doi.org/10.26434/chemrxiv.12780482.v1

Primary Prevention of Outdoor Lead (Pb) Exposure on ResidentialProperties in Rochester, NY, and Potential of a Sustainable RemediationSolution Involving the Reuse of Drinking Water Treatment Residual(WTR), a Waste Generated Daily by the CityPadmini Das, Stephanie Zamule, Deanna R. Bolduc, Michelle J. Patton, Meghan L. Mendola, Julia Penoyer,Benjamin Lyon, Hanna M. Chittenden, Alexander C. Hoyt, Cassandra V. Dupre, Jack J. Wessel, Ivan Gergi,Charlotte V. Buechi, Jane A. Shebert, Mandeep Chauhan, David Giacherio, Jacob C. Phouthavong-Murphy

Submitted date: 08/08/2020 • Posted date: 10/08/2020Licence: CC BY-NC-ND 4.0Citation information: Das, Padmini; Zamule, Stephanie; Bolduc, Deanna R.; Patton, Michelle J.; Mendola,Meghan L.; Penoyer, Julia; et al. (2020): Primary Prevention of Outdoor Lead (Pb) Exposure on ResidentialProperties in Rochester, NY, and Potential of a Sustainable Remediation Solution Involving the Reuse ofDrinking Water Treatment Residual (WTR), a Waste Generated Daily by the City. ChemRxiv. Preprint.https://doi.org/10.26434/chemrxiv.12780482.v1

Prevention of lead (Pb) poisoning is imperative for public health and environmental justice. This studypresents both assessment and affordable mitigation of Pb exposure risks at individual yard scales, which arecritically important to homeowners to attain primary prevention, but not yet explored thoroughly.

File list (2)

download fileview on ChemRxivChemRxiv Final Pb Manuscript_Das.pdf (3.43 MiB)

download fileview on ChemRxivChemRxiv Final Pb Manuscript_Das.docx (9.98 MiB)

1

Title 1

Primary prevention of outdoor lead (Pb) exposure on residential properties in Rochester, NY, 2

and potential of a sustainable remediation solution involving the reuse of drinking water 3

treatment residual (WTR) waste generated daily by the city. 4

The Authors 5

Padmini Das1, Ph.D.; Stephanie M. Zamule1, Ph.D.; Deanna R. Bolduc1; Michelle J. Patton1; 6

Meghan L. Mendola1; Julia M. Penoyer1, B.S.; Benjamin E. Lyon1; Hanna M. Chittenden1, B.S.; 7

Alexander C. Hoyt1 B.S.; Cassandra E. Dupre1, B.S.; Jack L. Wessel1, B.S.; Ivan Gergi1, B.S.; 8

Charlotte V. Buechi1, Jane A. Shebert1, M.S.; Mandeep Chauhan2, M.S.; David Giacherio1, 9

Ph.D.; and Jacob C. Phouthavong-Murphy3, OMS-III. 10

Affiliations 11

1Nazareth College of Rochester. 4245 East Ave, Rochester, NY 14618, USA; 2Northeastern 12

University, 360 Huntington Ave, Boston, MA 02115, USA; 3Touro College of Osteopathic 13

Medicine 230 West 125th Street, New York, NY 10027 14

15

Corresponding Author 16

Padmini Das, PhD. Email: pdas8@naz.edu, phone: 5855000424; 585-389-2552, fax: 585-389-17

2672; Address: 4245 East Ave, Rochester, NY 14618, USA. 18

Competing Interests: 19

The authors declare they have no actual or potential competing financial interests. 20

21

Keywords: Lead (Pb), primary prevention, in-situ risk assessment, XRF, soil, dust, urban 22

gardens, remediation, chemical immobilization, water treatment residual, WTR, Al-WTR 23

24

25

2

Abstract 26

Background 27

Prevention of lead (Pb) poisoning is imperative for public health and environmental justice. This 28

study presents both assessment and affordable mitigation of Pb exposure risks at individual yard 29

scales, which are critically important to homeowners to attain primary prevention, but not yet 30

explored thoroughly. 31

Objectives and Methods 32

In phase I, the potential risk of Pb exposure from soil, surface-dust, and vegetables grown in yard-33

gardens were measured along with secondary estimation of predicted blood Pb levels in children 34

in seven urban and six suburban residential properties in and around Rochester, NY, U.S. In phase 35

II, aluminum-based water treatment residuals (Al-WTR), a waste sludge generated by the city’s 36

drinking water treatment plant, was investigated for its potential to immobilize Pb as a remedial 37

measure. 38

Results 39

All tested properties with pre-1978 built structures were found to contain Pb in exterior paint at 40

varying depth, soil, surface dust, and garden vegetables. Soil Pb levels (SLL) were heterogeneous 41

(11 to 173,500 mg/kg) and higher in urban than suburban properties (mean 3,178 and 461 mg/kg 42

respectively), underscoring the inequitable risk. 71% of vegetables collected across four property 43

gardens exceeded the European Union’s health-based Pb guidelines and the median Pb in plants 44

was 72 times higher than the U.S. market-basket values. Very strong sorption (>90%), minimal 45

desorption (<2%), effective prevention (>85%) of Pb leaching from high Pb-containing exterior 46

paint chips and estimated Freundlich and Langmuir parameters suggest practically irreversible Pb-47

WTR binding. 48

Discussion 49

Primary prevention tools were developed to communicate risk with homeowners through GIS-50

maps that illustrate risk-categories and prediction models to calculate SLL and predicted Pb in 51

vegetables using distance from the house. Al-WTR showed very high Pb-immobilizing ability, 52

3

suggesting the implementation of this no-cost, zero-waste sorbent as a soil amendment would 53

provide sustainable primary prevention for low-income urban communities where Pb exposure 54

prevails. 55

1. Introduction 56

Lead (Pb) exposure continues to be the single most critical environmental health issue affecting 57

American children (CDC 2002), despite the ban on lead-based paint (LBP) and the phasing out of 58

leaded gasoline that took place four decades ago (USEPA 2011; Wagner and Langley-Turnbaugh 59

2007). Lead is a potent neurotoxicant particularly detrimental to developing fetuses and children 60

under the age of six, as it causes cognitive and behavioral impairments, learning disabilities, 61

attention deficit, juvenile delinquency, and violent or even criminal behavior (CDC 2002). The 62

number of children with elevated blood lead levels (EBLL) has been reduced due to the 63

implementation of secondary interventions such as testing BLL in children followed by tracking 64

and mitigating the hazard source. However, the problem remains unaddressed in two critical areas. 65

Firstly, it is increasingly evident from recent research that although EBLL-based secondary 66

prevention has been effective in reducing exposure to some extent, it comes at a cost to children 67

who have already been exposed, and who pay a steep price in irreparable systemic damage through 68

cumulative exposure (Crinnion and Pizzorno 2019). 80% of the total body Pb in children, and 90% 69

in adults, is stored in bone; and while the half-life of Pb is only thirty-five days in blood, it is eight 70

to forty years in bone (Hu et al. 2007). Pregnant and postmenopausal women undergo increased 71

bone turnover, causing Pb poisoning of the fetus and Pb-related age-associated problems in adult 72

women (Gulson et al. 2007; Nash et al. 2004). Cumulative exposure through bone-Pb turnover 73

continues throughout life causing neuroinflammatory diseases like Alzheimer’s and Parkinsonism 74

at variable BLL (Bakulski et al. 2012; Weisskopf et al. 2010); increased cardiovascular mortality 75

in BLL as low as 3.62 µg/dL (Menke et al. 2006); and anxiety and depressive disorders in young 76

adults with BLL as low as 1.7 µg/dL (Bouchard et al. 2009). The BLL action limit for children has 77

recently been reduced from 10 µg/dL to 5 µg/dL by U.S. Centers for Disease Control and 78

Prevention, despite their acknowledgment that there is no safe limit for BLL in children; however, 79

only 18 states (not including NY) have adopted this standard thus far (CDC 2018). Moreover, 80

USEPA regulatory limits for residential soil Pb levels (SLL) (400 and 1200 mg/kg for bare soil in 81

4

children’s play areas and the remainder of the yard, respectively; USEPA 2001) are associated 82

with BLL higher than 5 µg/dL according to two prediction models that independently showed a 83

steep increase in BLL at lower soil Pb levels (Mielke et al. 2007; Johnson and Bretsch 2002). 84

These guidelines, which are substantially higher than the soil Pb standards set by some European 85

nations (40 to 150 mg/kg) and Canada (140 mg/kg), have remained unchanged for decades, despite 86

recommendations by researchers to lower them to safer limits (Mielke et al. 2008; Reagan and 87

Silderbard 1989; Bell et al. 2010). 88

Secondly, the current state of residential Pb exposure poses critical questions related to equality 89

and social justice. Despite a substantial drop in the national average, EBLL still prevails in densely 90

populated urban communities, meaning that children from lower-income minority families suffer 91

most from Pb poisoning (Mielke 1999; Geltman et al. 2001). Fillippelli et al. 2005 reported that 92

15% of children living in urban communities exhibit BLL’s above 10 µg/dL, compared to only 93

2.2% of children nationally. Similar trends were observed in western New York state cities, like 94

Rochester and Buffalo, where most frequent Pb exposure was found to occur in low-income urban 95

neighborhoods, where people of color are highly concentrated (Magavern 2018). The established 96

correlation between Pb exposure and crime in teens and young adults in low-income urban 97

communities (Nevin et al. 2007) underscores the socioeconomic injustice of this issue and the 98

importance of developing a sustainable intervention that will address the environmental, social, 99

and economic aspects of this problem equitably. 100

Primary prevention, which directly measures and corrects hazards before an exposure happens 101

(CDC 2005), is the only process that can provide an equitable solution to the Pb exposure issue 102

without vulnerable populations being disproportionately affected as collateral damage. Sustainable 103

measures of primary prevention are not easy to attain, as they involve a number of multifaceted 104

tasks and challenges. A complete assessment of SLL distribution, outdoor surface dust, and 105

property garden vegetables at the individual property scale is critically important for homeowners 106

to understand the contamination pattern, associated health risks, and potential remedial strategies 107

(Clark and Knudsen 2013). Primary contributors of EBLL in children are Pb in soil and house dust 108

that consists of 20-80% resuspended Pb from the soil, along with other dietary sources. (Mielke et 109

al. 2007; Wu et al. 2008; Bugadalski et al. 2013). One-third of U.S. residents grow at least one 110

vegetable in their yard for intended consumption (National Gardening Association 2011), 111

5

heightening the risk of Pb exposure from contaminated yard soils (Clark and Knudsen 2013). This 112

percentage is expected to increase due to the recent COVID-19 pandemic given the disrupted 113

access to fresh and nutritious foods (Lal 2020). Further, an effective and affordable remediation 114

solution must be provided to the families of lower socioeconomic standing to prevent Pb exposure 115

at identified hot spots. Surface soils act as a sink for accumulating Pb over years and can remain 116

so for centuries in the absence of proper remediation and intervention (LaBelle et al. 1987; Datko-117

Williams et al. 2014). Although the excavation of the surface soil with high Pb is the best practice 118

to remove the risk (Mielke et al. 2011), this process is not economically feasible for homeowners 119

in lower socioeconomic urban neighborhoods. The lack of access to risk assessment measures and 120

economically-feasible alternatives for Pb removal contributes to the ongoing inequitable Pb 121

poisoning in children. 122

To address primary prevention, we investigated thirteen residential properties in and around 123

Rochester, NY, United States, utilizing three unique approaches that differ from earlier studies. As 124

opposed to evaluating commercial and residential locations on a city-wide scale, we focused on a 125

site-specific scale of individual properties with structures built prior to 1978 as the principal 126

outdoor Pb sources. Furthermore, we determined the spatial distribution of SLL, dividing every 127

property into broader risk categories such as: safe areas with minimal to no risk; low-risk areas 128

safer for children to play, but not safe enough to grow edible produce; and high-risk areas with Pb 129

hot spots, which should be avoided and considered for remedial actions. Using this data, primary 130

prevention tools including mathematical prediction models and Geographical Information System 131

(GIS) maps were developed to provide homeowners with a thorough understanding of the risk 132

distribution on their properties. Finally, to address the need for a cost-effective remediation 133

strategy to reduce exposure risk from the Pb hot spots, we explored the potential reuse of a waste 134

sludge as a soil amendment to chemically immobilize Pb. This water treatment residual (WTR) is 135

generated as a waste product on a continual basis by the city’s largest drinking water treatment 136

plants. Successful implementation of this technology will provide the city with a sustainable 137

solution for reducing costs of waste management as well as provide homeowners with a low-cost 138

solution to reduce the risk of Pb exposure from high-risk property hot spots. To the best of our 139

knowledge, this study is the first to focus on both aspects of primary prevention by directly 140

measuring the outdoor Pb levels at individual yard scales in urban and suburban residential 141

6

properties and proposing a low-cost, zero-waste solution for the homeowners by reusing a waste 142

that the city generates every day. 143

2. Methods 144

2.1. Phase I: Assessment 145

2.1.1. Field Site Descriptions 146

Seven urban and six suburban residential properties were selected based on varying characteristics 147

including the age of the property’s structures; socioeconomic categorizations of the homeowners 148

such as race, education, and household income; and the willingness of the homeowners to 149

participate in this study. Tableau data visualization software (version 2020.1) was used to create 150

site location maps with respect to the number of houses built prior to 1978 and the socioeconomic 151

standing of the corresponding areas of the sub-county (figure 1). The software was used to develop 152

a geographical map of Monroe and Ontario counties utilizing latitude and longitude coordinates 153

for each sub-county boundary. An urban location was defined by a geographical location within 154

the city limits of Rochester, NY, United States. A suburban location was defined as a geographical 155

location outside the city limits and within the county limits of Monroe or Ontario counties. The 156

socioeconomic categorization data was collected from the Census Bureau for each respective sub-157

county. 158

2.1.2. In-situ measurement of soil Pb using XRF 159

Soil lead concentrations were measured on-site using a USHUD certified Thermo ScientificTM 160

NitonTM XLp 300 series x-ray fluorescence analyzer (XRF). The XRF was set to “standard bulk” 161

mode while an uncovered soil surface was chosen to conduct the measurement in each location (n 162

= 20). The soil surface locations included: i) alongside structures including corners and beneath 163

doors, front stairs, and windows; ii) along a perpendicular line starting from a high soil Pb point 164

adjacent to the structure towards the property line; iii) adjacent to sidewalks and near structures of 165

neighboring properties. Additional parameters including variation in slopes and presence of loose 166

paint chips were noted if observed. Once recorded, each measurement was marked with flags with 167

different colors (figure S.1), a process used to communicate with the homeowners about the 168

distribution of soil Pb on their property and potential risk associated with it. 169

7

2.1.3. Measurement of Pb in paint and outdoor dust 170

Exterior paint was measured in-situ using the “paint mode” of the XRF on outdoor surfaces such 171

as siding, doors, steps, window sills, pillars, and old housing materials stored outside and inside a 172

barn, (site E only). Calibration of the XRF was performed according to the HUD calibration 173

protocol. Levels of Pb in the paint on a given surface area and a depth index indicating how close 174

the Pb was to the surface were obtained as part of each measurement. 175

Outdoor dust collection was completed using Fisher ScientificTM dust wipes (catalog number: 176

NC0309992) on any surface which registered an XRF lead reading. To ensure maximum 177

collection, the dust wipes were unfolded for the initial collection, then folded in half and used 178

within the same surface area. Pb levels on each dust wipe were measured in the laboratory using 179

the “dust wipe” mode of the XRF by placing each wipe in a Niton Portable Test Stand and scanning 180

for 80 seconds total (20 seconds for each position of the holder). 181

2.1.4. Collection of produce/plant samples and preparations 182

Four assessed properties (A, H, C, M) had home gardens. Produce and plant samples were 183

collected with the permission of the homeowners. Plant samples were cleaned and weighed using 184

a Fisher Science Education analytical balance, dried in a ThermoScientific Heratherm OGS100 185

oven at 105°C for 24 hours (or until completely dry), and then weighed again to calculate moisture 186

content. A dry weight (maximum of 0.5 g or based on the available mass of produce) of plant 187

sample was digested in Fisher Scientific Trace Metal Grade HNO3 following the US EPA 3051 188

protocol in a CEM MARS6 digestion microwave unit. Once digested, samples were filtered 189

through Thermo scientific 0.45 µm PTFE-L Luer-Lock syringe filters, diluted to 50 mL with RO 190

water from a Thermofisher Scientific Barnstead Nanopure purifier, and centrifuged at room 191

temperature in an Eppendorf 5424R for 15 minutes at 4,000 rpm. 192

2.2. Phase II: Remediation 193

2.2.1. WTR Collection, Characterization, and TCLP 194

Two batches of WTR were collected from two sludge lagoons of Monroe County Water 195

Authority’s drinking water treatment plant in Greece, NY, U.S. in 2017 and 2018. These lagoons 196

8

contained the post-flocculation waste sludge from Lake Ontario source water after being treated 197

with alum (aluminum sulfate). WTR samples were then thoroughly mixed, air-dried, ground into 198

powder, and passed through a 2 mm sieve. Triplicate samples collected from each batch of WTR 199

were used for determining physicochemical properties, including pH, EC, moisture content, and 200

organic matter content. Toxicity characteristic leaching procedures (TCLP) were performed using 201

EPA SW-846 methods 1311. 202

2.2.2. Sorption and Desorption of Pb by WTR 203

Kinetic and equilibrium sorption and desorption experiments were carried out in triplicate using a 204

1:20 solid to solution (m/v) ratio (SSR) of WTR to a lead-spiked solution made from lead nitrate 205

and 0.01M KCl as a background electrolyte. Kinetic sorption/desorption experiments were 206

conducted at four initial Pb concentrations (0, 25, 200, and 1000 mg/L) for 24h with six 207

intermediate sample collections after 0, 1, 2, 5, 10, and 24 h during end-over-end mixing on an 208

Analytical Testing Model DC-20 Rotary Agitator shaker at 250 rpm. Similarly, equilibrium 209

sorption experiments were conducted at six initial Pb concentrations (0, 200, 400, 600, 800, and 210

1000 mg/L) for 10 h using the same sampling method. Three subsequent desorption cycles (24h 211

each) were carried out to determine the extent of the potential release of pre-adsorbed Pb from 212

WTR. In addition to shaking time, the WTR remained in contact with the Pb solution for 213

approximately 8 min (0.13 h) during the sample preparation. This time in combination with the 214

shaking time was represented as the contact time for kinetic data. After the respective sorption or 215

desorption processes, samples were centrifuged using an Eppendorf 5804R for 15 minutes at 4000 216

rpm, filtered through Thermo scientific 0.45 µm PTFE-L Luer-Lock syringe filters, and the 217

supernatants were analyzed for Pb and Al. 218

2.2.3. Optimizing WTR to prevent paint chip Pb leaching 219

Loose paint chips were collected from the yard of site A both before and after the home was 220

renovated, in order to conduct two experiments. First, a TCLP test was conducted to estimate the 221

Pb leaching potential from these scattered paint chips in simulated landfill conditions following 222

the EPA SW-846 methods 1311. Second, a batch desorption experiment in the presence of varying 223

levels of WTR was performed to determine the optimal amount of WTR necessary to effectively 224

prevent Pb leaching. Four WTR-to-paint chips (m/m) ratios (10, 20, 50, and 100) were added to a 225

9

0.01 M KCl solution while maintaining a 1:20 SSR. Samples were equilibrated for 7 d with two 226

desorption cycles (1 and 6 d) using end-over-end mixing. The samples were centrifuged at 4000 227

rpm for 15 mins, filtered through a Thermo scientific 0.45 µm PTFE-L Luer-Lock syringe filter, 228

and the supernatants were analyzed for Pb. 229

2.3. Pb Analysis 230

In phase I, Plant samples were analyzed for lead content in triplicate using an Agilent Technologies 231

4210 MP-AES. The MP-AES was calibrated with standards ranging from 2500 to 10000 μg/L 232

prepared from a stock solution of Agilent Technologies 1000 μg/mL Pb in 5% HNO3. Method 233

analysis settings were as follows: pump speed of 25 rpm, sample uptake time of 15 seconds, 234

stabilization time of 15 seconds, read time of 30 seconds, and nebulizer flow of 0.8 L/min. 235

