collection of soil samples (for Sites 1, 2, and 3, respectively). These renovations included the removal of Pb-based paint from the exteriors and alterations to the home siding and roofs. Such actions are known to result in the rapid addition of high concentrations of Pb and Zn to adjacent soils, particularly in labile forms [60,61,62]. The impact of renovations was observed as a high proportion of Pb and Zn in the exchangeable fraction close in the soils immediately adjacent to the homes (Figure 2 and Figure 3). Furthermore, the speciation and total extractable amounts of Pb and Zn exhibited a linear change with time since renovation, with a decrease in M_T-EX and a shift in speciation from the exchangeable to the reducible fraction (Figure 10a,b). Given that the ongoing addition of Pb and Zn to the soils from the homes was likely minimal compared to the period during renovations, the decrease in total Pb and Zn likely reflects the leaching and erosion of metals as aqueous and/or colloidal species [11,63]. The continual leaching of Pb and Zn without the additional introduction of new high concentrations would result in Site 3 having the lowest M_T-EX concentrations and Site 1 having the highest concentrations based on the time since renovation and independent of the age of the house. Concurrent with decreases in the total extractable Pb and Zn, the proportions of exchangeable Pb and Zn in the soil decreased with corresponding increases in Pb and Zn in the reducible fraction (Figure 2 and Figure 3). This change in speciation from more labile phases (i.e., dominated by the exchangeable fraction) to less labile and slightly more recalcitrant phases (i.e., dominated by the reducible fraction) over time is consistent with previous observations of the weathering of Pb- and Zn-bearing phases in soil [64]. The primary source of Pb from the exterior of the homes would have been Pb-bearing paint, which can include Pb oxides, Pb carbonates, and Pb sulfates [17,65]. The weathering of deposited Pb-bearing paint chips results in the formation of Pb(OH)₂ and, ultimately, the release of aqueous Pb(II) [66]. The primary source of Zn from galvanized roofs is metallic zinc [67]. Similarly to Pb, the weathering of the Zn coating from roofing sheets produces Zn carbonates and oxides, which further liberate aqueous Zn(II) during weathering [68]. Over time, as labile Pb- and Zn-bearing proceed through weathering, a fraction of the Pb and Zn would be lost to leaching and mobilization, while some would be re-sequestered by Fe-(oxy)hydroxides, resulting in the observed decrease in [M_T-EX] and increase in percentage of M_EX over time. Furthermore, given the similar soil pH values of the three sites (Table S2), it is unlikely that changes in soil porewater dominate the observed changes in Pb and Zn speciation between the three homes.

The rapid addition of a pulse of Pb and Zn to soil at a known time provides a unique opportunity to calculate the subsequent loss rate of Pb and Zn for the broader region based on similar soil properties. The observed changes in total extractable Pb and Zn and changes in speciation exhibited linear changes over time, although it is possible that these trends will be non-linear over long time periods. The rates of Pb and Zn loss in the soils were 845 mg/kg/year (R² = 0.97) and 79 mg/kg/year (R² = 0.86), respectively, over a two-decade period. A related study conducted within an urban metropolis reported lower rates within a 15-year time period, with soil Pb loss at 2.4 mg/kg/year [28]. It is noteworthy that the soil Pb concentrations in that study ranged from 54 to 99 mg/kg, which are significantly lower than those observed in the Akron homes, indicating potential non-linearity in Pb loss over time when lower Pb concentrations are reached. Similarly, Zn has been reported to have a faster loss rate (up to 58%) in soils compared to other potentially toxic elements, such as Cd, Cu, and Pb [69,70]. It was also possible to determine the rate of change in Pb and Zn from the exchangeable to reducible form, which was 0.79%/year for Pb (R² = 0.96) and 1.37%/year for Zn (R² = 0.98). Ultimately, the relative proportions of Pb and Zn in the exchangeable and reducible fractions at the three homes suggest that speciation is changing over the decadal time scale. Further work is warranted to find homes with a broader time range of renovations and metal speciation to determine the relationship in metal loss and potential non-linear behavior over longer time periods and/or a broader range of soil Pb and Zn concentrations.