In phase II, Pb in 5% of in-situ soil samples (following USEPA method 6200), Pb and Al from the 236

batch sorption/desorption, and Pb from leaching experiments were measured using a Perkin Elmer 237

400 graphite flame Atomic Absorption Spectrophotometer (AAS). After TCLP, the eight RCRA 238

metals (Arsenic, Barium, Cadmium, Chromium, Lead, Mercury, Selenium, and Silver) were 239

measured using a Perkin Elmer Optima 8000 inductively coupled plasma optical emission 240

spectroscopy (ICP-OES) to attain data at lower detection limits. Calibration coefficient limits were 241

set to 0.995, with an allowed calibration error of 10%, and quality control samples were accepted 242

within +5% error margins. 243

2.4. Data analyses, Modeling, and Mapping 244

Statistical analyses were carried out using JMP IN version 13.0 (Sall et al. 2005). One-way and 245

Two-way ANOVA were performed as required, followed by mean comparisons using the Tukey–246

Kramer honest significant difference (HSD) test. Multivariate correlation analyses of Pb in surface 247

dust were conducted with Pb in paint and depth index; bivariate correlation analyses of Pb in edible 248

produce or plant parts were conducted with the adjacent soil Pb levels. Two types of prediction 249

models were developed using the relationship between i) soil Pb over distance from the structure(s) 250

at each site and ii) Pb in produce/plants grown in the yard gardens with the adjacent soil Pb 251

concentrations. Sorption data were fit to both Freundlich and Langmuir isotherm models to gain 252

insight on possible mechanisms of sorption between Pb and WTR and the reversibility of this 253

10

binding. XRF measurements of SLL were taken along with the latitude and longitude of each 254

sampling location and the collected readings were imported into ArcGIS Pro 2.5.2 and portrayed 255

with proportional symbols, color categories, and heat maps. 256

3. Results 257

3.1. Outdoor Pb-exposure in urban and suburban residential properties 258

As described in fig. 1 and table S.1, seven urban and six suburban residential properties built 259

between 1830 and 2003 were tested for outdoor Pb exposure sources. Among those, the house at 260

site F was the only structure built after 1978. As expected, there was no Pb found in exterior paint, 261

dust, or natural soil at this property. Interestingly, low soil Pb levels (27.8 + 0.98 mg/kg) were 262

found at two sites in the garden area, both of which were covered with store-bought potting soil 263

and mulch, suggesting that either may have contributed Pb to the garden. Two suburban sites (I 264

and J) were found to have Pb in exterior paint and soil; however, the SLL in these two properties 265

were below 400 mg/kg. All seven urban (A, B, C, G, H, K, M) sites and one suburban (L) site were 266

found to have SLL higher than 400 mg/kg. Table S.1 shows the Pb in paint, its depth index, and 267

the maximum concentration of Pb in soil and dust. 268

3.1.1. Distribution of Pb in soils 269

The distribution of SLL was extremely heterogeneous in all properties assessed and thus was 270

explored further in eight properties that surpassed the federal limit. Three factors were found to be 271

controlling this distribution; i) distance from the most prominent source of LBP, ii) slopes and iii) 272

the presence of a secondary source at close proximity. Among these, the distance from the primary 273

source of LBP was found to be the dominant factor controlling the distribution of SLL. The 274

maximum SLL were found alongside the source structures at all sites, indicating the most likely 275

primary source was LBP that leached into the soil over the years (figure 2). The data shows a 276

continual decrease of soil Pb over distance, dropping below 100 mg/kg in all properties at variable 277

distances from the houses. However, a few exceptions were noted in A2, L1, G1, G2, and K. These 278

variations in SLL against the effect of distance were observed due to the presence of loose paint 279

chips at close proximity to the measured locations in A2 and slopes in L1. Both sites G and K were 280

11

inner-city residential properties with smaller yards where the structures of the neighboring 281

property influenced the distribution of SLL by contributing as secondary sources of Pb (figure 3). 282

In each yard, SLL showed a biphasic pattern: initially steep and then a gradual decrease over 283

distance from the source structures. This trend was utilized to create site-specific and overall 284

regression models. Site-specific models based on polynomial cubic regression equations are 285

presented for seven yards in table 1, which exhibited both significance as well as strength (R2 286

ranging from 0.84 to 0.99). In contrast, the overall bivariate regression models across all sites were 287

significant for linear (p = 0.0011) as well as nonlinear polynomial (n = 2 to 6) fit (table S.2), but 288

neither were acceptable to be used, as the regression coefficients were too low (no more than 0.24). 289

3.1.2. Pb in outdoor surface dust 290

Figure 4 shows the multivariate correlation of surface dust collected from different surfaces with 291

Pb in paint and depth index. All surfaces showed a significant (p < 0.05) correlation with both Pb 292

in paint and depth index. Interestingly, Pb was found in all of these surfaces at a low depth index, 293

suggesting the presence of Pb closer to the surface that impacted its loading in the dust. Figure S.3 294

shows a similar multivariate correlation for surfaces with a higher depth index and the surface dust 295

only showed a significant correlation with Pb in paint but not with depth index. Surface dust 296

measurements in all urban properties assessed (with the exception of site C) exceeded HUD’s 297

action level of 800 µg/ft2 for exterior concrete. In contrast, surface dust measurements were within 298

this limit in all suburban properties tested, with the exception of a barn at site E which showed a 299

very high risk of potential Pb exposure through inhalation of dust. The assessment of Pb was 300

conducted inside the barn built in 1830; however, the barn is considered an outdoor source as it 301

was not located adjacent to the house built in 1966. The barn is currently used to store materials 302

from a previously demolished structure and seldom used as a play area for visiting grandchildren 303

of the current property owners. All 19 objects tested in this barn (including the wall, doors, wall 304

decor, and various old materials) contained Pb in the paint with an average depth index of 1.66 + 305

0.62, showing most of the Pb is on the surface. All 19 dust samples collected from this site were 306

found to contain Pb averaging 1,551 µg/ft2, almost twice the level of HUD’s action level. The 307

highest concentration of Pb-containing surface dust reported in this study (site E) was 8,612 µg/ft2, 308

12

which is 11 times higher than the federal guideline, showing the extent of the hidden risk of 309

potential Pb exposure through inhalation of dust. 310

3.1.3. Pb in produce/plants in outdoor home gardens 311

Table 2 shows the plant Pb concentrations (mg/kg) and the Pb levels of the adjacent soil in four 312

yard-gardens at sites A, C, H, and M. Tomatoes collected from the garden of Nazareth College of 313

Rochester, NY, which had adjacent SLL below detection limits, were used to calculate the 314

background plant Pb levels of plants from the property sites. Pb concentrations in tomatoes from 315

the Nazareth College garden were subtracted from the Pb levels found in plants from the properties 316

tested to focus on the effect of SLL on the Pb uptake and accumulation in plant tissues. Tomato 317

plants were found in three gardens (sites A, C, and M), and the lead accumulation in tomatoes 318

significantly (p<0.0001; F ratio=750.7) increased in a linear manner (R2 = 0.99) with increasing 319

SLL. Site A had several garden beds filled with potting soil which contained soil Pb, but at much 320

lower levels than 400 mg/kg. Currently, there is no existing guideline defining the permissible 321

level of Pb in the soil for gardening. All plants collected from property garden beds, including 322

arugula, lettuce, radish, garlic, tomato, potato, and broccoli, were found to contain high 323

concentrations of Pb, ranging from 1.25 mg/kg in garlic to 4.67 mg/kg in radish (by dry weight). 324

Sites H and M had gardens in locations with higher SLL levels compared to other locations in their 325

respective properties. Consequently, the tomatoes from site M and elephant foot yam from site H 326

exhibited a very high accumulation of Pb. Overall, as shown in figure S.4, the increase in 327

accumulated Pb in plants collected from all four yard-gardens over adjacent SLL followed a 328

polynomial square fit. Equations 1 and 2 present two prediction models that were derived based 329

on the strong correlation between the soil Pb and Pb in plant tissues per dry weight (d.w.) and fresh 330

weight (f.w.) respectively. 331

332

𝑦 = 2.331 − 0.005 ∗ 𝑥 + 5.74𝑒−7 ∗ (𝑥 − 812.1)2 --------Equation 1 333

𝑦 = 2.338 − 0.0008 ∗ 𝑥 + 6.99 𝑒−8 ∗ (𝑥 − 812.1)2 --------Equation 2 334

13

Where y is Pb concentration in (mg/kg) in plant tissues and x is Pb concentration (mg/kg) in the 335

adjacent soils. Both models were significant (p < 0.0001; F ratio = 120.5 and 47.7 respectively) 336

with strong regression coefficients (R2 = 0.85 and 0.89 respectively). 337

3.2. Chemical immobilization of Pb by WTR 338

3.2.1. Physico-chemical and TCLP characteristics of WTR 339

Table 3 presents the relative physicochemical properties and the results of the TCLP of the two 340

batches of WTR. Both exhibited almost neutral pH, high moisture content, and low organic matter 341

content. TCLP levels of the RCRA metals revealed a complete absence of lead and mercury and 342

the presence of other metals in substantially lower concentrations than the criteria set by the EPA. 343

Both physicochemical properties and TCLP data for these two batches of WTR were statistically 344

similar, which was evident in p > 0.1 and low F ratio. 345

3.2.2. Pb kinetic and equilibrium sorption/desorption by WTR 346

Kinetic sorption experiments showed that the binding of Pb by WTR was rapid and varied with 347

the initial Pb concentrations in solution, while the Pb levels in the control solutions with no WTR 348

remained unchanged (fig. 5). Equilibrium times could not be determined in lower Pb 349

concentrations because of WTR’s extremely rapid and strong sorption for Pb, as the sorption 350

reached 100% instantaneously (within 8 minutes of contact time with no shaking) for 25 mg/L and 351

within 5.13 h contact time in 200 mg/L initial Pb concentrations. At 1000 mg/L initial Pb in 352

solution, the Pb sorption by WTR reached equilibrium within 5.13 h contact time and did not 353

exhibit any significant change in sorbed Pb up to 24.13 h contact time. Absolutely no desorption 354

occurred in any of these initial concentrations within 24 h. 355

The extent of equilibrium sorption of Pb by WTR increased with increasing Pb concentration in 356

solution (fig. 6a), but the percent sorption showed a reverse trend. However, very high sorption 357

affinities (> 90%) were exhibited in all initial loads within 10h (fig. 6d). The data followed an L-358

type curve (fig. 6a) and showed strong fit to the linearized Freundlich (Eqn. 2; fig. 6b; R2 = 0.99) 359

as well as (Eqn. 3; fig. 6c; R2 = 0.98) Langmuir equations and the associated parameters were 360

calculated accordingly. 361

14

𝑙𝑜𝑔 𝑄 = 𝑙𝑜𝑔 𝐾𝐹 + (1 / 𝑛) 𝑙𝑜𝑔 𝐶𝑒𝑞-------Equation 3 362

Where KF (mg/kg) and n are the Freundlich equilibrium constants representing the sorption 363

capacity and adsorption intensity respectively; Q (mg/kg) is the adsorbed amount of Pb by WTR 364

at equilibrium; Ceq (mg/L) is the equilibrium concentration of Pb in solution. The value of KF was 365

determined as 1.6 mg/g, which exhibited a very high sorption capacity of Pb by WTR. The n value 366

determined in this study was above unity (1.82), indicating favorable adsorption of Pb onto WTR 367

through physical processes (McKay 1980; Özer and Pirinççi 2006). 368

(Ce / Q) = (1 / 𝐾𝐿𝑄𝑚) + (Ce / 𝑄𝑚) --------Equation 4 369

Where Qm (mg/kg) is the maximum adsorption capacity and KL (L/mg) represents the Langmuir 370

constant related to the energy of adsorption. A good fit to the Langmuir isotherm indicates the 371

monolayer coverage and homogeneous distribution of Pb on the active binding sites of the WTR 372

surfaces with an adsorption maximum of 25 mg/g, that exhibits a very high sorption capacity. 373

Additionally, another Langmuir constant named separation factor (RL) was calculated following 374

equation 4. 375

𝑅𝐿 = 1 / 1 + 𝐾𝐿𝐶0--------Equation 5 376

Where C0 (mg/L) is the initial Pb concentration. The RL values indicates the nature and the 377

feasibility of the adsorption process, such as unfavorable sorption if RL > 1, linear sorption if RL = 378

1, favorable sorption if 0 < RL <1, or irreversible sorption if RL = 0 (Meroufel et al. 2013). The 379

value of RL calculated for all tested initial concentrations of Pb in this study ranged from 0.03 to 380

0.15, which were between zero to unity, suggesting favorable sorption of Pb on WTR. In addition, 381

these values were much closer to zero than one, indicating more irreversibility in adsorption of Pb 382

on WTR. This was evident in the equilibrium desorption of Pb from WTR. Figure 6e shows the 383

total desorbed Pb after three desorption cycles (24 h each). No desorption was noted in 72 h up to 384

600 mg/L initial Pb concentration, indicating complete hysteresis or complete irreversible binding 385

of Pb by WTR at these initial loads. Although with the further increase of initial Pb, the extent of 386

desorption increased; it remained restricted to < 2% of the sorbed Pb in WTR even after 72 h of 387

shaking. 388

389

15

3.2.3. WTR concentration effects on paint chip Pb leaching 390

As evident from figure S.1, the numerous loose paint chips found in the yard at site A both before 391

and after the renovation of the structure resulted in an extremely high SLL. Locations with buried 392

paint chips exhibited an average of 12 and 5 times higher SLL as compared to the surrounding soil 393

with no paint chips before and after renovation, respectively. The surface soil with both loose and 394

buried paint chips posed a very high risk of Pb exposure through dust loading as well as leaching 395

into the stormwater runoff. A TCLP assessment of these paint chips collected from the yard 396

revealed that they released 47.9 + 1.71 mg/L (n=6) of Pb in solution, which is 9.6 times higher 397

than the EPA permissible Pb levels for TCLP. This data highlights the need for determining 398

potential Pb leaching from these paint chips in a stormwater solution and provided us with an ideal 399

opportunity to test the effectiveness of WTR to prevent the mobilization of Pb from the paint chips 400

with extremely high Pb-leaching potential. 401

The results were highly promising as WTR concentrations significantly (p<0.0001 and F ratio = 402

90.7) reduced Pb leaching from loose paint chips. 43, 61, 84, and 93% prevention of Pb leaching 403

was achieved by 10, 20, 50, and 100% WTR to paint chips (m/m) application after two desorption 404

cycles of 1 and 6 d. Effect of time was masked (p=0.2) by WTR’s very strong affinity for Pb; no 405

significant difference between the higher WTR treatments suggests the application of WTR in a 1 406

to 2 ratio to paint chips would be enough to achieve the optimum prevention. As these paint chips 407

exhibited the maximum Pb concentrations and acted as the major contributor of SLL in site A, a 408

lower dose of WTR would be enough for achieving optimum prevention of Pb leaching in soils at 409

lower SSR. 410

4. Discussion 411

4.1 Field Assessment 412

Although the City of Rochester succeeded in reducing the number of children with EBLL 413

(compared to neighboring cities like Buffalo and Syracuse) through its adoption of an aggressive 414

secondary intervention approach, the city continues to face significant issues related to Pb 415

exposure (Magavern 2018). This is consistent with our findings showing that the overall SLL 416

medians across all tested urban and suburban properties with structures built prior to 1978 were 417

16

413 and 155 mg/kg, respectively, with mean SLL as high as 3,178 and 461 mg/kg, respectively. 418

This study also documented the highest maximum residential SLL (173,500 mg/kg), as compared 419

to those reported over the last two decades in urban and suburban areas of thirty-one U.S. cities 420

(data compared with 36 research articles; not shown). Risk of Pb exposure through soil Pb, 421

potential dust loading, and produce grown in property gardens was found to be much higher in the 422

urban residential properties where residents already face significant socioeconomic challenges. 423

According to the 2010 Rochester census data, urban neighborhoods in the City of Rochester have 424

the highest percentage of residents living below the poverty line, the lowest percentage of those 425

with above-average income, and the broadest distribution of race with the highest percentage of 426

African American residents when compared to suburban sub-counties. The unjust burden of Pb 427

exposure in urban neighborhoods in which many low-income minority families reside has been 428

documented in twenty studies conducted over the last four decades using data from sixteen U.S. 429

cities (Datko-Williams et al. 2014). 430

The prevalence and correlation of SLL and BLL are well established in the literature (Bickel 2010; 431

Zahran et al. 2010; Wu et al. 2010). The overall median SLL in urban properties exceeding the 432

federal standard clearly underscores this risk. However, the lower median SLL of suburban 433

properties does not represent a risk-free condition either. Adverse health effects of Pb are 434

associated with BLL as low as 2 mg/dL, which is associated with SLL < 100 mg/kg (Canfield et 435

al. 2003; Schwarz et al. 2012). Two existing prediction models by Mielke et al. (2007) and Johnson 436

and Bretsch (2002) independently showed a steep increase in BLL at SLL <100 mg/kg and a 437

gradual rise at SLL >300 mg/kg. To assess the exposure risk more specifically, we used both of 438

these models (equations 5 and 6) to predict the potential BLL if residents (especially children) are 439

being exposed at tested residential properties. 440

𝐵𝐿𝐿 = 2.038 + 0.172 ∗ (𝑆𝐿𝐿 𝑚𝑒𝑑𝑖𝑎𝑛)0.5 -------Equation 6 (Mielke et al. 2007) 441

𝐵𝐿𝐿 = 2.097 ∗ 𝐿𝑛(𝑆𝐿𝐿 𝑚𝑒𝑑𝑖𝑎𝑛) − 3.6026 -------Equation 7 (Johnson and Bretsch 2002) 442

SLL in all but one suburban property was predicted to result in BLL > 5 µg/dL according to 443

equation 7, whereas five out of the seven urban properties and one out of the five suburban 444

properties were predicted to result in BLL > 5 µg/dL according to equation 6 (table 4). We 445

observed that the predicted BLL based on the median SLL of the entire property underestimates 446

17

the predicted risk for properties with high variation in SLL throughout the property. For instance, 447

as discussed earlier, site A showed a high variation in SLL throughout the property, both before 448

and after renovation. The predicted BLL of the front yard after the renovation was as high as 20.9 449

µg/dL (equation 6) and 16.1 µg/dL (equation 7), both of which are much higher than the predicted 450

BLL based on the median SLL of the entire property. In such cases, yard-specific predictions 451

would be a more appropriate measure of risk assessment. Racial and socioeconomic distribution 452

in the City of Rochester, along with the SLL and estimated BLL documented in this study, echo 453

the findings of Mielke et al. (1999) and Geltman et al. (2001), who reported that higher SLL in 454

urban soils result in inequitable EBLL in children of lower-income families, minorities, and recent 455

immigrants. This is also evident from the data shown by Korfmacher (2007) that low-income 456

families in the crescent area in urban Rochester characterized by high poverty, crime, and pre-457

1950 rental housing correlated with 23.6% of children under 6 years of age having blood lead 458

levels greater than 10 µg/dL. 459

In accordance with this study, extreme heterogeneous distribution of soil Pb has been documented 460

by other researchers, and has been observed to decrease with increasing distance from three Pb 461

source categories: i) industrial sites or brownfields with residues of historic Pb use (Marshall and 462

Thornton 1993; Wu et al. 2010), ii) areas near highways or roads with intense traffic density 463

(Mielke et al. 2008; Laidlaw 2001; Filippelli et.al. 2005), and iii) aged structures with exterior 464

LBP (USEPA 2000; Litt et al. 2002; and Clark and Knudson 2013). Some studies have reported 465

that the combustion of leaded gasoline is the primary source of Pb in larger urban settings, as noted 466

in Baltimore (Mielke et al. 1983), New Orleans (Mielke et al. 2008), and Indianapolis (Laidlaw 467

2001; Filippelli et.al. 2005). Clark and Knudson (2013) argue that exterior LBP dominates the 468

distribution of soil Pb in smaller urban communities like Appleton, WI. Our study showed that 469

even in a medium-sized urban community like Rochester, with a history of past industrialized use 470

of heavy metals, the primary source of Pb in all tested sites was exterior LBP from structures built 471

prior to 1978, as none of these sites are located in neighborhoods with high traffic density or at 472

close proximity of any brownfield or superfund site. The building structures can also act as barriers 473

and facilitate atmospheric deposition of Pb (Mielke et al. 1983), leaving a pattern of higher Pb 474

trapping in the street side of the front yard as compared to the back or side yard. A combined effect 475

of preferential trapping and LBP results in a U-shaped pattern of higher SLL near the house and 476

the road, and lower SLL in the middle of the yard (Olszowy et al. 1995). Lack of these types of 477

18

spatial patterns negates the effect of preferential trapping on properties we assessed. Based on the 478

findings of our current study, we argue that even in medium urban communities, leaded exterior 479

paint can be the dominant source of soil Pb if the property is located at a certain distance from the 480

industrial sites, freeways, or dense traffic area. The location and age of the building structure are 481

the two most important parameters that influence which source will dominate the distribution of 482

soil Pb and to what extent in any given community, irrespective of its size. 483

Nationwide studies conducted by U.S. Housing and Urban Development report that 76% of homes 484

built prior to 1960 have Pb in exterior paint and 83% of pre-1980 housing stock contains Pb in the 485

paint (USHUD 1990, 1995). This study found Pb in exterior paint at varying depths for all tested 486

urban and suburban properties built prior to 1978, supporting the observation of Clark and 487

Knudson (2013) that all communities of similar age and site throughout the US show a similar 488

trend of contamination. SLL can exist around all sides of the structures and extend far into the 489

yard (Clark and Knudson 2013; USEPA 2000, Litt et al. 2002). In all properties, the SLL was 490

observed to be very high alongside the structures and decreased with increasing distance from the 491

homes. In addition, we also noted that in inner-city properties like site A, G, and K, the neighboring 492

structures, adjacent to the tested yards, also acted as prominent sources of Pb. Buried or scattered 493

paint chips in the soil of site A increased the SLL in some locations against the influence of 494

distance. 495

Pb contamination in residential soil is so widespread that it can be found in unapparent locations, 496

increasing the threat of continual risk of exposure. In accordance with previous studies, we have 497

observed three such unforeseen scenarios: i) high soil Pb in a suburban house with a well-498

maintained exterior (Clark and Knudson 2013) was observed in site L, which exhibited SLL as 499

high as 3000 mg/kg; ii) high soil Pb in a property with an unpainted exterior was noted in site B, 500

where the SLL surpassed the federal guideline only beneath the windows, a trend also observed 501

by Linton et al. (1980); and iii) increase in SLL after renovation and repainting of a house (Clark 502

and Knudson 2013), which is duly noted in site A. Table S.3 presents a comparative SLL, estimated 503

BLL, and potential dust loading values in the three yards of site A that illustrates the detrimental 504

effects of renovation activity on soil Pb. 505

19

Dust is a major source of concern for children, as inhalation is a major route of Pb exposure. The 506

respiratory tract is responsible for the absorption of lead particles into the systemic circulation, and 507

respiration and metabolic rates are faster in children, resulting in increased absorption of lead via 508

dust (Menezes-Filho 2018). Two tools were used in this study to estimate the risk of Pb exposure 509

from outdoor dust: i) direct measurement of accumulated dust on the outdoor surfaces, and ii) 510

secondary estimation of potential lead on play surfaces (PLOPS). Overall, accumulated Pb in dust 511

showed significant (p<0.001) positive correlation with Pb in paint and negative correlation with 512

depth index, suggesting flaking LBP on surfaces such as window sills, doors, steps, and old 513

building materials can potentially contribute Pb to dust. 514

Lead loading on exterior surfaces has been found to be a continuous process, primarily due to 515

resuspension of soil Pb (Caravanos et al. 2006; Mielke et al. 2006). Researchers identified the 516

resuspension of soil Pb during dry seasons as the major driving force behind the seasonally EBLL 517

in children in New Orleans, Syracuse, and Indianapolis (Laidlaw et al. 2005). Mielke et al. (2006) 518

developed a new tool for measuring the potential Pb surface loading per area (mg/ft2) of the soil. 519

Potential Pb on play surfaces, a measure of the Pb concentration of a soil surface per surface area 520

as a source of Pb dust for children who would play on that surface, was calculated based on a 521

prediction model created by Mielke et al. (2006) (equation 8). 522

𝑃𝐿𝑂𝑃𝑆 = −43.74 + 24.85 ∗ (𝑆𝐿𝐿 𝑚𝑒𝑑𝑖𝑎𝑛)0.69 -------Equation 8 523

The PLOPS values for all urban and suburban properties are presented in table 5 to provide insight 524

into the overall potential of Pb exposure from these surfaces, both directly through hand-to-mouth 525

activity, or indirectly through resuspension of Pb dust. Mielke et al. (2006) also showed a large 526

drop in PLOPS in 136 properties and vacant lots in New Orleans after the contaminated surfaces 527

were treated with a soil cover. Similar measures should be considered for the properties assessed 528

in our study, as the large PLOPS values indicate a high potential risk of Pb exposure to children if 529

they play in these areas. 530

The relationship between Pb in produce and the adjacent soil Pb is equivocal in the literature; with 531

some studies reporting maximum Pb levels in vegetables associated with the highest SLL 532

(Bielinska 2009; Huang et al. 2012; Moir and Thornton 1989) and other studies finding only weak 533

correlations between crop Pb levels and the adjacent soil Pb, due to unverified reasons (Hough et 534

20

al. 2004; Murray et al. 2011; McBride et al. 2014). Our study found strong and significant (r = 535

0.92 based on fresh weight and r = 0.85 based on dry weight; p < 0.001) correlations between the 536

plant-accumulated Pb and SLL. Ten out of fourteen edibles tested exceeded the health-based 537

guidance values set by the European Union (EU) (EC 2006). There are no such standards currently 538

in the U. S., but the median Pb in edible plants found in this study was 72 times higher than the 539

median market basket concentrations of Pb reported by the U. S. Food and Drug Administration’s 540

(FDA) Total Diet Study. (USFDA 2010). The FDA recommended the maximum ingestion lead 541

level for a child is 3 µg/day; eating between four and five grams of contaminated produce would 542

reach that limit (Frank 2019). Interestingly, Pb levels exceeding the EU guideline standards were 543

found in plants grown in both indigenous soils (sites H and M) of these properties as well as potting 544

soils (site C and side yard of site A). The median Pb in plants grown in indigenous soils (1.05 545

mg/kg) was higher than those grown in the potting soil (0.21 mg/kg) and showed a stronger 546

correlation with the Pb concentrations in adjacent soils (r = 0.91 and 0.69 for indigenous and 547

potting soils respectively). However, it is deeply concerning that the edible plants grown in potting 548

soils containing Pb lower than 60 mg/kg were found to contain Pb levels exceeding the health 549

guidelines. The federal standard of 400 mg/kg soil Pb provides a misconception of safety for soils 550

containing lower Pb concentrations; there are no current levels set by either U.S. Environmental 551

Protection Agency (EPA) or HUD for urban gardening. Moreover, each species of produce has 552

different biological properties that determine their uptake, absorption, and bioavailability of Pb, 553

demonstrated by the varying lead concentrations in them (USFDA 2010). Differential uptake of 554

Pb by different vegetables at similar SLL was noted in the current study (site A) in Rochester as 555

well as by McBride et al. (2014) in New York City and Buffalo, NY; and Misenheimer et al. 556

(2018) in Puerto Rico. These findings strongly support the development of new plant-specific 557

federal guidelines for soil Pb to promote safer gardening practices. 558

4.2 Safer use of yards by primary prevention tools 559

560

Based on results of the field assessment, we developed three primary prevention tools to help 561

homeowners understand the Pb exposure risks throughout their properties as the first step of 562

primary prevention: i) a prediction model to calculate soil Pb using the measured distance from 563

the structure; ii) a prediction model to estimate Pb in edible produce grown in yard gardens from 564

the adjacent soil Pb that can be calculated from the distance from the built structure, iii) GIS maps 565

21

showing Pb hot spots with high risk, low risk, and safe areas in the yards. Prediction models and 566

GIS maps presented by prior research, covering a larger city area, are crucial for bringing necessary 567

changes in the policy (Wu et al. 2010), but this is an ongoing, long-term process. Considering the 568

continual Pb exposure and elevated BLL in families with lower socioeconomic standing, it is of 569

utmost importance to develop an easier process for the homeowners to understand the risk to be 570

able to avoid it. To the best of our knowledge, the present study is the first attempt to develop site-571

specific prediction models and precise GIS maps illustrating the potential risk through mapping 572

the SLL in the exact latitude and longitude of the sampling locations on individual residential 573

properties. The GIS maps also mark the safer locations to create gardens based on the two 574

prediction models we developed. These measures would provide the homeowners with an 575

opportunity to predict soil Pb levels to plan safer uses of their yards. 576

577

City-wide larger models to calculate soil Pb based on distance from urban centers of big cities or 578

roads with high traffic density considered the atmospheric deposition of Pb as the dominant source 579

(Yaylah-Abanuz 2019; Brown et al. 2008; Zhang et al. 2015; Wu et al. 2010). However, residential 580

soil Pb with exterior LBP as the major source varies widely from one property to another, 581

depending on the age of the structure(s), presence and types of siding, past land use, and renovation 582

processes. Site-specific prediction models would provide homeowners with an opportunity to 583

predict soil Pb levels using the distance from a structure to utilize their yards more safely. For 584

instance, a play area should be planned in soil that contains less than 400 mg/kg Pb, according to 585

federal guidelines. Based on our site-specific models, SLL drops from 400 mg/kg to < 100 mg/kg 586

if the play area is located 2 m further from the house. Considering the steep increase of BLL 587

between 100 to 300 mg/kg SLL, simply moving a play area a few meters further from the structure 588

would increase the chance of primary prevention at no cost. Although we highly recommend at 589

least a one-time assessment of soil Pb for each property, it is likely that one prediction model 590

would be valid for one block or neighborhood if the homes were built at a similar time; however, 591

further investigation of this assumption is necessary. 592

593

This study also presents prediction models exhibiting a strong correlation (r = 0.85 and 0.92) 594

between Pb accumulation (per d.w. and f.w.) in produce/plant parts and the Pb levels in the 595

adjacent soil. One limitation of these prediction models is that they will not be able to provide any 596

22

insight into the plant-specific differential Pb uptake, as the pronounced effect of soil Pb may have 597

masked the effects of plant-specific uptake in the current study. However, utilization of these 598

equations would be able to provide a rough estimate of the potential Pb accumulation into 599

vegetables from the adjacent SLL to estimate a safer location for a garden. For instance, the 600

vegetable gardens in sites M and H were located in the areas with the highest SLL, resulting in 601

median Pb levels in edible parts of 1.58 and 0.56 mg/kg, respectively, which exceed the EU 602

guideline by 15.8 and 5.6 folds, respectively (EC 2006). Both of these properties have areas with 603

much lower soil Pb further into the yard. Simply moving these gardens to such locations would be 604

a useful measure of intervention to reduce the Pb exposure risk through the consumption of Pb-605

contaminated vegetables from the property gardens. We estimated an SLL value associated with 606

the EU guideline of 0.1 mg/kg Pb per plant f.w. using a reverse fit of equation 2 (R2 = 0.97, 607

p<0.0001) which associates with a SLL of 64.7 mg/kg; a slightly lower value of 0.09 mg/kg Pb 608

per plant f.w. associates with an SLL of 58.2 mg/kg. Using the reverse fit of the site-specific 609

models, we calculated the associated distance with 58.2 mg/kg soil Pb in site M and site H, and 610

these distances are found to be 6.44 and 3.85 m respectively. Thus, our combined models indicate 611

that to attain a minimum requirement for the primary prevention, a home garden should be set up 612

at least 6.44 m away in site M and 3.85 m away in site H from their respective structures. Similarly, 613

a safer distance can be calculated for any property garden to avoid consumption of Pb through 614

homegrown vegetables. 615

616

Figure 8 presents two reference GIS maps of site A showing the Pb exposure risk associated with 617

soil Pb distribution in three areas of the property before and after the renovation of the home. To 618

simply and clearly communicate risk levels to homeowners, we developed a visual tool utilizing a 619

graded color scheme of green to black representing risk levels. Table S.4 presents a guideline 620

created for the homeowners to explain each color category, associated risk, and the recommended 621

use of that location on their properties. In addition to using the federal standards for assigning 622

colors to the range of Pb concentrations, we have also used our prediction models to mark areas 623

of the property where it would be safer to create a garden based on the EU health guidelines for 624

edible produce. The areas marked in red to black colors represent hotspots with very high to 625

extreme risk of potential Pb exposure and thus must be avoided and should be considered for 626

remedial actions. 627

23

4.3 Sustainable Remediation 628

629

The second and final step of primary prevention is to correct the hazards through remedial actions, 630

which led us to investigate an affordable process that would provide substantial risk reduction 631

without economically burdening the homeowners. WTR collected from Rochester’s largest 632

drinking water treatment plant not only exhibited a very high Pb-binding ability but was also found 633

to be safe when used as a soil amendment according to the federal TCLP guidelines. Past research 634

documented varying effectiveness of different sorbents to immobilize lead, including activated 635

carbon (Moyo et al. 2013), natural and engineered clay-like zeolite (Wingenfelder et al. 2005), 636

mordenite (Turkyilmaz et al. 2014), montmorillonite (Zhang and Hou 2008), bentonite (Inglezakis 637

et al. 2007), palygorskite, and sepiolite (Shirvani et al. 2006). Although these technologies provide 638

in-situ alternatives to the current expensive ex-situ practices, the cost of these natural or engineered 639

sorbents in bulk could become cost-prohibitive to low-income urban families. WTR is a no-cost 640

sorbent which would provide these homeowners with an affordable remediation alternative to be 641

utilized as soil cover or amendment in the most concerning areas of their properties. The only 642

required cost is associated with its implementation, which could be further reduced by conducting 643

a proper assessment to mark the risk hotspots (as shown in the GIS maps for site A), where the 644

remediation process is absolutely needed. Very few studies have investigated the effectiveness of 645

Pb adsorption by Fe and Al-based WTR collected from different parts of the world. Pb sorption 646

by Al-WTR showed a similar pattern of L type sorption isotherm (Castaldi et al. 2015; Zhou et al. 647

2011). Adsorption maximum was reported to be 17.55, 15.66, and 15.13 mg/g by WTRs collected 648

from Miyamachi and Nishino in Japan and Italy, respectively, as compared to 24.4 mg/g found in 649

this study (Putra and Tanaka 2011). Furthermore, Soleimanifer et al. (2019) reported the unique 650

approach of reducing metal contamination in urban stormwater with WTR coated mulch; a similar 651

approach could be further investigated for its potential to reduce plant-Pb uptake if used in home 652

gardens. The sorption desorption data along with the calculated Freundlich and Langmuir isotherm 653

parameters suggests very strong, favorable, and more irreversible adsorption of Pb on WTR 654

through a physical process, promising an effective option for reusing this waste byproduct that the 655

city generates every day to immobilize Pb in the high-risk soils of many contaminated residential 656

properties. 657

658

24

Using WTR as a soil amendment will prevent the runoff and leaching of soil Pb and also decrease 659

the Pb available to be accumulated in garden vegetables. However, the implementation of only 660

WTR will not prevent the resuspension of Pb into dust. Our goal is to implement a dual-process 661

remediation system using WTR as a soil amendment with the addition of a plant cover to decrease 662

the dust loading of Pb. Several studies have shown the phytoremediation potential of Pb (Jagetiya 663

and Kumar 2020), but it is of utmost importance to use a non-edible plant in the Pb hot spot areas 664

to prevent its further entry into the food chain. Vetiver, a perennial non-edible and non-invasive 665

grass has been shown to be highly effective for Pb removal (Andra et al. 2009); however, the harsh 666

winters of western NY could be a serious impediment for its implementation. Our group has tested 667

17 native inedible plant species for their potential tolerance of high Pb concentration, but all plants 668

showed severe phytotoxic effects of Pb in hydroponic media, where 100% Pb was available for 669

plant uptake (data not shown). This further underscores the utility of WTR for Pb removal, which 670

according to the results of this study would retain more than 90% of total soil Pb, making only a 671

small fraction available for the plants to tolerate. The ongoing studies in our greenhouse are 672

characterizing this dual process technique using WTR as an amendment to the soils from the 673

backyard of site A, with an SLL median of 2900 mg/kg, and switchgrass (Panicum virgatum), high 674

biomass, perennial, fast-growing, inedible grass, which is also native to western NY and which 675

was previously studied for its ability to remove other environmental contaminants (Phouthavong-676

Murphy 2020). This combined remediation process would provide the homeowners with an 677

effective and affordable primary prevention alternative to intervene with the Pb exposure from the 678

risk hotspots of their property. 679

680

Acknowledgements 681

682

The authors acknowledge Mr. Larry Peckham and Mrs. Nancy Peckham for funding the High 683

School Summer Research Program in 2018 and Nazareth College Summer Opportunities for 684

Activities in Research and Sponsorship (SOARS) in 2018 for partially funding this project. 685

686

We acknowledge Mr. John C. Kelly Jr. and other officials from Monroe County Water Authority 687

(MCWA) for their help in WTR collection; our community partners and homeowners for their 688

communication and participation; Dr. Callie Babbitt from Rochester Institute of Technology for 689

25

conducting the ICP-OES analyses of TCLP samples; Mrs. Sharon Luxmore for assistance with 690

ordering materials; and Ms. Aelis Spiller, Ms. Ariel Struzyk, Mr. LaMar Jackson, Mr. Sean Dillon, Ms. 691

Maria Vargas Morales, and Ms. Bailey Groth for assistance in sample preparation and field measurements; 692

and Mr. Shyamal K. Mitra and Dr. Ananta Paine for providing feedback about this manuscript. 693

694

Author contributions 695

P.D. designed the research, field assessment, and experiments. P.D., D.R.B., M.J.P, B.E.L., A.C.T., 696

J.M.P., H.M.C., J.L.W., I.G., and D.G. conducted the field assessment; B.E.L. and A.C.T analyzed Pb in 697

surface dust, J. S., C.V.B., M.L.M., D. R. B., M.J.P., C.E.D. analyzed Pb in plants with the supervision of 698

S.M.Z. and P.D.; J.L.W., I.G., J.P.M., J.M.P., H.M.C., A.C.T. conducted phase II experiments with the 699

supervision of S.M.Z. and P.D; P.D. conducted the secondary analyses, isotherm modeling, statistical data 700

analyses, and developed the prediction models. M.C. created the GIS maps and D.R.B. created the 701

Tableau maps. P.D. wrote the manuscript with S.M.Z. and J.P.M. with contributions and feedback from 702

all other authors. 703

704

Abbreviations: 705

Pb: lead; WTR: drinking water treatment residual; XRF: x-ray Fluorescence Analyzer; Al-WTR: 706

aluminum-based water treatment residual; SLL: soil lead level; mg/kg: milligram per kilogram; 707

WTR: water treatment residual; GIS: geographical information systems; LBP: lead based paint; 708

EBLL: elevated blood lead level; BLL: blood lead level; µ𝑔/𝑑𝐿: micrograms per deciliter; 709

USEPA: United States Environmental Protection Agency; USHUD: United States Department of 710

Housing and Urban Development; n: number of samples; HUD: Department of Housing and Urban 711

Development; TCLP: toxicity characteristic leaching procedures; SSR: solid to solution ratio; KCl: 712

potassium chloride; mg/L: milligrams per liter; HNO3: nitric acid; AAS: atomic absorption 713

spectrophotometer; RCRA metals: 8 metals (Arsenic, Barium, Cadmium, Chromium, Lead, 714

Mercury, Selenium, Silver) listed under Resource Conservation and Recovery Act; ICP-OES: 715

individually coupled plasma optical emission spectroscopy; Tukey-Kramer HSD: Tukey Kramer 716

honest significant difference. 717

718

719

26

References 720

1. Andra SS, Datta R, Sarkar D, Makris KC, Mullens CP, Sahi SV, et al. 2009. Induction of 721

lead-binding phytochelatins in vetiver grass (Vetiveria zizanioides L.). J Environ Qual 722

38:868–877, PMID:19329675, https://doi.org10.2134/jeq2008.0316. 723

724

2. Bakulski KM, Rozek LS, Dolinoy DC, Paulson HL, Hu H. 2012. Alzheimer’s disease and 725

environmental exposure to lead: the epidemiologic evidence and potential role of epigenetics. 726