4.3. Anthropogenic Impacts on Other Trace Metals in Urban Soils

Other metals, such as Cu, Mn, Co, Cr, and Cd, are also prone to anthropogenic impact in soils. The maximum total extractable concentrations of these metals in the Akron topsoil were 186 ppm for Cu; 1862 ppm for Mn; 116 ppm for Co; 86 ppm for Cr; and 8 ppm for Cd (with the highest values being observed at Site 1) (Figure 4, Figure 5, Figure 7, Figure 8 and Figure 9). All of these values exceeded their background concentrations in soils from this metropolitan area, namely Cu (37 ppm), Mn (30 ppm), Co (20 ppm), Cr (19 ppm), and Cd (0.8 ppm) [71], which is consistent with evidence of anthropogenic impact at varying degrees. The elevated concentrations of Cu_T-EX and Cr_T-EX above the background have been linked to metal release during the wear and tear of the braking system in automobiles [72,73]. The elevated Mn concentrations could be the result of a complex mix of anthropogenic and natural sources, including electronic wastes, pesticides, plant organic matter, and Mn-rich fertilizers for gardening [74]. The speciation of Cu, Cr, and Mn also varied in the soils. Compared to other metals, large fractions of Cr and Cu were observed in the oxidizable fraction of the Akron soils (Figure 4 and Figure 8), which is consistent with findings from similar studies which observed the high affinity these metals have with organic matter [75,76]. The large pools of Cu and Cr in the residual fraction are also consistent with previous studies of urban soils [69,77,78,79]. The labile fractions of Cu and Cr positively correlated with Pb and Zn, which seem to be driven by combustion emissions and vehicle wear-off. The concentration of Mn in the Akron soils dominated the labile fractions, aligning with another study which investigated the fractionation of heavy metals in urban soils [69]. Conversely, another study found Mn in the soil predominantly in the reducible fraction [80]. Manganese may have become labile upon release into the Akron soil, subsequently being associated with both exchangeable and reducible fractions (Figure 5). The exchangeable fraction of Mn positively correlates with LOI (electronic annex), suggesting that Mn is associated with the organic content around the homes and is readily released into the soils. The reactive portion of Cd in the soils is dominated by the reducible fraction (Figure 9), which seems to include contributions from anthropogenic sources such nickel–cadmium batteries and landfill leachates [81,82]. Cadmium in urban soils has also been linked to its historical use in producing glass and paint pigments [83,84]. The positive correlation among Cd, Zn, and Pb, coupled with the elevated concentration of labile Cd near the homes, indicates that there is a common source, which is likely to be the homes. Ultimately, in comparison to Pb and Zn, the Akron soils are less anthropogenically influenced by Cu, Mn, Co, Cr, and Cd, as shown by their total extractable concentrations and distribution profiles.

5. Conclusions

This study revealed that the distribution and chemical speciation of Pb and Zn at the house lot scale vary both spatially and temporally. With respect to space, the Pb and Zn concentrations are the highest closest to the homes, with speciation being dominated by more labile fractions in soils immediately adjacent to the homes, and more recalcitrant fractions are dominant towards the edge of the home properties. These results suggest that the homes are the dominant source of Pb and Zn leaching into soils as a result of exterior renovations and the flaking off of Pb-based paint and Zn-bearing materials. With respect to temporal variation, longer times since renovations resulted in lower Pb and Zn concentrations as well as smaller fractions of labile Pb and Zn. These results are consistent with previous studies that have shown the importance of the release of Pb from the exterior of older homes as an important source of Pb to soils; however, the current results bring to focus the acute risk of exposure to more labile Pb immediately following exterior renovations and damage to home exteriors in areas of older housing stock. These results suggest that screening soils as part of a hazard assessment should not be a one-time occurrence but rather be conducted in response to changes in housing development. Furthermore, screening should take into account speciation and the fraction of total Pb that is labile. For example, the U.S. Environmental Protection Agency’s nationwide screening level for Pb in residential properties is 200 mg/kg, and for properties with multiple Pb sources (e.g., lead water service lines and lead-based paint), it is 100 mg/kg [85]. However, these standards refer to the total Pb concentration and do not account for the potentially bioaccessible fractions. Assuming that the total extractable amount of Pb is approximately one third of the total Pb [16], then approximately half of all the soil samples in the current study (and all of the samples within 4 m of the homes) would exceed the screening level. However, given that speciation changes over time, it would be most critical to assess the potential risk of exposure immediately after performing renovations.

The results presented here also highlight the value of using sequential extraction methods as a cost-effective method for determining metal speciation, which can inform remediation efforts. For example, based on the ‘bullseye’ pattern of Pb speciation and distribution surrounding the homes, targeted remediation would only need to treat or cover soils closest to the homes within approximately 5 m, where Pb concentrations are the highest (i.e., above the EPA action limit of 200 mg/kg), which accounts for a fraction of the total property (approximately 30% for Site 1 and approximately 10% for Sites 2 and 3) rather than needing to treat soil across the entire property. These results also highlight the need for a balanced sampling density plan across an urban landscape, where low-density sampling could result in the risk of Pb exposure being underestimated (if the samples are only taken near the edges of properties) or overestimated (if the samples are only taken from areas immediately adjacent to homes). In areas of older housing stock with a high proportion of homes that have recently undergone renovation or have been damaged, it would be critical to balance broad soil sampling across a neighborhood with a more targeted sampling effort at some homes to determine if the bullseye patterns of Pb speciation and distribution are observed to better constrain the risk of exposure. Finally, given the high fraction of labile Pb in the soils, a more thorough assessment of physical speciation (i.e., discrete grains of Pb-bearing phases versus aggregates of Pb-bearing grains coating larger grains) is warranted to determine the risk of exposure through the remobilization of Pb into wind-blown dust and aerosols.