Current Alzheimer Research 9(5):563-573, PMID:22272628. 727

728

3. Bell T, Campbell S, Liverman DGE, Allison D, Sylvester P. 2010. Environmental and 729

potential human health legacies of non-industrial sources of lead in a Canadian urban 730

landscape – the case study of St John's, Newfoundland. International Geology Review 52(7-731

8):771-800, https://doi.org/10.1080/00206811003679786. 732

733

4. Bickel JM. 2010. Spatial and temporal relationships between blood lead and soil lead 734

concentrations in Detroit, Michigan. Master’s Thesis, Wayne State University: Detroit, MI. 735

736

5. Bielinska EJ. 2009. Influence of the root layer on the content of cadmium and lead in soils and 737

vegetable plants in regions with diverse anthropogenic impact. J Res Appl Agric Eng 54:16–738

20. 739

740

6. Bouchard MF, Bellinger DC, Weuve J, Matthews-Bellinger J, Gilman SE, Wright RO, et al. 741

2009. Blood lead levels and major depressive disorder, panic disorder, and generalized 742

anxiety disorder in US young adults. Archives of General Psychiatry 66(12):1313-1319, 743

https://doi.org/10.1001/archgenpsychiatry.2009.164. 744

745

746

7. Brown RW, Gonzales C, Hooper MJ, Bayat AC, Fornerette AM, Mcbride TJ, et al. 2008. Soil 747

lead (Pb) in residential transects through Lubbock, Texas: a preliminary assessment. Environ 748

Geochem Health 30:541–547, PMID:18546057, https://doi.org/10.1007/s10653-008-9180-y. 749

750

8. Canfield R, Henderson CR, Cory-Slechta DA, Cox C, Jusko TA, Lanphear BP. 2003. 751

Intellectual impairment in children with blood lead concentrations below 10 mg/dL: the 752

Rochester cohort study. N. Engl. J. Med 348:1517–1526, 753

https://doi.org/10.1056/NEJMoa022848. 754

755

9. Caravanos J, Weiss AL, Blaise MJ, Jaeger RJ. 2006. A survey of spatially distributed exterior 756

dust lead loadings in New York City. Environmental Research 100(2):165-172, 757

https://doi.org/10.1016/j.envres.2005.05.001. 758

759

760

27

10. Castaldi P, Silvetti M, Garau G, Demurtas D, Deiana S. 2015. Copper(II) and lead(II) 761

removal from aqueous solution by water treatment residues. Journal of Hazardous Materials 762

283:140-147, PMID: 25262486, https://doi.org/10.1016/j.jhazmat.2014.09.019. 763

764

11. CDC (Centers for Disease Control and Prevention). 2002. Managing Elevated Blood Lead 765

Levels Among Young Children: Recommendations from the Advisory Committee on 766

Childhood Lead Poisoning Prevention. 767

https://www.cdc.gov/nceh/lead/casemanagement/casemanage_main.htm [Accessed 12 July 768

2020]. 769

770

12. CDC (Centers for Disease Control and Prevention). 2005. Building Blocks for Primary 771

Prevention: Protecting Children from Lead-Based Paint Hazards. 772

https://www.cdc.gov/nceh/lead/publications/building_blocks_for_primary_prevention.pdf 773

[Accessed 28 July 2020]. 774

775

13. CDC (Centers for Disease Control and Prevention). 2018. Safe Blood Lead Testing Laws 776

Requiring 5 ug/dL & CDC Reference Rule. https://www.cdc.gov/phlp/docs/laws-bll.pdf. 777

[Accessed Jul. 28, 2020]. 778

779

780

14. Clark JJ, Knudsen AC. 2013. Extent, characterization, and sources of soil lead contamination 781

in small-urban residential neighborhoods. Journal of Environmental Quality 42:1498-1506, 782

https://doi.org/10.2134/jeq2013.03.0100. 783

15. Crinnion WJ, Pizzorno JE. Clinical Environmental Medicine. Elsevier Health 784

Sciences:Kindle Edition. 8535-8537. 785

786

16. Datko-Williams L, Wilkie A, Richmond-Bryant J. 2014. Analysis of U.S. soil lead (Pb) 787

studies from 1970 to 2012. Science of the Total Environment 468-469:854-863, 788

https://doi.org/10.1016/j.scitotenv.2013.08.089. 789

790

791

17. Dixon DL, Gaitens JM, Jacobs DE, Strauss W, Nagaraja J, Pivetz T, et al. 2009. Exposure of 792

U.S. children to residential dust lead, 1999–2004: II. The contribution of lead-contaminated 793

dust to children’s blood lead levels. Environmental Health Perspectives 117(3):468-474, 794

https://doi.org/10.1289/ehp.11918. 795

796

797

18. EC (Commission Regulation EC). 2006. No 1881/2006 of 19 December 2006 Setting 798

Maximum Levels for Certain Contaminants in Foodstuffs. Off J Eur Union L 364:5–24. 799

https://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2006:364:0005:0024:EN:PDF. 800

[Accessed 12 July 2020]. 801

802

28

19. Filippelli GM, Laidlaw MAS, Latimer JC, Raftis R. 2005. Urban lead poisoning and medical 803

geology: an unfinished story. Geological Society of America Today 15(1):4-11, 804

https://doi.org/10.1130/1052-5173(2005)015<4:ULPAMG>2.0.CO;2. 805

806

20. Frank JJ, Poulakos AG, Tornero-Velez R, Xue J. 2019. Systematic review and meta-analyses 807

of lead (Pb) concentrations in environmental media (soil, dust, water, food, and air) reported 808

in the United States from 1996 to 2016. Science of The Total Environment 694:133489, 809

https://doi.org/10.1016/j.scitotenv.2019.07.295. 810

811

21. Geltman PL, Brown MJ, Cochran J. 2001. Lead poisoning among refugee children resettled 812

in Massachusetts, 1995 to 1999. Pediatrics 108(1):158–162, PMID: 11433069, 813

https://doi.org/10.1542/peds.108.1.158. 814

815

22. Gulson BL, Mizon KJ, Korsch MJ, Palmer JM, Donnelly JB. 2003. Mobilization of lead 816

from human bone tissue during pregnancy and lactation—a summary of long-term research. 817

Science of the Total Environment 303(1–2):79–104, PMID: 12568766, 818

https://doi.org/10.1016/s0048-9697(02)00355-8. 819

820

23. Hough RL, Breward N, Young SD, Crout NMJ, Tye AM, Moir AM, et al. 2004. Assessing 821

potential risk of heavy metal exposure from consumption of home-produced vegetables by 822

urban populations. Environ Health Perspect 112:215–221, PMID: 14754576, 823

https://doi.org/10.1289/ehp.5589. 824

825

24. Hu H, Shih R, Rothenberg S, Schwartz BS. 2007. The epidemiology of lead toxicity in 826

adults: measuring dose and consideration of other methodologic issues 115(3):455-462, 827

PMID:17431499, https://doi.org/10.1289/ehp.9783. 828

829

25. Huang Z, Chen T, Yu J, Qin D, Chen L. 2012. Lead contamination and its potential sources 830

in vegetables and soils of Fujian, China. Environ Geochem Health 34:55–65, 831

https://doi.org/10.1007/s10653-011-9390-6. 832

26. Inglezakis VJ, Stylianou MA, Gkantzou D, Loizidou MD. 2007. Removal of Pb(II) from 833

aqueous solutions by using clinoptilolite and bentonite as adsorbents. Desalination 210(1-834

3):248-256, https://doi.org/10.1016/j.desal.2006.05.049 835

836

837

27. Jagetiya B, Kumar S. 2020. Phytoremediation of lead: a review. In: Lead in Plants and the 838

Environment. Radionuclides and Heavy Metals in the Environment. (Gupta D, Chatterjee S, 839

Walther C, eds). Springer. 179-202. 840

841

28. Johnson DL, Bretsch JK. 2002. Soil lead and children’s blood lead levels in Syracuse, FNY, 842

USA. Environmental Geochemistry and Health 24:375-385, 843

https://doi.org/10.1023/A:1020500504167. 844

29

845

29. Kastury F, Smith E, Lombi E, Donnelley MW, Cmielewski PL, Parsons DW, et al. 2019. 846

Dynamics of lead bioavailability and speciation in indoor dust and x‑ray spectroscopic 847

investigation of the link between ingestion and inhalation pathways. Environmental Science 848

and Technology 53:11486-11495, https://doi.org/10.1021/acs.est.9b03249. 849

850

851

30. Korfmacher KS, Kuholski K. 2007. Do the same houses poison many children? An 852

investigation of lead poisoning in Rochester, New York, 1993–2004. Public Health Reports 853

122:482-487, https://doi.org/10.1177/003335490712200409. 854

855

31. LaBelle SJ, Lindahl PC, Hinchman RR, Ruskamp J, McHugh K. 1987. Pilot Study of the 856

Relationship of Regional Road Traffic to Surface-soil Lead Levels in Illinois. Argonne, IL: 857

Argonne National Laboratory. 858

https://ntrl.ntis.gov/NTRL/dashboard/searchResults.xhtml?searchQuery=DE88013158. 859

[Accessed 12 July 2020]. 860

861

32. Laidlaw M. 2001. Distribution and sources of metals in soils of Marion County, Indiana. 862

Master’s Thesis, Indiana University: Indianapolis, IN. 863

864

33. Laidlaw MA, Mielke HW, Filippelli GM, Johnson DL, Gonzales CR. 2005. Seasonality and 865

children's blood lead levels: developing a predictive model using climatic variables and blood 866

lead data from Indianapolis, Indiana, Syracuse, New York, and New Orleans, Louisiana 867

(USA). Environmental Health Perspectives 113(6):783-800, PMID: 15929906, 868

https://doi.org/10.1289/ehp.7759. 869

870

34. Lal R. 2020. Home gardening and urban agriculture for advancing food and nutritional 871

security in response to the COVID-19 pandemic. Food Sec 2020. 872

https://doi.org/10.1007/s12571-020-01058-3. 873

874

35. Linton RW, Farmer ME, Hopke PK, Natusch DFS. 1980. Determination of the sources of 875

toxic elements in environmental particles using microscopic and statistical analysis 876

techniques. Environmental International 4(5-6):453-461, https://doi.org/10.1016/0160-877

4120(80)90026-4. 878

879

36. Litt JS, Hynes P, Carroll P, Maxfield R, McLaine P, Kawecki C. 2002. Safe yards: improving 880

urban health through lead-safe yards. J. Urban Technol 9:71–93, 881

https://doi.org/10.1080/1063073022000016487. 882

883

37. Magavern S. 2018. Policies to reduce lead exposure: lessons from Buffalo and Rochester. 884

International Journal of Environmental Research and Public Health 15(10):2197, PMID: 885

30304767, https://doi.org/10.3390/ijerph15102197. 886

887

30

38. Maskall JE, Thornton I. 1993. Metal contamination of soils at historical lead smelting sites. 888

Land Contam Reclam 1:92–100. 889

890

39. McBride MB, Shayler HA, Spliethoff HM, Mitchell RG, Marquez-Bravo LG, Ferenz GS, et 891

al. 2014. Concentrations of lead, cadmium and barium in urban garden grown vegetables: the 892

impact of soil variables. Environ Pollut 194:254-261, PMID: 25163429, 893

https://doi.org/10.1016/j.envpol.2014.07.036. 894

895

40. McKay G, Otterburn MS, Sweeney AG. 1980. The removal of colour from effluent using 896

various adsorbents III. Silica: rate processes. Water Research 14(1):15–20, 897

https://doi.org/10.1016/0043-1354(80)90037-8. 898

41. Menezes-Filho J, Carvalho CF, Rodrigues JLG, Araújo CFS, Dos Santos NR, Lima CS, et 899

al. 2018. Environmental co-exposure to lead and manganese and intellectual deficit in school-900

aged children. International Journal of Environmental Research and Public Health 901

15(11):2418, https://doi.org/10.3390/ijerph15112418. 902

903

42. Menke A, Muntner P, Batuman V, Silbergeld EK, Guallar E. 2006. Blood lead below 0.48 904

μmol/L (10 μg/dL) and mortality among US adults. Circulation 114(13):1388-1394, 905

https://doi.org/10.1161/CIRCULATIONAHA.106.628321. 906

907

43. Meroufel B, Benali O, Benyahia M, Benmoussa Y, Zenasni MA. 2013. Adsorptive removal 908

of anionic dye from aqueous solutions by Algerian kaolin: characteristics, isotherm, kinetic 909

and thermodynamic studies B. J. Mater. Environ. Sci 4(3):482-491, ISSN: 2028-2508, 910

CODEN: JMESCN. 911

912

44. Mielke HW. 1999. Lead in the inner cities: policies to reduce children's exposure to lead may 913

be overlooking a major source of lead in the environment. American Scientist 87(1): 62-73. 914

915

45. Mielke HW, Anderson JC, Berry KJ, Mielke PW, Chaney RL, Leech M. 1983. Lead 916

concentrations in inner-city soils as a factor in the child lead problem. American Journal of 917

Public Health 73(12):1366-1369, PMID: 6638229, https://doi.org/10.2105/ajph.73.12.1366. 918

919

46. Mielke HW, Covington TP, Mielke PW, Wolman FJ, Powell ET, Gonzales CR. 2011. Soil 920

intervention as a strategy for lead exposure prevention: the New Orleans lead-safe childcare 921

playground project. Environ Pollut 159:2071–2077, 922

https://doi.org/10.1016/j.envpol.2010.11.008. 923

924

925

47. Mielke HW, Gonzales C, Mielke Jr. PW 2008. Urban soil-lead (Pb) footprint: retrospective 926

comparison of public and private properties in New Orleans. Environmental Geochemistry 927

and Health 30(3):231-242, PMID:17687626, https://doi.org/10.1007/s10653-007-9111-3. 928

929

48. Mielke HW, Gonzales CR, Powell E, Jartun M, Mielke Jr. PW. 2007. Nonlinear association 930

between soil lead and blood lead of children in metropolitan New Orleans Louisiana: 2000-931

2005. Science of the Total Environment 388:43-53, PMID: 17884147, 932

https://doi.org/10.1016/j.scitotenv.2007.08.012. 933

31

934

49. Mielke HW, Powell ET, Gonzales CR, Mielke Jr PW. 2006. Potential lead on play surfaces: 935

evaluation of the “PLOPS” sampler as a new tool for primary lead prevention. Environmental 936

Research 103(2):154-159, https://doi.org/10.1016/j.envres.2006.08.007. 937

938

939

50. Misenheimer J, Nelson C, Huertas E, Medina-Vera M, Prevatte A, Bradham K. 2018. Total 940

and bioaccessible soil arsenic and lead levels and plant uptake in three urban community 941

gardens in Puerto Rico. Geosciences 8(2):43-53, 942

https://doi.org/10.3390/geosciences8020043. 943

944

51. Moir AM, Thornton I. 1989. Lead and cadmium in urban allotment and garden soils and 945

vegetables in the United Kingdom. Environ Geochem Health 11:113-119, 946

https://doi.org/10.1007/BF01758660. 947

948

52. Moyo M, Chikazaza L, Chomunorwa Nyamunda B, Guyo U. 2013. Adsorption batch studies 949

on the removal of Pb(II) using maize tassel based activated carbon. Journal of Chemistry 950

2013, Article ID: 508934. 951

952

53. Murray H, Pinchin TA, Macfie SM. 2011. Compost application affects metal uptake in plants 953

grown in urban garden soils and potential human health risk. J Soils Sediments 11:815–829, 954

https://doi.org/10.1007/s11368-011-0359-y. 955

956

54. Nash D, Magder LS, Sherwin R, Rubin RJ, Silbergeld EK. 2004. Bone density-related 957

predictors of blood lead level among peri- and postmenopausal women in the United States: 958

the third national health and nutrition examination survey, 1988–1994. American Journal of 959

Epidemiology 160(9):901-911, https://doi.org/10.1093/aje/kwh296. 960

961

55. National Gardening Association. 2011. Impact of home and community gardening in 962

America. [Accessed 12 July 2020]. https://garden.org/learn/articles/view/3126/. 963

964

56. Nevin R. 2007. Understanding international crime trends: the legacy of preschool lead 965

exposure. Environmental Research 104(3):315–336, PMID: 17451672, 966

https://doi.org/10.1016/j.envres.2007.02.008. 967

968

57. Olszowy H, Torr P, Imray P. 1995. Trace element concentrations in soil from rural and urban 969

areas of Australia. In: Contaminated Sites Monograph Series 4. South Australian Health 970

Commission. Glenelg Press: Glenelg, Australia. 971

972

58. Özer A, Pirinççi HB. 2006. The adsorption of Cd(II) ions on sulphuric acid-treated wheat 973

bran. Journal of Hazardous Materials 137(2):849–855, 974

https://doi.org/10.1016/j.jhazmat.2006.03.009. 975

976

32

59. Phouthavong-Murphy JC, Merrill AK, Zamule S, Giacherio D, Brown B, Roote C, et al. 2020. 977

Phytoremediation potential of switchgrass (Panicum virgatum), two United States native 978

varieties, to remove bisphenol-A (BPA) from aqueous media. Sci Rep 10:835, 979

https://doi.org/10.1038/s41598-019-56655-w. 980

981

60. Putra RS, Tanaka S. 2011. Aluminum drinking water treatment residuals (Al-WTRs) as an 982

entrapping zone for lead in soil by electrokinetic remediation. Separation and Purification 983

Technology 79(2011):208-215, https://doi.org/10.1016/j.seppur.2011.02.015. 984

985

61. Reagan P, Silbergeld E. 1989. Establishing a health based standard for lead in residential 986

soils. In: Trace Substances in Environmental Health (Hemphill and Cothern, eds). Suppl 987

Environmental Geochemistry and Health 23:199-238. 988

989

62. Sall J, Creighton L, Lehman A. 2005. JMP Start Statistics, third ed. SAS Institute: Cary, NC 990

and Pacific Grove, CA. 991

992

993

63. Schwarz K, Pickett ST, Lathrop RG, Weathers KC, Pouyat RV, Cadenasso ML. 2012. The 994

effects of the urban built environment on the spatial distribution of lead in residential soils. 995

Environ Pollut 163:32–39, PMID: 22325428, https://doi.org/10.1016/j.envpol.2011.12.003. 996

997

64. Soleimanifar H, Deng Y, Barrett K, Feng H, Li X, Sarkar D. 2019. Water treatment residual-998

coated wood mulch for addressing urban stormwater pollution. Water Environ Res 91(6):523-999

535, https://doi.org/10.1002/wer.1055. 1000

1001

65. Turkyilmaz H, Kartal T, Yigitarslan Yildiz S. 2014. Optimization of lead adsorption of 1002

mordenite by response surface methodology: characterization and modification. J Environ 1003

Health Sci Engineer 12(5), https://doi.org/10.1186/2052-336X-12-5. 1004

1005

66. USEPA (United States Environmental Protection Agency). 2000. America’s Children and the 1006

Environment: A First View of Available Measures. 1007

https://www.epa.gov/americaschildrenenvironment [Accessed 13 July 2020]. 1008

1009

67. USEPA (United States Environmental Protection Agency). 2001. Lead;Identification of 1010

Dangerous Levels of Lead;Final Rule. Federal Register 66(4):1206-1240. 1011

https://www.govinfo.gov/content/pkg/FR-2001-01-05/pdf/01-84.pdf. [Accessed 13 July 1012

2020]. 1013

1014

68. USEPA (United States Environmental Protection Agency). 2011. National Emissions 1015

Inventory Data and Documentation. https://www.epa.gov/air-emissions-inventories/2011-1016

national-emissions-inventory-nei-data. [Accessed 13 July 2020]. 1017

1018

1019

69. USFDA (United States Food and Drug Administration). 2010. Total Diet Study Statistics on 1020

Element Results Market Baskets 2006–1 through 2013–4. 1021

https://www.fda.gov/media/77948/download. [Accessed 13 July 2020]. 1022

33

1023

70. USHUD (United States Department of Housing and Urban Development). 1990. 1024

Comprehensive and Workable Plan for the Abatement of Lead-based Paint in Privately 1025

Owned Housing. Report to Congress. http://www.huduser.org/Publications/pdf/HUD-1026

5716.pdf. [Accessed 13 July 2020]. 1027

1028

71. USHUD (United States Department of Housing and Urban Development). 1995. Report on 1029

the National Survey of Lead-based Paint in Housing. http://www2.epa.gov/lead/report-1030

national-survey-lead-based-paint-housing-base-report-june-1995. [Accessed 13 July 2020]. 1031

1032

1033

72. Wagner T, Langley-Turnbaugh S. 2008. Case study: examining the contribution of historical 1034

sources of lead in urban soils in Portland, Maine, USA. Journal of Environmental Planning 1035

and Management 51(4):525-541, https://doi.org/10.1080/09640560802117002. 1036

1037

1038

73. Weisskopf MG, Weuve J, Nie H, Saint-Hilaire M, Sudarsky L, Simon DK, et al. 2010. 1039

Association of cumulative lead exposure with Parkinson’s Disease. Environmental Health 1040

Perspectives 118(11):1609-1613, https://doi.org/10.1289/ehp.1002339. 1041

1042

74. Wu J, Edwards R, He X, Liu Z, Kleinman M. 2010. Spatial analysis of bioavailable soil lead 1043

concentrations in Los Angeles, California. Environmental Research 110:309-317. PMID: 1044

20219189, https://doi.org/10.1016/j.envres.2010.02.004. 1045

1046

75. Yaylah-Abanuz G. 2019. Application of multivariate statistics in the source identification of 1047

heavy-metal pollution in roadside soils of Bursa, Turkey. Arabian Journal of Geosciences 1048

12(12):382, https://doi.org/10.1007/s12517-019-4545-3. 1049

1050

1051

76. Zahran S, Mielke HW, Gonzales CR, Powell ET, Weiler S. 2010. New Orleans before and 1052

after Hurricanes Katrina/Rita: a quasi-experiment of the association between soil lead and 1053

children’s blood lead. Environ. Sci. Technol 44:4433–4440. 1054

https://doi.org/10.1021/es100572s. 1055

77. Zhang SQ, Hou WG. 2008. Adsorption behavior of Pb(II) on montmorillonite. Colloids and 1056

Surfaces A: Physicochemical and Engineering Aspects 320(1-3):92-97, 1057

https://doi.org/10.1016/j.colsurfa.2008.01.038. 1058

1059

1060

78. Zhang H, Wang Z, Zhang Y, Ding M, Li L. 2015. Identification of traffic-related metals and 1061

the effects of different environments on their enrichment in roadside soils along the Qinghai–1062

Tibet highway. Science of The Total Environment 521-522:160–172, PMID: 25835375, 1063

https://doi.org/10.1016/j.scitotenv.2015.03.054. 1064

1065

34

79. Zhou Y, Haynes RJ. 2011. Removal of Pb(II), Cr(III) and Cr(VI) from aqueous solutions 1066

using alum-derived water treatment sludge. Water, Air and Soil Pollution 215(1-4):631-643, 1067

https://doi.org/10.1007/s11270-010-0505-y. 1068

35

Table 1: Polynomial cubic fit of lead concentrations (mg/kg) in residential soil by distance from potential source of Pb based paint.

The nonlinear correlation fit, regression equation, and parameter estimates were conducted using JMP (version 15). Parameters of the

regression equations were estimated and the significance levels of the parameters were expressed as * (p < 0.05), ** (p < 0.01), *** (p <

0.001), **** (p < 0.0001).

Site R2 F

ratio p value Regression Equations

Significant

Parameters

A1 0.9994 632.9 0.0292 𝑦 = 3684 – 575.9 ∗ x + 89.45 ∗ (x − 4.321)2 – 2.277 ∗ (x − 4.321)3 Intercept**, x2**

A3 0.9366 14.78 0.0266 𝑦 = 2141 – 1307 ∗ x + 8370 ∗ (x − 1.045)2 – 5519 ∗ (x − 1.045)3 x2*

A4 0.9868 50.02 0.0197 𝑦 = 2226 – 282.6 ∗ x + 126.5 ∗ (x − 3.569)2 – 28.10 ∗ (x − 3.569)3 x2*

L1 0.9567 22.13 0.0151 𝑦 = 388.3 – 249.5 ∗ x + 497.1 ∗ (x − 2.199)2 – 108.1 ∗ (x − 2.199)3 x2*, x3*

L2 0.8593 18.32 0.0004 𝑦 = 420.3 – 170.1 ∗ x + 277.5 ∗ (x − 2.110)2 – 61.83 ∗ (x − 2.110)3 x2***, x3*

M 0.8402 21.03 <0.0001 𝑦 = 394.3 – 68.31 ∗ x + 135.8 ∗ (x − 3.029)2 – 28.03 ∗ (x − 3.029)3 x2***, x3**

H 0.9820 218.5 <0.0001 𝑦 = 752.9 – 220.8 ∗ x + 127.8 ∗ (x − 2.286)2 – 22.73 ∗ (x − 2.286)3 Intercept****,

x****, x2****, x3*

C1 0.9607 32.63 0.0029 𝑦 = 232.8 – 183.5 ∗ x + 486.1 ∗ (x − 1.181)2 – 246.8 ∗ (x − 1.181)3 x2**

C2 0.9868 150.4 <0.0001 𝑦 = 153.3 – 71.01 ∗ x + 362.7 ∗ (x − 1.371)2 – 268.6 ∗ (x − 1.371)3 x2****, x3**

B 0.9865 97.77 0.0003 𝑦 = 1175 – 713.6 ∗ x + 105.9 ∗ (x − 1.124)2 – 150.5 ∗ (x − 1.124)3 Intercept***, x***,

x3*

G1 0.8887 13.31 0.0081 𝑦 = 361.8 – 71.36 ∗ x + 97.78 ∗ (x − 1.439)2 – 41.25 ∗ (x − 1.439)3 Intercept*, x2*

G2 0.9848 86.41 0.0004 𝑦 = 508.9 – 71.63 ∗ x + 13.23 ∗ (x − 3.458)2 – 0.4656 ∗ (x − 3.458)3 Intercept***, x**,

x2**

Table 2. Pb concentrations in produce or parts of plants (mg/kg) grown in the gardens of the tested residential properties. Data are

expressed as mean (n=3) + one standard deviation. Mean comparison is conducted by one-way ANOVA (F ratio = 121.4; p <0.0001)

followed by Tukey-Kramer honest significant difference test. Mean comparison letters are expressed as superscript and levels not

connected by same letters are significantly different from one another.

36

Produce/Plants Parts of plants

analyzed Site

Pb

concentrati

on in soil

(mg/kg)

Pb

concentration

in plants by dry

weight (mg/kg)

Pb concentration

in plants by

fresh weight

(mg/kg)

Guidance Value

(EC, 2006)

Tomato

(Solanum lycopersicum) Tomato (fruit) Garden bed in C 8.5 0 + 0G 0.02 + 0.009I 0.1

Arugula

(Eruca vesicaria) Leaves Garden bed 1 in A 11.2 3.52 + 0.22 DE 0.19 + 0.016H 0.3

Lettuce

(Lactuca sativa)

Leaves Garden bed 8 in A 28.3 2.98 + 0.10 DE 0.01 + 0.005I 0.3

Radish

(Raphanus sativus) Radish (root) Garden bed 8 in A 28.3 4.67 + 0.33 DE 0.08 + 0.014I 0.1

Garlic

(Allium sativum) Garlic (bulb) Garden bed 7 in A 30.4 1.25 + 0.02 F 0.34 + 0.004G 0.3

Tomato

(Solanum lycopersicum) Stem Garden bed 7 in A 30.4 1.97 + 0.04 F 0.37 + 0.006G 0.3

Potato

(Solanum tuberosum) Root Garden bed 2 in A 32.6 1.88 + 0.07 F 0.23 + 0.008H 0.1

Broccoli

(Brassica oleracea var. italica) Leaves Garden bed 5 in A 52.8 1.79 + 0.07 F 0.48 + 0.012EF 0.3

Ground Ivy

(Glechoma hederacea) Whole plant Front yard in A 7400 16.07 + 0.37 A 3.35 + 0.59A N/A

Elephant Foot Yam

(Amorphophallus paeoniifolius) Whole plant Front yard garden in H 358.7 3.42 + 0.13 E 0.34 + 0.013G 0.1

Elephant Foot Yam

(Amorphophallus paeoniifolius) Whole plant Front yard garden in H 511.4 4.42 + 0.01 DE 1.05 + 0.003D 0.1

Elephant Foot Yam

(Amorphophallus paeoniifolius) Whole plant Front yard garden in H 663.3 4.89 + 0.08D 0.55 + 0.008E 0.1

Elephant Foot Yam

(Amorphophallus paeoniifolius) Whole plant Front yard garden in H 998.9 4.54 + 0.14DE 0.40 + 0.013FG 0.1

Elephant Foot Yam

(Amorphophallus paeoniifolius) Whole plant Front yard garden in H 1042 10.89 + 0.56B 1.22 + 0.054C 0.1

Tomato

(Solanum lycopersicum) Tomato (fruit) Back yard garden in M 985.1 8.40 + 0.30C 1.59 + 0.056B 0.1

37

Table 3. Physicochemical characteristics and TCLP of WTR collected from lagoon 1 in 2017 and lagoon 2 in 2018. Data are

expressed as mean (n = 3) + one standard deviation. Mean comparison is conducted between WTR collected from different space and

time.

TCLP RCRA Metals (mg/L)

Moisture

Content

(%)

pH

Electrical

Conductivit

y (𝞵s/cm)

Organic

matter

content

(%)

Ba Ag As Cr Pb Cd Se Hg

WTR from

Lagoon 1

collected in

2017

48.32 +

1.997

6.99 +

0.064

371.8 +

38.09

2.195 +

0.166

1.157 +

0.0102

0.001 +

0.000

0.086 +

0.023

0.033 +

0.006

0 +

0

0.003 +

0.001

0.036 +

0.006 0 + 0

WTR from

Lagoon 2

collected in

2018

47.99 +

1.962

6.98 +

0.035

450.3 +

52.50

2.259 +

0.157

1.047 +

0.021

0.001 +

0.000

0.098 +

0.035

0.04 +

0.001

0 +

0

0.002 +

0.001

0.036 +

0.006 0 + 0

EPA

permissible

Limit

100 5 5 5 5 1 1 0.2

F ratio 0.042 0.155 4.398 0.235 3.381 NA 0.273 3.811 NA 0.500 0.018 NA

p value 0.848 0.714 0.104 0.653 0.139 NA 0.629 0.123 NA 0.519 0.899 NA

38

Table 4. Percentile, mean, standard deviation (SD) of soil lead level (SLL) (mg/kg) and potential blood lead level (BLL) (g/dL) in

children and potential lead on play surfaces (PLOPS) (g /ft2) using three prediction models developed by Mielke et al., (2007),

Mielke et al., (2006), and Johnson and Bretsch (2002) respectively.

Percentile SLL (mg/kg) SLL Statistics

(mg/kg) Potential BLL (g/dL)

PLOPS

(g /ft2)

Max 75% Median 25% Min Mean SD N Equation 6 Equation 7 Equation 8

Urban

A after

Renovation 173500 7346 1665 455 57 9619 25045 79 9.1 12 4107

A before

Renovation 50000 6811 963 257 82 6857 11768 42 7.4 10.8 2800

C 1553 992 578 144 11 619 469 58 6.2 9.7 1956

M 3743 990 442 253 58 707 759 36 5.7 9.2 1617

H 2378 530 357 181 112 502 490 45 5.3 8.7 1389

G 1200 427 276 198 75 328 206 54 4.9 8.2 1156

B 1077 504 187 131 65 329 326 16 4.4 7.4 874

K 847.7 248 109 82 32 190 180 43 3.8 6.2 589

Suburban

L 3088 1099 558 201 34 753 720 45 6.1 9.7 1909

I 1295 150 116 80 65 202 301 16 3.9 6.4 615

J 115.7 97 82 47 37 76 28 8 3.6 5.6 478

E 248.4 63 45 35 25 66 61 15 3.2 4.4 301

Overall

Urban 173500 1073 413 185 11 3178 12768 373 5.5 9.0 1542

Suburban 3088 711 155 66 25 461 627 84 4.2 7.0 763

39

Figure 1. Field assessment site maps with respect to number of pre-1978 built structures (a), racial diversity (b), education (c), and

annual household income (d) of the sub-counties. These maps were created using Tableau software, which developed a geographical

map of Monroe and Ontario County utilizing latitude and longitude coordinates for each sub-county boundary. In subfigure 1a, the

size of the bubble depicts the number of pre-1978 houses found within that sub-county, and the color of the bubble denotes the

respective sub-county. In subfigure 1b, 1c, and 1d, the distribution of racial diversity, education, and annual income data in their

respective categories was represented by overlaying pie-charts into each sub-county.

Figure 2. Effect of distance (m) from potential source of lead-based paint on the lead concentrations (mg/kg) of residential soils in the

urban or suburban neighborhoods of Rochester, NY; Data are expressed as mean (n = 20) + one standard deviation. Each subfigure

expressed data for one residential property except site A, which was expressed in two subfigures (a, b). The subfigures are arranged

left to right in rows based on their decreasing Y axis values, indicating the magnitude of soil Pb, of which the maximum was recorded

in the front yard of site A after renovation (a) followed by the back yard of site A (b) before and after renovation; site L (c); site M (d);

site H (e); site C (f), site B (g); site G (h); and site K (i).

Figure 3. SLL measurements in the backyard of site G showing two assessment lines; G1 (pink line), which was measured from the

corner of the garage at site G, showed a decrease in soil Pb with distance except on one location (X), which was found to be

horizontally aligned to the corner of the neighbor’s house, indicating it to be an additional source of LBP (fig. S.2). The line G2 was

measured from a point adjacent to the property line between these two houses and was horizontally aligned to the corner of the

neighbor’s house. G2 also showed a continual decrease with distance in the soil Pb concentration, except the increase at the last

measured location point, which is the closest to the main house in site G, suggesting an additional source of LBP. The photograph is

included in the manuscript with the permission of the homeowner.

Figure 4. Multivariate correlation among Pb in outdoor dust, paint and the depth index indicating how deep Pb is in the paint from a

window (a), door (b), and front stairs (c) of site G; discarded old house materials from site A (d); pillars from the front porch in site H

(e); and discarded old house materials left in a barn at site E (f).

40

Figure 5. Kinetic sorption (%) of Pb in presence or absence of WTR. Data are expressed as mean (n = 3) + one standard deviation.

Mean comparison is conducted by two-way ANOVA followed by Tukey-Kramer honest significant difference test. It is conducted

separately for each initial Pb concentration. Levels not connected by same letters are significantly different.

Figure 6. Equilibrium sorption and desorption of Pb in 5% WTR. Linear (a), Freundlich (b), and Langmuir (C) isotherms of Pb-

sorption by WTR in 24h. Data are expressed as mean (n = 3) + one standard deviation. % sorption (d) and % total desorption (e) of Pb

by WTR in three desorption cycles of 24h each. Desorption of Al (mg/kg) from WTR during sorption of Pb (f). Mean comparison is

conducted by one-way ANOVA followed by Tukey-Kramer honest significant difference test. Levels not connected by same letters

are significantly different.

Figure 7. Effect of WTR concentrations on the leaching of lead (µg/L) from the paint chips collected from site A. Data are expressed

as mean (n = 3). Mean comparison is conducted by two-way ANOVA followed by Tukey-Kramer honest significant difference test.

Levels not connected by same letters are significantly different.

Figure 8: GIS maps showing distribution of SLL (mg/kg) in site A before (8a) and after (8b) the restoration/repainting of the built

structure. These maps were created using ArcGIS Pro 2.5.2 by plotting the SLL (mg/kg) measurements into their respective latitude

and longitude. An approximate property-line (dotted line) was overlaid to show the boundaries of the property segregating it from the

side walk in south and west and three built structures of neighboring property in north and east. Based on different guidelines for

safety, 7 sizes and color categories were used, each representing a range of SLL.

41

Figure 1:

Figure 2:

a b

c d

42

0

5000

10000

15000

20000

25000

30000

35000

40000

0 0.5 1 1.5 2 2.5

Le

ad

Co

nc

en

tra

tio

n in

so

il

(mg

/kg

)

Distance from potential source of Pb based Paint (m)

A5 (after renovation)

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

0 2 4 6 8 10Le

ad

Co

nc

en

tra

tio

n in

so

il

(mg

/kg

)

Distance from potential source of Pb based Paint (m)

A1 (beforerenovation)A2 (afterrenovation)A3 (afterrenovation)A4 (afterrenovation)

0

500

1000

1500

2000

2500

3000

3500

0 2 4 6

Le

ad

Co

nc

en

tra

tio

n in

so

il

(mg

/kg

)

Distance from potential source of Pb based Paint (m)

L1 L2

0

500

1000

1500

2000

2500

3000

3500

0 2 4 6 8

Le

ad

Co

nc

en

tra

tio

n in

s

oil

(m

g/k

g)

Distance from potential source of Pb based paint (m)

M

0200400600800100012001400160018002000

0 2 4Le

ad

Co

nc

en

tra

tio

n in

s

oil

(m

g/k

g)

Distance from potential source of Pb based Paint (m)

H

0

200

400

600

800

1000

1200

1400

1600

1800

0 1 2 3

Le

ad

Co

nc

en

tra

tio

n in

s

oil

(m

g/k

g)

Distance from potential source of Pb based Paint (m)

C1 C2

0

200

400

600

800

1000

1200

0 1 2 3Le

ad

Co

nc

en

tra

tio

n in

s

oil

(m

g/k

g)

Distance from potential source of Pb based Paint (m)

B

0

100

200

300

400

500

600

700

800

900

0 2 4 6 8

Le

ad

Co

nc

en

tra

tio

n in

s

oil

(m

g/k

g)

Distance from potential source of Pb based Paint (m)

G1 G2

0

100

200

300

400

500

600

700

0 5

Le

ad

Co

nc

en

tra

tio

n in

s

oil

(m

g/k

g)

Distance from potential source of Pb based Paint (m)

K

a b

c

d

D e f

g h i

43

Figure 3:

44

Figure 4:

500

750

1000

1250

1500

1750

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

1

1.2

1.4

1.6

1.8

Pb in dust

r=0.8202

r=-0.6845

500 1000 1500

r=0.8202

Pb in paint

r=-0.9757

0.1 0.3 0.5 0.7

r=-0.6845

r=-0.9757

Depth index

1 1.2 1.4 1.6 1.8

50

100

150

200

250

300

350

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

1.5

2

2.5

3

Pb in dust

r=-0.6914

r=-0.7013

50 150 250 350

r=-0.6914

Pb in paint

r=0.7785

0.15 0.25 0.35 0.45

r=-0.7013

r=0.7785

Depth index

1.5 2 2.5 3

30

35

40

45

0.15

0.2

0.25

0.3

0.35

0.4

0.45

2

2.5

3

Pb in dust

r=0.8736

r=0.7256

30 35 40 45

r=0.8736

Pb in paint

r=0.9368

0.15 0.25 0.35 0.45

r=0.7256

r=0.9368

Depth index

2 2.5 3

0

500

1000

1500

2000

2500

3000

3500

0

5

10

15

20

1

1.5

2

2.5

Pb in dust

r=0.8894

r=-0.2848

0 1000 2000 3000

r=0.8894

Pb in paint

r=-0.3090

0 5 10 15 20

r=-0.2848

r=-0.3090

Depth index

1 1.5 2 2.5

0

1000

2000

3000

4000

5000

6000

0

5

10

15

0.9

0.95

1

1.05

1.1

Pb in dust

r=0.9128

r=0.0000

0 2000 4000 6000

r=0.9128

Pb in paint

r=0.0000

0 5 10 15

r=0.0000

r=0.0000

Depth index

0.9 0.95 1 1.05 1.1

a b c

d e

f

45

Figure 5.

0

20

40

60

80

100

120

0 5 10 15 20 25

% S

orp

tio

n o

f P

b

Contact Time (h)

0 mg/L Pb 5% WTR 25 mg/L Pb 5%WTR

200 mg/L Pb 5%WTR 1000 mg/L Pb 5%WTR

0 mg/L Pb 0%WTR 25 mg/L Pb 0%WTR

200 mg/L Pb 0%WTR 1000 mg/LPb 0% WTR

Tukey-Kramer HSD

% WTR*Contact time (h)

Initial

Pb

(mg/L)

Contact

time

(h)

WTR

(%)

Mean

Comparison

Letters

25 0.13 5 A

25 1.13 5 A

25 2.13 5 A

25 5.13 5 A

25 10.13 5 A

25 24.13 5 A

25 0.13 0 B

25 1.13 0 B

25 2.13 0 B

25 5.13 0 B

25 10.13 0 B

25 24.13 0 B

220 0.13 5 D

220 1.13 5 C

220 2.13 5 B

220 5.13 5 A

220 10.13 5 A

220 24.13 5 A

220 0.13 0 E

220 1.13 0 E

220 2.13 0 E

220 5.13 0 E

220 10.13 0 E

220 24.13 0 E

1025 0.13 5 D

1025 1.13 5 C

1025 2.13 5 B

1025 5.13 5 A

1025 10.13 5 A

1025 24.13 5 A

1025 0.13 0 FG

1025 1.13 0 G

1025 2.13 0 EFG

1025 5.13 0 E

1025 10.13 0 EFG

1025 24.13 0 EF

46

Figure 6.

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

0 20 40 60 80 100

So

rbe

d P

b b

y W

TR

Q (

mg

/kg

)

Equilibrium Pb Concentration in

Solution Ceq (mg/L)

y = 0.5481x + 3.2092

R² = 0.99743.5

3.6

3.7

3.8

3.9

4.0

4.1

4.2

4.3

4.4

0 0.5 1 1.5 2 2.5

log

Q

log Ceq

y = 4E-05x + 0.0014

R² = 0.9808

0.0000

0.0010

0.0020

0.0030

0.0040

0.0050

0.0060

0 50 100

Ce

q/Q

Ceq

a b c

d e f

47

Figure 7.

0

1000

2000

3000

4000

5000

6000

7000

8000

010

2050

100

Le

ac

he

d L

ea

d in

so

luti

on

(µ

g/L

)

% WTR

1 Day 7 Days

Tukey-Kramer HSD

(WTR % * Time in days)

WTR

%

Time

(days)

Mean

Comparison

Letters

0 1 A

10 1 AB

20 1 BC

50 1 DE

100 1 DE

0 7 A

10 7 CD

20 7 CDE

50 7 E

100 7 E

48

Figure 8:

Figure 8 Footnote: Table S.4 includes a guideline created for the homeowners to explain each color category used in these maps,

associated risk, and the recommended use of that location in their yards.

a b

49

Supplementary Information

Table S.1: Overview of outdoor Pb exposure in the urban and suburban residential properties in Rochester, NY.

Residential

Properties Neighborhood Year

Pb in Paint

(mg/cm2) Depth Index

Maximum

Pb in soil

(mg/kg)

Maximum Pb in

outdoor dust

(g/ft2)

A (Before

Renovation) Urban Unknown 33.7 1.0 50000 Not Measured

A (After

Renovation) Urban Unknown 20.3 6.3 173500 3635

B Urban 1923 2.3 1.0 1077 Not Measured

C Urban 1930 5.9 10 1553 43.39

E Suburban barn 1830 14.1 1.6 NA 8612

F Suburban 2001 0 1.0 0 0

G Urban 1910 6.4 4.4 1200 1969

H Urban 1883 27.7 5.5 2378 1479

1I Suburban 1900 23.0 1.9 390.5 290.2

J Suburban 1954 8.4 6.5 115.7 83.86

K Urban 1930 19.4 9.0 847.7 156.4

L Suburban 1923 33.8 10.0 3088 80.42

M Urban 1910 38.2 4.7 3743 Not Measured

50

Table S.2: Parameters of linear and nonlinear bivariate regression fit of Pb concentrations (mg/kg) in residential soil by distance (m) from

potential source of Pb based paint using data collected from all properties.

Degrees of Fit R2 F ratio p value

1 0.074 11.13 0.0011

2 0.138 11.04 <0.0001

3 0.179 9.992 <0.0001

4 0.209 9.017 <0.0001

5 0.229 8.016 <0.0001

6 0.238 6.986 <0.0001

51

Table S.3: Distribution of soil Pb and associated risks in three yards of site A. Percentile, mean, standard deviation (SD) of soil lead level (SLL)

(mg/kg) and potential blood lead level (BLL) (g/dL) in children and potential lead on play surfaces (PLOPS) (g /ft2) using three prediction

models developed by Mielke et al., (2007), Johnson and Bretsch (2002), and Mielke et al., (2006), respectively.

Percentile SLL

(mg/kg)

SLL Statistics

(mg/kg) Potential BLL (g/dL) PLOPS

(g/ft2)

Equation 8 Max 75% Median 25% Min Mean Std Dev N Equation 6 Equation 7

Front

yard

Before

Renovation 4130 3516 854 392 211 1667 1568 7 7.1 10.6 2575

After

Renovation 173500 26350 12050 3660 373 26167 39086 26 20.9 16.1 16220

Side yard

Before

Renovation 1044 621.3 345.4 201 82 410 286 11 5.2 8.7 1358

After

Renovation 740.6 550.95 367.5 121 61 362 237 9 5.3 8.8 1420

Back yard

Before

Renovation 50000 19625 2899.5 645 232 11326 14045 24 11.3 13.1 6042

After

Renovation 19000 2534.75 1644.5 469 68 2340 3211 42 9.0 11.9 4071

52

Table S.4: A guideline created for the homeowners to explain each color category used in the GIS map (figure 8), associated risk, and the

recommended use of that location in their yards.

Color

categories Based on SLL

Risk

Categories Recommended use

Dark Green

58 mg/kg, which associates with the European Union’s

advisory limit for Pb in vegetables, calculated according to

the prediction models developed in this study.

Minimal to

No Risk

Safer locations to set up yard

gardens

Light Green 100 mg/kg, which is Minnesota’s current soil Pb advisory

and recommended federal standard by scientists

Very Low

Risk

Safer area for the children to

play

Not safe for home gardens

Yellow 400 mg/kg, current federal limit for children’s play area Low Risk

Low exposure risk for the

children to play

Not safe for home gardens

Orange 1200 mg/kg, current federal limit for remaining yard Moderate

Risk

Not safe for the children to

play

Not safe for home gardens

Red Three extents of high concentrations exceeding all

regulatory threshold Risk Hotspots

Should be completely avoided.

Remedial actions are required. Maroon

Black

53

Figure S.1: In situ measurements of soil Pb by XRF and marking the assessment with colored flags, a tool that was used to communicate with the

homeowners to show them the distribution of soil Pb and risk hotspots instantly. Red, green, and blue flags were used to mark SLL > 400, 80 –

400, and < 80 mg/kg, respectively. The photograph is be included in the manuscript with the permission of the homeowner. Photo courtesy:

Michael Fisher.

54

Figure S.2: Lead concentration (mg/kg) in soil at Location A before (a) and after (b) renovation in 2018 and 2019 respectively. Data are expressed

as mean (n=20) and + one standard deviation.

1

2

3

4

5

Pb concentration of soil (mg/kg)

Bulk Soil

Soil with burried paintchips

1

2

3

4

5

6

7

8

9

Pb concentration of soil (mg/kg)

Bulk Soil

Soil with burried paintchips

a b

55

Figure S.3: Multivariate correlation among Pb in outdoor dust, paint and the depth index on discarded painted pieces of wood from site A (a), a

window in site G (b), and a window in site H (c).

10

20

30

40

50

60

70

80

0.5

1

1.5

2

2.5

3

3.5

4

4.5

1.5

2

2.5

Pb in dust

r=0.1586

r=0.0527

10 30 50 70

r=0.1586

Pb in paint

r=0.9484

0.5 1.5 2.5 3.5 4.5

r=0.0527

r=0.9484

Depth index

1.5 2 2.5

56

Figure S.4: Effect of soil Pb (mg/kg) on Pb concentration in plants (mg/kg) collected from all four home gardens of the tested residential

properties and its bivariate regression with a polynomial square fit.

download fileview on ChemRxivChemRxiv Final Pb Manuscript_Das.pdf (3.43 MiB)

Title

Primary prevention of outdoor lead (Pb) exposure on residential properties in Rochester, NY, and

potential of a sustainable remediation solution involving the reuse of drinking water treatment

residual (WTR) waste generated daily by the city.

The Authors

Padmini Das1, Ph.D.; Stephanie M. Zamule1, Ph.D.; Deanna R. Bolduc1; Michelle J. Patton1;

Meghan L. Mendola1; Julia M. Penoyer1, B.S.; Benjamin E. Lyon1; Hanna M. Chittenden1, B.S.;

Alexander C. Hoyt1 B.S.; Cassandra E. Dupre1, B.S.; Jack L. Wessel1, B.S.; Ivan Gergi1, B.S.;

Charlotte V. Buechi1, Jane A. Shebert1, M.S.; Mandeep Chauhan2, M.S.; David Giacherio1, Ph.D.;

and Jacob C. Phouthavong-Murphy3, OMS-III.

Affiliations

1Nazareth College of Rochester. 4245 East Ave, Rochester, NY 14618, USA; 2Northeastern University, 360 Huntington Ave, Boston, MA 02115, USA; 3Touro College of Osteopathic Medicine 230 West 125th Street, New York, NY 10027

Corresponding Author

Padmini Das, PhD. Email: pdas8@naz.edu, phone: 5855000424; 585-389-2552, fax: 585-389-

2672; Address: 4245 East Ave, Rochester, NY 14618, USA.

Competing Interests:

The authors declare they have no actual or potential competing financial interests.

Keywords: Lead (Pb), primary prevention, in-situ risk assessment, XRF, soil, dust, urban

gardens, remediation, chemical immobilization, water treatment residual, WTR, Al-WTR

1

1

2

3

4

5

6

7

8

9

10

11

12

13

1415

16

17

18

19

20

21

22

23

24

25

1

Abstract

Background

Prevention of lead (Pb) poisoning is imperative for public health and environmental justice. This

study presents both assessment and affordable mitigation of Pb exposure risks at individual yard

scales, which are critically important to homeowners to attain primary prevention, but not yet

explored thoroughly.

Objectives and Methods

In phase I, the potential risk of Pb exposure from soil, surface-dust, and vegetables grown in

yard-gardens were measured along with secondary estimation of predicted blood Pb levels in

children in seven urban and six suburban residential properties in and around Rochester, NY,

U.S. In phase II, aluminum-based water treatment residuals (Al-WTR), a waste sludge generated

by the city’s drinking water treatment plant, was investigated for its potential to immobilize Pb as

a remedial measure.

Results

All tested properties with pre-1978 built structures were found to contain Pb in exterior paint at

varying depth, soil, surface dust, and garden vegetables. Soil Pb levels (SLL) were

heterogeneous (11 to 173,500 mg/kg) and higher in urban than suburban properties (mean 3,178

and 461 mg/kg respectively), underscoring the inequitable risk. 71% of vegetables collected

across four property gardens exceeded the European Union’s health-based Pb guidelines and the

median Pb in plants was 72 times higher than the U.S. market-basket values. Very strong

sorption (>90%), minimal desorption (<2%), effective prevention (>85%) of Pb leaching from

high Pb-containing exterior paint chips and estimated Freundlich and Langmuir parameters

suggest practically irreversible Pb-WTR binding.

Discussion

Primary prevention tools were developed to communicate risk with homeowners through GIS-

maps that illustrate risk-categories and prediction models to calculate SLL and predicted Pb in

vegetables using distance from the house. Al-WTR showed very high Pb-immobilizing ability,

2

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

2

suggesting the implementation of this no-cost, zero-waste sorbent as a soil amendment would

provide sustainable primary prevention for low-income urban communities where Pb exposure

prevails.

1. Introduction

Lead (Pb) exposure continues to be the single most critical environmental health issue affecting

American children (CDC 2002), despite the ban on lead-based paint (LBP) and the phasing out

of leaded gasoline that took place four decades ago (USEPA 2011; Wagner and Langley-

Turnbaugh 2007). Lead is a potent neurotoxicant particularly detrimental to developing fetuses

and children under the age of six, as it causes cognitive and behavioral impairments, learning

disabilities, attention deficit, juvenile delinquency, and violent or even criminal behavior (CDC

2002). The number of children with elevated blood lead levels (EBLL) has been reduced due to

the implementation of secondary interventions such as testing BLL in children followed by

tracking and mitigating the hazard source. However, the problem remains unaddressed in two

critical areas.

Firstly, it is increasingly evident from recent research that although EBLL-based secondary

prevention has been effective in reducing exposure to some extent, it comes at a cost to children

who have already been exposed, and who pay a steep price in irreparable systemic damage

through cumulative exposure (Crinnion and Pizzorno 2019). 80% of the total body Pb in

children, and 90% in adults, is stored in bone; and while the half-life of Pb is only thirty-five

days in blood, it is eight to forty years in bone (Hu et al. 2007). Pregnant and postmenopausal

women undergo increased bone turnover, causing Pb poisoning of the fetus and Pb-related age-

associated problems in adult women (Gulson et al. 2007; Nash et al. 2004). Cumulative exposure

through bone-Pb turnover continues throughout life causing neuroinflammatory diseases like

Alzheimer’s and Parkinsonism at variable BLL (Bakulski et al. 2012; Weisskopf et al. 2010);

increased cardiovascular mortality in BLL as low as 3.62 µg/dL (Menke et al. 2006); and anxiety

and depressive disorders in young adults with BLL as low as 1.7 µg/dL (Bouchard et al. 2009).

The BLL action limit for children has recently been reduced from 10 µg/dL to 5 µg/dL by U.S.

Centers for Disease Control and Prevention, despite their acknowledgment that there is no safe

limit for BLL in children; however, only 18 states (not including NY) have adopted this standard

3

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

3

thus far (CDC 2018). Moreover, USEPA regulatory limits for residential soil Pb levels (SLL)

(400 and 1200 mg/kg for bare soil in children’s play areas and the remainder of the yard,

respectively; USEPA 2001) are associated with BLL higher than 5 µg/dL according to two

prediction models that independently showed a steep increase in BLL at lower soil Pb levels

(Mielke et al. 2007; Johnson and Bretsch 2002). These guidelines, which are substantially higher

than the soil Pb standards set by some European nations (40 to 150 mg/kg) and Canada (140

mg/kg), have remained unchanged for decades, despite recommendations by researchers to lower

them to safer limits (Mielke et al. 2008; Reagan and Silderbard 1989; Bell et al. 2010).

Secondly, the current state of residential Pb exposure poses critical questions related to equality

and social justice. Despite a substantial drop in the national average, EBLL still prevails in

densely populated urban communities, meaning that children from lower-income minority

families suffer most from Pb poisoning (Mielke 1999; Geltman et al. 2001). Fillippelli et al.

2005 reported that 15% of children living in urban communities exhibit BLL’s above 10 µg/dL,

compared to only 2.2% of children nationally. Similar trends were observed in western New

York state cities, like Rochester and Buffalo, where most frequent Pb exposure was found to

occur in low-income urban neighborhoods, where people of color are highly concentrated

(Magavern 2018). The established correlation between Pb exposure and crime in teens and

young adults in low-income urban communities (Nevin et al. 2007) underscores the

socioeconomic injustice of this issue and the importance of developing a sustainable intervention

that will address the environmental, social, and economic aspects of this problem equitably.

Primary prevention, which directly measures and corrects hazards before an exposure happens

(CDC 2005), is the only process that can provide an equitable solution to the Pb exposure issue

without vulnerable populations being disproportionately affected as collateral damage.

Sustainable measures of primary prevention are not easy to attain, as they involve a number of

multifaceted tasks and challenges. A complete assessment of SLL distribution, outdoor surface

dust, and property garden vegetables at the individual property scale is critically important for

homeowners to understand the contamination pattern, associated health risks, and potential

remedial strategies (Clark and Knudsen 2013). Primary contributors of EBLL in children are Pb

in soil and house dust that consists of 20-80% resuspended Pb from the soil, along with other

dietary sources. (Mielke et al. 2007; Wu et al. 2008; Bugadalski et al. 2013). One-third of U.S.

4

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

4

residents grow at least one vegetable in their yard for intended consumption (National Gardening

Association 2011), heightening the risk of Pb exposure from contaminated yard soils (Clark and

Knudsen 2013). This percentage is expected to increase due to the recent COVID-19 pandemic

given the disrupted access to fresh and nutritious foods (Lal 2020). Further, an effective and

affordable remediation solution must be provided to the families of lower socioeconomic

standing to prevent Pb exposure at identified hot spots. Surface soils act as a sink for

accumulating Pb over years and can remain so for centuries in the absence of proper remediation

and intervention (LaBelle et al. 1987; Datko-Williams et al. 2014). Although the excavation of

the surface soil with high Pb is the best practice to remove the risk (Mielke et al. 2011), this

process is not economically feasible for homeowners in lower socioeconomic urban

neighborhoods. The lack of access to risk assessment measures and economically-feasible

alternatives for Pb removal contributes to the ongoing inequitable Pb poisoning in children.

To address primary prevention, we investigated thirteen residential properties in and around

Rochester, NY, United States, utilizing three unique approaches that differ from earlier studies.

As opposed to evaluating commercial and residential locations on a city-wide scale, we focused

on a site-specific scale of individual properties with structures built prior to 1978 as the principal

outdoor Pb sources. Furthermore, we determined the spatial distribution of SLL, dividing every

property into broader risk categories such as: safe areas with minimal to no risk; low-risk areas

safer for children to play, but not safe enough to grow edible produce; and high-risk areas with

Pb hot spots, which should be avoided and considered for remedial actions. Using this data,

primary prevention tools including mathematical prediction models and Geographical

Information System (GIS) maps were developed to provide homeowners with a thorough

understanding of the risk distribution on their properties. Finally, to address the need for a cost-

effective remediation strategy to reduce exposure risk from the Pb hot spots, we explored the

potential reuse of a waste sludge as a soil amendment to chemically immobilize Pb. This water

treatment residual (WTR) is generated as a waste product on a continual basis by the city’s

largest drinking water treatment plants. Successful implementation of this technology will

provide the city with a sustainable solution for reducing costs of waste management as well as

provide homeowners with a low-cost solution to reduce the risk of Pb exposure from high-risk

property hot spots. To the best of our knowledge, this study is the first to focus on both aspects

of primary prevention by directly measuring the outdoor Pb levels at individual yard scales in

5

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

5

urban and suburban residential properties and proposing a low-cost, zero-waste solution for the

homeowners by reusing a waste that the city generates every day.

2. Methods

2.1. Phase I: Assessment

2.1.1. Field Site Descriptions

Seven urban and six suburban residential properties were selected based on varying

characteristics including the age of the property’s structures; socioeconomic categorizations of

the homeowners such as race, education, and household income; and the willingness of the

homeowners to participate in this study. Tableau data visualization software (version 2020.1)

was used to create site location maps with respect to the number of houses built prior to 1978 and

the socioeconomic standing of the corresponding areas of the sub-county (figure 1). The

software was used to develop a geographical map of Monroe and Ontario counties utilizing

latitude and longitude coordinates for each sub-county boundary. An urban location was defined

by a geographical location within the city limits of Rochester, NY, United States. A suburban

location was defined as a geographical location outside the city limits and within the county

limits of Monroe or Ontario counties. The socioeconomic categorization data was collected from

the Census Bureau for each respective sub-county.

2.1.2. In-situ measurement of soil Pb using XRF

Soil lead concentrations were measured on-site using a USHUD certified Thermo ScientificTM

NitonTM XLp 300 series x-ray fluorescence analyzer (XRF). The XRF was set to “standard bulk”

mode while an uncovered soil surface was chosen to conduct the measurement in each location

(n = 20). The soil surface locations included: i) alongside structures including corners and

beneath doors, front stairs, and windows; ii) along a perpendicular line starting from a high soil

Pb point adjacent to the structure towards the property line; iii) adjacent to sidewalks and near

structures of neighboring properties. Additional parameters including variation in slopes and

presence of loose paint chips were noted if observed. Once recorded, each measurement was

marked with flags with different colors (figure S.1), a process used to communicate with the

6

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

6

homeowners about the distribution of soil Pb on their property and potential risk associated with

it.

2.1.3. Measurement of Pb in paint and outdoor dust

Exterior paint was measured in-situ using the “paint mode” of the XRF on outdoor surfaces such

as siding, doors, steps, window sills, pillars, and old housing materials stored outside and inside a

barn, (site E only). Calibration of the XRF was performed according to the HUD calibration

protocol. Levels of Pb in the paint on a given surface area and a depth index indicating how close

the Pb was to the surface were obtained as part of each measurement.

Outdoor dust collection was completed using Fisher ScientificTM dust wipes (catalog number:

NC0309992) on any surface which registered an XRF lead reading. To ensure maximum

collection, the dust wipes were unfolded for the initial collection, then folded in half and used

within the same surface area. Pb levels on each dust wipe were measured in the laboratory using

the “dust wipe” mode of the XRF by placing each wipe in a Niton Portable Test Stand and

scanning for 80 seconds total (20 seconds for each position of the holder).

2.1.4. Collection of produce/plant samples and preparations

Four assessed properties (A, H, C, M) had home gardens. Produce and plant samples were

collected with the permission of the homeowners. Plant samples were cleaned and weighed using

a Fisher Science Education analytical balance, dried in a ThermoScientific Heratherm OGS100

oven at 105°C for 24 hours (or until completely dry), and then weighed again to calculate

moisture content. A dry weight (maximum of 0.5 g or based on the available mass of produce) of

plant sample was digested in Fisher Scientific Trace Metal Grade HNO3 following the US EPA

3051 protocol in a CEM MARS6 digestion microwave unit. Once digested, samples were filtered

through Thermo scientific 0.45 µm PTFE-L Luer-Lock syringe filters, diluted to 50 mL with RO

water from a Thermofisher Scientific Barnstead Nanopure purifier, and centrifuged at room

temperature in an Eppendorf 5424R for 15 minutes at 4,000 rpm.

2.2. Phase II: Remediation

2.2.1. WTR Collection, Characterization, and TCLP

7

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

7

Two batches of WTR were collected from two sludge lagoons of Monroe County Water

Authority’s drinking water treatment plant in Greece, NY, U.S. in 2017 and 2018. These lagoons

contained the post-flocculation waste sludge from Lake Ontario source water after being treated

with alum (aluminum sulfate). WTR samples were then thoroughly mixed, air-dried, ground into

powder, and passed through a 2 mm sieve. Triplicate samples collected from each batch of WTR

were used for determining physicochemical properties, including pH, EC, moisture content, and

organic matter content. Toxicity characteristic leaching procedures (TCLP) were performed

using EPA SW-846 methods 1311.

2.2.2. Sorption and Desorption of Pb by WTR

Kinetic and equilibrium sorption and desorption experiments were carried out in triplicate using

a 1:20 solid to solution (m/v) ratio (SSR) of WTR to a lead-spiked solution made from lead

nitrate and 0.01M KCl as a background electrolyte. Kinetic sorption/desorption experiments

were conducted at four initial Pb concentrations (0, 25, 200, and 1000 mg/L) for 24h with six

intermediate sample collections after 0, 1, 2, 5, 10, and 24 h during end-over-end mixing on an

Analytical Testing Model DC-20 Rotary Agitator shaker at 250 rpm. Similarly, equilibrium

sorption experiments were conducted at six initial Pb concentrations (0, 200, 400, 600, 800, and

1000 mg/L) for 10 h using the same sampling method. Three subsequent desorption cycles (24h

each) were carried out to determine the extent of the potential release of pre-adsorbed Pb from

WTR. In addition to shaking time, the WTR remained in contact with the Pb solution for

approximately 8 min (0.13 h) during the sample preparation. This time in combination with the

shaking time was represented as the contact time for kinetic data. After the respective sorption or

desorption processes, samples were centrifuged using an Eppendorf 5804R for 15 minutes at

4000 rpm, filtered through Thermo scientific 0.45 µm PTFE-L Luer-Lock syringe filters, and the

supernatants were analyzed for Pb and Al.

2.2.3. Optimizing WTR to prevent paint chip Pb leaching

Loose paint chips were collected from the yard of site A both before and after the home was

renovated, in order to conduct two experiments. First, a TCLP test was conducted to estimate the

Pb leaching potential from these scattered paint chips in simulated landfill conditions following

the EPA SW-846 methods 1311. Second, a batch desorption experiment in the presence of

8

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

8

varying levels of WTR was performed to determine the optimal amount of WTR necessary to

effectively prevent Pb leaching. Four WTR-to-paint chips (m/m) ratios (10, 20, 50, and 100)

were added to a 0.01 M KCl solution while maintaining a 1:20 SSR. Samples were equilibrated

for 7 d with two desorption cycles (1 and 6 d) using end-over-end mixing. The samples were

centrifuged at 4000 rpm for 15 mins, filtered through a Thermo scientific 0.45 µm PTFE-L Luer-

Lock syringe filter, and the supernatants were analyzed for Pb.

2.3. Pb Analysis

In phase I, Plant samples were analyzed for lead content in triplicate using an Agilent

Technologies 4210 MP-AES. The MP-AES was calibrated with standards ranging from 2500 to

10000 μg/L prepared from a stock solution of Agilent Technologies 1000 μg/mL Pb in 5%

HNO3. Method analysis settings were as follows: pump speed of 25 rpm, sample uptake time of

15 seconds, stabilization time of 15 seconds, read time of 30 seconds, and nebulizer flow of 0.8

L/min.

In phase II, Pb in 5% of in-situ soil samples (following USEPA method 6200), Pb and Al from

the batch sorption/desorption, and Pb from leaching experiments were measured using a Perkin

Elmer 400 graphite flame Atomic Absorption Spectrophotometer (AAS). After TCLP, the eight

RCRA metals (Arsenic, Barium, Cadmium, Chromium, Lead, Mercury, Selenium, and Silver)

were measured using a Perkin Elmer Optima 8000 inductively coupled plasma optical emission

spectroscopy (ICP-OES) to attain data at lower detection limits. Calibration coefficient limits

were set to 0.995, with an allowed calibration error of 10%, and quality control samples were

accepted within +5% error margins.

2.4. Data analyses, Modeling, and Mapping

Statistical analyses were carried out using JMP IN version 13.0 (Sall et al. 2005). One-way and

Two-way ANOVA were performed as required, followed by mean comparisons using the Tukey–

Kramer honest significant difference (HSD) test. Multivariate correlation analyses of Pb in

surface dust were conducted with Pb in paint and depth index; bivariate correlation analyses of

Pb in edible produce or plant parts were conducted with the adjacent soil Pb levels. Two types of

prediction models were developed using the relationship between i) soil Pb over distance from

9

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

9

the structure(s) at each site and ii) Pb in produce/plants grown in the yard gardens with the

adjacent soil Pb concentrations. Sorption data were fit to both Freundlich and Langmuir isotherm

models to gain insight on possible mechanisms of sorption between Pb and WTR and the

reversibility of this binding. XRF measurements of SLL were taken along with the latitude and

longitude of each sampling location and the collected readings were imported into ArcGIS Pro

2.5.2 and portrayed with proportional symbols, color categories, and heat maps.

3. Results

3.1. Outdoor Pb-exposure in urban and suburban residential properties

As described in fig. 1 and table S.1, seven urban and six suburban residential properties built

between 1830 and 2003 were tested for outdoor Pb exposure sources. Among those, the house at

site F was the only structure built after 1978. As expected, there was no Pb found in exterior

paint, dust, or natural soil at this property. Interestingly, low soil Pb levels (27.8 + 0.98 mg/kg)

were found at two sites in the garden area, both of which were covered with store-bought potting

soil and mulch, suggesting that either may have contributed Pb to the garden. Two suburban sites

(I and J) were found to have Pb in exterior paint and soil; however, the SLL in these two

properties were below 400 mg/kg. All seven urban (A, B, C, G, H, K, M) sites and one suburban

(L) site were found to have SLL higher than 400 mg/kg. Table S.1 shows the Pb in paint, its

depth index, and the maximum concentration of Pb in soil and dust.

3.1.1. Distribution of Pb in soils

The distribution of SLL was extremely heterogeneous in all properties assessed and thus was

explored further in eight properties that surpassed the federal limit. Three factors were found to

be controlling this distribution; i) distance from the most prominent source of LBP, ii) slopes and

iii) the presence of a secondary source at close proximity. Among these, the distance from the

primary source of LBP was found to be the dominant factor controlling the distribution of SLL.

The maximum SLL were found alongside the source structures at all sites, indicating the most

likely primary source was LBP that leached into the soil over the years (figure 2). The data

shows a continual decrease of soil Pb over distance, dropping below 100 mg/kg in all properties

at variable distances from the houses. However, a few exceptions were noted in A2, L1, G1, G2,

10

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

10

and K. These variations in SLL against the effect of distance were observed due to the presence

of loose paint chips at close proximity to the measured locations in A2 and slopes in L1. Both

sites G and K were inner-city residential properties with smaller yards where the structures of the

neighboring property influenced the distribution of SLL by contributing as secondary sources of

Pb (figure 3).

In each yard, SLL showed a biphasic pattern: initially steep and then a gradual decrease over

distance from the source structures. This trend was utilized to create site-specific and overall

regression models. Site-specific models based on polynomial cubic regression equations are

presented for seven yards in table 1, which exhibited both significance as well as strength (R2

ranging from 0.84 to 0.99). In contrast, the overall bivariate regression models across all sites

were significant for linear (p = 0.0011) as well as nonlinear polynomial (n = 2 to 6) fit (table

S.2), but neither were acceptable to be used, as the regression coefficients were too low (no more

than 0.24).

3.1.2. Pb in outdoor surface dust

Figure 4 shows the multivariate correlation of surface dust collected from different surfaces with

Pb in paint and depth index. All surfaces showed a significant (p < 0.05) correlation with both Pb

in paint and depth index. Interestingly, Pb was found in all of these surfaces at a low depth index,

suggesting the presence of Pb closer to the surface that impacted its loading in the dust. Figure

S.3 shows a similar multivariate correlation for surfaces with a higher depth index and the

surface dust only showed a significant correlation with Pb in paint but not with depth index.

Surface dust measurements in all urban properties assessed (with the exception of site C)

exceeded HUD’s action level of 800 µg/ft2 for exterior concrete. In contrast, surface dust

measurements were within this limit in all suburban properties tested, with the exception of a

barn at site E which showed a very high risk of potential Pb exposure through inhalation of dust.

The assessment of Pb was conducted inside the barn built in 1830; however, the barn is

considered an outdoor source as it was not located adjacent to the house built in 1966. The barn

is currently used to store materials from a previously demolished structure and seldom used as a

play area for visiting grandchildren of the current property owners. All 19 objects tested in this

barn (including the wall, doors, wall decor, and various old materials) contained Pb in the paint

11

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

11

with an average depth index of 1.66 + 0.62, showing most of the Pb is on the surface. All 19 dust

samples collected from this site were found to contain Pb averaging 1,551 µg/ft2, almost twice

the level of HUD’s action level. The highest concentration of Pb-containing surface dust reported

in this study (site E) was 8,612 µg/ft2, which is 11 times higher than the federal guideline,

showing the extent of the hidden risk of potential Pb exposure through inhalation of dust.

3.1.3. Pb in produce/plants in outdoor home gardens

Table 2 shows the plant Pb concentrations (mg/kg) and the Pb levels of the adjacent soil in four

yard-gardens at sites A, C, H, and M. Tomatoes collected from the garden of Nazareth College of

Rochester, NY, which had adjacent SLL below detection limits, were used to calculate the

background plant Pb levels of plants from the property sites. Pb concentrations in tomatoes from

the Nazareth College garden were subtracted from the Pb levels found in plants from the

properties tested to focus on the effect of SLL on the Pb uptake and accumulation in plant

tissues. Tomato plants were found in three gardens (sites A, C, and M), and the lead

accumulation in tomatoes significantly (p<0.0001; F ratio=750.7) increased in a linear manner

(R2 = 0.99) with increasing SLL. Site A had several garden beds filled with potting soil which

contained soil Pb, but at much lower levels than 400 mg/kg. Currently, there is no existing

guideline defining the permissible level of Pb in the soil for gardening. All plants collected from

property garden beds, including arugula, lettuce, radish, garlic, tomato, potato, and broccoli,

were found to contain high concentrations of Pb, ranging from 1.25 mg/kg in garlic to 4.67

mg/kg in radish (by dry weight). Sites H and M had gardens in locations with higher SLL levels

compared to other locations in their respective properties. Consequently, the tomatoes from site

M and elephant foot yam from site H exhibited a very high accumulation of Pb. Overall, as

shown in figure S.4, the increase in accumulated Pb in plants collected from all four yard-

gardens over adjacent SLL followed a polynomial square fit. Equations 1 and 2 present two

prediction models that were derived based on the strong correlation between the soil Pb and Pb

in plant tissues per dry weight (d.w.) and fresh weight (f.w.) respectively.

x−812.1¿2

y=2.331−0.005∗x+5.74e−7∗¿ --------Equation 1

12

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

12

x−812.1¿2

y=2.338−0.0008∗x+6.99e−8∗¿ --------Equation 2

Where y is Pb concentration in (mg/kg) in plant tissues and x is Pb concentration (mg/kg) in the

adjacent soils. Both models were significant (p < 0.0001; F ratio = 120.5 and 47.7 respectively)

with strong regression coefficients (R2 = 0.85 and 0.89 respectively).

3.2. Chemical immobilization of Pb by WTR

3.2.1. Physico-chemical and TCLP characteristics of WTR

Table 3 presents the relative physicochemical properties and the results of the TCLP of the two

batches of WTR. Both exhibited almost neutral pH, high moisture content, and low organic

matter content. TCLP levels of the RCRA metals revealed a complete absence of lead and

mercury and the presence of other metals in substantially lower concentrations than the criteria

set by the EPA. Both physicochemical properties and TCLP data for these two batches of WTR

were statistically similar, which was evident in p > 0.1 and low F ratio.

3.2.2. Pb kinetic and equilibrium sorption/desorption by WTR

Kinetic sorption experiments showed that the binding of Pb by WTR was rapid and varied with

the initial Pb concentrations in solution, while the Pb levels in the control solutions with no WTR

remained unchanged (fig. 5). Equilibrium times could not be determined in lower Pb

concentrations because of WTR’s extremely rapid and strong sorption for Pb, as the sorption

reached 100% instantaneously (within 8 minutes of contact time with no shaking) for 25 mg/L

and within 5.13 h contact time in 200 mg/L initial Pb concentrations. At 1000 mg/L initial Pb in

solution, the Pb sorption by WTR reached equilibrium within 5.13 h contact time and did not

exhibit any significant change in sorbed Pb up to 24.13 h contact time. Absolutely no desorption

occurred in any of these initial concentrations within 24 h.

The extent of equilibrium sorption of Pb by WTR increased with increasing Pb concentration in

solution (fig. 6a), but the percent sorption showed a reverse trend. However, very high sorption

affinities (> 90%) were exhibited in all initial loads within 10h (fig. 6d). The data followed an L-

type curve (fig. 6a) and showed strong fit to the linearized Freundlich (Eqn. 2; fig. 6b; R2 = 0.99)

13

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

13

as well as (Eqn. 3; fig. 6c; R2 = 0.98) Langmuir equations and the associated parameters were

calculated accordingly.

log Q=log K F+(1/n) logC eq -------Equation 3

Where KF (mg/kg) and n are the Freundlich equilibrium constants representing the sorption

capacity and adsorption intensity respectively; Q (mg/kg) is the adsorbed amount of Pb by WTR

at equilibrium; Ceq (mg/L) is the equilibrium concentration of Pb in solution. The value of KF

was determined as 1.6 mg/g, which exhibited a very high sorption capacity of Pb by WTR. The n

value determined in this study was above unity (1.82), indicating favorable adsorption of Pb onto

WTR through physical processes (McKay 1980; Özer and Pirinççi 2006).

(Ce/Q)=(1 /K L Qm)+(Ce /Qm) --------Equation 4

Where Qm (mg/kg) is the maximum adsorption capacity and KL (L/mg) represents the Langmuir

constant related to the energy of adsorption. A good fit to the Langmuir isotherm indicates the

monolayer coverage and homogeneous distribution of Pb on the active binding sites of the WTR

surfaces with an adsorption maximum of 25 mg/g, that exhibits a very high sorption capacity.

Additionally, another Langmuir constant named separation factor (RL) was calculated following

equation 4.

RL=1/1+K LC0 --------Equation 5

Where C0 (mg/L) is the initial Pb concentration. The RL values indicates the nature and the

feasibility of the adsorption process, such as unfavorable sorption if RL > 1, linear sorption if RL

= 1, favorable sorption if 0 < RL <1, or irreversible sorption if RL = 0 (Meroufel et al. 2013). The

value of RL calculated for all tested initial concentrations of Pb in this study ranged from 0.03 to

0.15, which were between zero to unity, suggesting favorable sorption of Pb on WTR. In

addition, these values were much closer to zero than one, indicating more irreversibility in

adsorption of Pb on WTR. This was evident in the equilibrium desorption of Pb from WTR.

Figure 6e shows the total desorbed Pb after three desorption cycles (24 h each). No desorption

was noted in 72 h up to 600 mg/L initial Pb concentration, indicating complete hysteresis or

complete irreversible binding of Pb by WTR at these initial loads. Although with the further

14

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

14

increase of initial Pb, the extent of desorption increased; it remained restricted to < 2% of the

sorbed Pb in WTR even after 72 h of shaking.

3.2.3. WTR concentration effects on paint chip Pb leaching

As evident from figure S.1, the numerous loose paint chips found in the yard at site A both before

and after the renovation of the structure resulted in an extremely high SLL. Locations with

buried paint chips exhibited an average of 12 and 5 times higher SLL as compared to the

surrounding soil with no paint chips before and after renovation, respectively. The surface soil

with both loose and buried paint chips posed a very high risk of Pb exposure through dust

loading as well as leaching into the stormwater runoff. A TCLP assessment of these paint chips

collected from the yard revealed that they released 47.9 + 1.71 mg/L (n=6) of Pb in solution,

which is 9.6 times higher than the EPA permissible Pb levels for TCLP. This data highlights the

need for determining potential Pb leaching from these paint chips in a stormwater solution and

provided us with an ideal opportunity to test the effectiveness of WTR to prevent the

mobilization of Pb from the paint chips with extremely high Pb-leaching potential.

The results were highly promising as WTR concentrations significantly (p<0.0001 and F ratio =

90.7) reduced Pb leaching from loose paint chips. 43, 61, 84, and 93% prevention of Pb leaching

was achieved by 10, 20, 50, and 100% WTR to paint chips (m/m) application after two

desorption cycles of 1 and 6 d. Effect of time was masked (p=0.2) by WTR’s very strong affinity

for Pb; no significant difference between the higher WTR treatments suggests the application of

WTR in a 1 to 2 ratio to paint chips would be enough to achieve the optimum prevention. As

these paint chips exhibited the maximum Pb concentrations and acted as the major contributor of

SLL in site A, a lower dose of WTR would be enough for achieving optimum prevention of Pb

leaching in soils at lower SSR.

4. Discussion

4.1 Field Assessment

15

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

15

Although the City of Rochester succeeded in reducing the number of children with EBLL

(compared to neighboring cities like Buffalo and Syracuse) through its adoption of an aggressive

secondary intervention approach, the city continues to face significant issues related to Pb

exposure (Magavern 2018). This is consistent with our findings showing that the overall SLL

medians across all tested urban and suburban properties with structures built prior to 1978 were

413 and 155 mg/kg, respectively, with mean SLL as high as 3,178 and 461 mg/kg, respectively.

This study also documented the highest maximum residential SLL (173,500 mg/kg), as

compared to those reported over the last two decades in urban and suburban areas of thirty-one

U.S. cities (data compared with 36 research articles; not shown). Risk of Pb exposure through

soil Pb, potential dust loading, and produce grown in property gardens was found to be much

higher in the urban residential properties where residents already face significant socioeconomic

challenges. According to the 2010 Rochester census data, urban neighborhoods in the City of

Rochester have the highest percentage of residents living below the poverty line, the lowest

percentage of those with above-average income, and the broadest distribution of race with the

highest percentage of African American residents when compared to suburban sub-counties. The

unjust burden of Pb exposure in urban neighborhoods in which many low-income minority

families reside has been documented in twenty studies conducted over the last four decades using

data from sixteen U.S. cities (Datko-Williams et al. 2014).

The prevalence and correlation of SLL and BLL are well established in the literature (Bickel

2010; Zahran et al. 2010; Wu et al. 2010). The overall median SLL in urban properties exceeding

the federal standard clearly underscores this risk. However, the lower median SLL of suburban

properties does not represent a risk-free condition either. Adverse health effects of Pb are

associated with BLL as low as 2 mg/dL, which is associated with SLL < 100 mg/kg (Canfield et

al. 2003; Schwarz et al. 2012). Two existing prediction models by Mielke et al. (2007) and

Johnson and Bretsch (2002) independently showed a steep increase in BLL at SLL <100 mg/kg

and a gradual rise at SLL >300 mg/kg. To assess the exposure risk more specifically, we used

both of these models (equations 5 and 6) to predict the potential BLL if residents (especially

children) are being exposed at tested residential properties.

SLL median¿0.5

BLL=2.038+0.172∗¿ -------Equation 6 (Mielke et al. 2007)

16

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

16

BLL=2.097∗ln(SLL median)−3.6026 -------Equation 7 (Johnson and Bretsch 2002)

SLL in all but one suburban property was predicted to result in BLL > 5 µg/dL according to

equation 7, whereas five out of the seven urban properties and one out of the five suburban

properties were predicted to result in BLL > 5 µg/dL according to equation 6 (table 4). We

observed that the predicted BLL based on the median SLL of the entire property underestimates

the predicted risk for properties with high variation in SLL throughout the property. For instance,

as discussed earlier, site A showed a high variation in SLL throughout the property, both before

and after renovation. The predicted BLL of the front yard after the renovation was as high as

20.9 µg/dL (equation 6) and 16.1 µg/dL (equation 7), both of which are much higher than the

predicted BLL based on the median SLL of the entire property. In such cases, yard-specific

predictions would be a more appropriate measure of risk assessment. Racial and socioeconomic

distribution in the City of Rochester, along with the SLL and estimated BLL documented in this

study, echo the findings of Mielke et al. (1999) and Geltman et al. (2001), who reported that

higher SLL in urban soils result in inequitable EBLL in children of lower-income families,

minorities, and recent immigrants. This is also evident from the data shown by Korfmacher

(2007) that low-income families in the crescent area in urban Rochester characterized by high

poverty, crime, and pre-1950 rental housing correlated with 23.6% of children under 6 years of

age having blood lead levels greater than 10 µg/dL.

In accordance with this study, extreme heterogeneous distribution of soil Pb has been

documented by other researchers, and has been observed to decrease with increasing distance

from three Pb source categories: i) industrial sites or brownfields with residues of historic Pb use

(Marshall and Thornton 1993; Wu et al. 2010), ii) areas near highways or roads with intense

traffic density (Mielke et al. 2008; Laidlaw 2001; Filippelli et.al. 2005), and iii) aged structures

with exterior LBP (USEPA 2000; Litt et al. 2002; and Clark and Knudson 2013). Some studies

have reported that the combustion of leaded gasoline is the primary source of Pb in larger urban

settings, as noted in Baltimore (Mielke et al. 1983), New Orleans (Mielke et al. 2008), and

Indianapolis (Laidlaw 2001; Filippelli et.al. 2005). Clark and Knudson (2013) argue that exterior

LBP dominates the distribution of soil Pb in smaller urban communities like Appleton, WI. Our

study showed that even in a medium-sized urban community like Rochester, with a history of

past industrialized use of heavy metals, the primary source of Pb in all tested sites was exterior

17

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

17

LBP from structures built prior to 1978, as none of these sites are located in neighborhoods with

high traffic density or at close proximity of any brownfield or superfund site. The building

structures can also act as barriers and facilitate atmospheric deposition of Pb (Mielke et al.

1983), leaving a pattern of higher Pb trapping in the street side of the front yard as compared to

the back or side yard. A combined effect of preferential trapping and LBP results in a U-shaped

pattern of higher SLL near the house and the road, and lower SLL in the middle of the yard

(Olszowy et al. 1995). Lack of these types of spatial patterns negates the effect of preferential

trapping on properties we assessed. Based on the findings of our current study, we argue that

even in medium urban communities, leaded exterior paint can be the dominant source of soil Pb

if the property is located at a certain distance from the industrial sites, freeways, or dense traffic

area. The location and age of the building structure are the two most important parameters that

influence which source will dominate the distribution of soil Pb and to what extent in any given

community, irrespective of its size.

Nationwide studies conducted by U.S. Housing and Urban Development report that 76% of

homes built prior to 1960 have Pb in exterior paint and 83% of pre-1980 housing stock contains

Pb in the paint (USHUD 1990, 1995). This study found Pb in exterior paint at varying depths for

all tested urban and suburban properties built prior to 1978, supporting the observation of Clark

and Knudson (2013) that all communities of similar age and site throughout the US show a

similar trend of contamination. SLL can exist around all sides of the structures and extend far

into the yard (Clark and Knudson 2013; USEPA 2000, Litt et al. 2002). In all properties, the SLL

was observed to be very high alongside the structures and decreased with increasing distance

from the homes. In addition, we also noted that in inner-city properties like site A, G, and K, the

neighboring structures, adjacent to the tested yards, also acted as prominent sources of Pb.

Buried or scattered paint chips in the soil of site A increased the SLL in some locations against

the influence of distance.

Pb contamination in residential soil is so widespread that it can be found in unapparent locations,

increasing the threat of continual risk of exposure. In accordance with previous studies, we have

observed three such unforeseen scenarios: i) high soil Pb in a suburban house with a well-

maintained exterior (Clark and Knudson 2013) was observed in site L, which exhibited SLL as

high as 3000 mg/kg; ii) high soil Pb in a property with an unpainted exterior was noted in site B,

18

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

18

where the SLL surpassed the federal guideline only beneath the windows, a trend also observed

by Linton et al. (1980); and iii) increase in SLL after renovation and repainting of a house

(Clark and Knudson 2013), which is duly noted in site A. Table S.3 presents a comparative SLL,

estimated BLL, and potential dust loading values in the three yards of site A that illustrates the

detrimental effects of renovation activity on soil Pb.

Dust is a major source of concern for children, as inhalation is a major route of Pb exposure. The

respiratory tract is responsible for the absorption of lead particles into the systemic circulation,

and respiration and metabolic rates are faster in children, resulting in increased absorption of

lead via dust (Menezes-Filho 2018). Two tools were used in this study to estimate the risk of Pb

exposure from outdoor dust: i) direct measurement of accumulated dust on the outdoor surfaces,

and ii) secondary estimation of potential lead on play surfaces (PLOPS). Overall, accumulated

Pb in dust showed significant (p<0.001) positive correlation with Pb in paint and negative

correlation with depth index, suggesting flaking LBP on surfaces such as window sills, doors,

steps, and old building materials can potentially contribute Pb to dust.

Lead loading on exterior surfaces has been found to be a continuous process, primarily due to

resuspension of soil Pb (Caravanos et al. 2006; Mielke et al. 2006). Researchers identified the

resuspension of soil Pb during dry seasons as the major driving force behind the seasonally

EBLL in children in New Orleans, Syracuse, and Indianapolis (Laidlaw et al. 2005). Mielke et

al. (2006) developed a new tool for measuring the potential Pb surface loading per area (mg/ft2)

of the soil. Potential Pb on play surfaces, a measure of the Pb concentration of a soil surface per

surface area as a source of Pb dust for children who would play on that surface, was calculated

based on a prediction model created by Mielke et al. (2006) (equation 8).

SLL median¿0.69

PLOPS=−43.74+24.85∗¿ -------Equation 8

The PLOPS values for all urban and suburban properties are presented in table 5 to provide

insight into the overall potential of Pb exposure from these surfaces, both directly through hand-

to-mouth activity, or indirectly through resuspension of Pb dust. Mielke et al. (2006) also showed

a large drop in PLOPS in 136 properties and vacant lots in New Orleans after the contaminated

surfaces were treated with a soil cover. Similar measures should be considered for the properties

19

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

19

assessed in our study, as the large PLOPS values indicate a high potential risk of Pb exposure to

children if they play in these areas.

The relationship between Pb in produce and the adjacent soil Pb is equivocal in the literature;

with some studies reporting maximum Pb levels in vegetables associated with the highest SLL

(Bielinska 2009; Huang et al. 2012; Moir and Thornton 1989) and other studies finding only

weak correlations between crop Pb levels and the adjacent soil Pb, due to unverified reasons

(Hough et al. 2004; Murray et al. 2011; McBride et al. 2014). Our study found strong and

significant (r = 0.92 based on fresh weight and r = 0.85 based on dry weight; p < 0.001)

correlations between the plant-accumulated Pb and SLL. Ten out of fourteen edibles tested

exceeded the health-based guidance values set by the European Union (EU) (EC 2006). There

are no such standards currently in the U. S., but the median Pb in edible plants found in this

study was 72 times higher than the median market basket concentrations of Pb reported by the U.

S. Food and Drug Administration’s (FDA) Total Diet Study. (USFDA 2010). The FDA

recommended the maximum ingestion lead level for a child is 3 µg/day; eating between four and

five grams of contaminated produce would reach that limit (Frank 2019). Interestingly, Pb levels

exceeding the EU guideline standards were found in plants grown in both indigenous soils (sites

H and M) of these properties as well as potting soils (site C and side yard of site A). The median

Pb in plants grown in indigenous soils (1.05 mg/kg) was higher than those grown in the potting

soil (0.21 mg/kg) and showed a stronger correlation with the Pb concentrations in adjacent soils

(r = 0.91 and 0.69 for indigenous and potting soils respectively). However, it is deeply

concerning that the edible plants grown in potting soils containing Pb lower than 60 mg/kg were

found to contain Pb levels exceeding the health guidelines. The federal standard of 400 mg/kg

soil Pb provides a misconception of safety for soils containing lower Pb concentrations; there are

no current levels set by either U.S. Environmental Protection Agency (EPA) or HUD for urban

gardening. Moreover, each species of produce has different biological properties that determine

their uptake, absorption, and bioavailability of Pb, demonstrated by the varying lead

concentrations in them (USFDA 2010). Differential uptake of Pb by different vegetables at

similar SLL was noted in the current study (site A) in Rochester as well as by McBride et al.

(2014) in New York City and Buffalo, NY; and Misenheimer et al. (2018) in Puerto Rico. These

findings strongly support the development of new plant-specific federal guidelines for soil Pb to

promote safer gardening practices.

20

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

20

4.2 Safer use of yards by primary prevention tools

Based on results of the field assessment, we developed three primary prevention tools to help

homeowners understand the Pb exposure risks throughout their properties as the first step of

primary prevention: i) a prediction model to calculate soil Pb using the measured distance from

the structure; ii) a prediction model to estimate Pb in edible produce grown in yard gardens from

the adjacent soil Pb that can be calculated from the distance from the built structure, iii) GIS

maps showing Pb hot spots with high risk, low risk, and safe areas in the yards. Prediction

models and GIS maps presented by prior research, covering a larger city area, are crucial for

bringing necessary changes in the policy (Wu et al. 2010), but this is an ongoing, long-term

process. Considering the continual Pb exposure and elevated BLL in families with lower

socioeconomic standing, it is of utmost importance to develop an easier process for the

homeowners to understand the risk to be able to avoid it. To the best of our knowledge, the

present study is the first attempt to develop site-specific prediction models and precise GIS maps

illustrating the potential risk through mapping the SLL in the exact latitude and longitude of the

sampling locations on individual residential properties. The GIS maps also mark the safer

locations to create gardens based on the two prediction models we developed. These measures

would provide the homeowners with an opportunity to predict soil Pb levels to plan safer uses of

their yards.

City-wide larger models to calculate soil Pb based on distance from urban centers of big cities or

roads with high traffic density considered the atmospheric deposition of Pb as the dominant

source (Yaylah-Abanuz 2019; Brown et al. 2008; Zhang et al. 2015; Wu et al. 2010). However,

residential soil Pb with exterior LBP as the major source varies widely from one property to

another, depending on the age of the structure(s), presence and types of siding, past land use, and

renovation processes. Site-specific prediction models would provide homeowners with an

opportunity to predict soil Pb levels using the distance from a structure to utilize their yards more

safely. For instance, a play area should be planned in soil that contains less than 400 mg/kg Pb,

according to federal guidelines. Based on our site-specific models, SLL drops from 400 mg/kg to

< 100 mg/kg if the play area is located 2 m further from the house. Considering the steep

increase of BLL between 100 to 300 mg/kg SLL, simply moving a play area a few meters further

21

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594

595

596

21

from the structure would increase the chance of primary prevention at no cost. Although we

highly recommend at least a one-time assessment of soil Pb for each property, it is likely that one

prediction model would be valid for one block or neighborhood if the homes were built at a

similar time; however, further investigation of this assumption is necessary.

This study also presents prediction models exhibiting a strong correlation (r = 0.85 and 0.92)

between Pb accumulation (per d.w. and f.w.) in produce/plant parts and the Pb levels in the

adjacent soil. One limitation of these prediction models is that they will not be able to provide

any insight into the plant-specific differential Pb uptake, as the pronounced effect of soil Pb may

have masked the effects of plant-specific uptake in the current study. However, utilization of

these equations would be able to provide a rough estimate of the potential Pb accumulation into

vegetables from the adjacent SLL to estimate a safer location for a garden. For instance, the

vegetable gardens in sites M and H were located in the areas with the highest SLL, resulting in

median Pb levels in edible parts of 1.58 and 0.56 mg/kg, respectively, which exceed the EU

guideline by 15.8 and 5.6 folds, respectively (EC 2006). Both of these properties have areas with

much lower soil Pb further into the yard. Simply moving these gardens to such locations would

be a useful measure of intervention to reduce the Pb exposure risk through the consumption of

Pb-contaminated vegetables from the property gardens. We estimated an SLL value associated

with the EU guideline of 0.1 mg/kg Pb per plant f.w. using a reverse fit of equation 2 (R2 = 0.97,

p<0.0001) which associates with a SLL of 64.7 mg/kg; a slightly lower value of 0.09 mg/kg Pb

per plant f.w. associates with an SLL of 58.2 mg/kg. Using the reverse fit of the site-specific

models, we calculated the associated distance with 58.2 mg/kg soil Pb in site M and site H, and

these distances are found to be 6.44 and 3.85 m respectively. Thus, our combined models

indicate that to attain a minimum requirement for the primary prevention, a home garden should

be set up at least 6.44 m away in site M and 3.85 m away in site H from their respective

structures. Similarly, a safer distance can be calculated for any property garden to avoid

consumption of Pb through homegrown vegetables.

Figure 8 presents two reference GIS maps of site A showing the Pb exposure risk associated with

soil Pb distribution in three areas of the property before and after the renovation of the home. To

simply and clearly communicate risk levels to homeowners, we developed a visual tool utilizing

22

597

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

619

620

621

622

623

624

625

626

627

22

a graded color scheme of green to black representing risk levels. Table S.4 presents a guideline

created for the homeowners to explain each color category, associated risk, and the

recommended use of that location on their properties. In addition to using the federal standards

for assigning colors to the range of Pb concentrations, we have also used our prediction models

to mark areas of the property where it would be safer to create a garden based on the EU health

guidelines for edible produce. The areas marked in red to black colors represent hotspots with

very high to extreme risk of potential Pb exposure and thus must be avoided and should be

considered for remedial actions.

4.3 Sustainable Remediation

The second and final step of primary prevention is to correct the hazards through remedial

actions, which led us to investigate an affordable process that would provide substantial risk

reduction without economically burdening the homeowners. WTR collected from Rochester’s

largest drinking water treatment plant not only exhibited a very high Pb-binding ability but was

also found to be safe when used as a soil amendment according to the federal TCLP guidelines.

Past research documented varying effectiveness of different sorbents to immobilize lead,

including activated carbon (Moyo et al. 2013), natural and engineered clay-like zeolite

(Wingenfelder et al. 2005), mordenite (Turkyilmaz et al. 2014), montmorillonite (Zhang and Hou

2008), bentonite (Inglezakis et al. 2007), palygorskite, and sepiolite (Shirvani et al. 2006).

Although these technologies provide in-situ alternatives to the current expensive ex-situ

practices, the cost of these natural or engineered sorbents in bulk could become cost-prohibitive

to low-income urban families. WTR is a no-cost sorbent which would provide these homeowners

with an affordable remediation alternative to be utilized as soil cover or amendment in the most

concerning areas of their properties. The only required cost is associated with its implementation,

which could be further reduced by conducting a proper assessment to mark the risk hotspots (as

shown in the GIS maps for site A), where the remediation process is absolutely needed. Very few

studies have investigated the effectiveness of Pb adsorption by Fe and Al-based WTR collected

from different parts of the world. Pb sorption by Al-WTR showed a similar pattern of L type

sorption isotherm (Castaldi et al. 2015; Zhou et al. 2011). Adsorption maximum was reported to

be 17.55, 15.66, and 15.13 mg/g by WTRs collected from Miyamachi and Nishino in Japan and

Italy, respectively, as compared to 24.4 mg/g found in this study (Putra and Tanaka 2011).

23

628

629

630

631

632

633

634

635

636

637

638

639

640

641

642

643

644

645

646

647

648

649

650

651

652

653

654

655

656

657

658

23

Furthermore, Soleimanifer et al. (2019) reported the unique approach of reducing metal

contamination in urban stormwater with WTR coated mulch; a similar approach could be further

investigated for its potential to reduce plant-Pb uptake if used in home gardens. The sorption

desorption data along with the calculated Freundlich and Langmuir isotherm parameters suggests

very strong, favorable, and more irreversible adsorption of Pb on WTR through a physical

process, promising an effective option for reusing this waste byproduct that the city generates

every day to immobilize Pb in the high-risk soils of many contaminated residential properties.

Using WTR as a soil amendment will prevent the runoff and leaching of soil Pb and also

decrease the Pb available to be accumulated in garden vegetables. However, the implementation

of only WTR will not prevent the resuspension of Pb into dust. Our goal is to implement a dual-

process remediation system using WTR as a soil amendment with the addition of a plant cover to

decrease the dust loading of Pb. Several studies have shown the phytoremediation potential of Pb

(Jagetiya and Kumar 2020), but it is of utmost importance to use a non-edible plant in the Pb hot

spot areas to prevent its further entry into the food chain. Vetiver, a perennial non-edible and

non-invasive grass has been shown to be highly effective for Pb removal (Andra et al. 2009);

however, the harsh winters of western NY could be a serious impediment for its implementation.

Our group has tested 17 native inedible plant species for their potential tolerance of high Pb

concentration, but all plants showed severe phytotoxic effects of Pb in hydroponic media, where

100% Pb was available for plant uptake (data not shown). This further underscores the utility of

WTR for Pb removal, which according to the results of this study would retain more than 90% of

total soil Pb, making only a small fraction available for the plants to tolerate. The ongoing

studies in our greenhouse are characterizing this dual process technique using WTR as an

amendment to the soils from the backyard of site A, with an SLL median of 2900 mg/kg, and

switchgrass (Panicum virgatum), high biomass, perennial, fast-growing, inedible grass, which is

also native to western NY and which was previously studied for its ability to remove other

environmental contaminants (Phouthavong-Murphy 2020). This combined remediation process

would provide the homeowners with an effective and affordable primary prevention alternative

to intervene with the Pb exposure from the risk hotspots of their property.

Acknowledgements

24

659

660

661

662

663

664

665

666

667

668

669

670

671

672

673

674

675

676

677

678

679

680

681

682

683

684

685

686

687

688

689

24

The authors acknowledge Mr. Larry Peckham and Mrs. Nancy Peckham for funding the High

School Summer Research Program in 2018 and Nazareth College Summer Opportunities for

Activities in Research and Sponsorship (SOARS) in 2018 for partially funding this project.

We acknowledge Mr. John C. Kelly Jr. and other officials from Monroe County Water Authority

(MCWA) for their help in WTR collection; our community partners and homeowners for their

communication and participation; Dr. Callie Babbitt from Rochester Institute of Technology for

conducting the ICP-OES analyses of TCLP samples; Mrs. Sharon Luxmore for assistance with

ordering materials; and Ms. Aelis Spiller, Ms. Ariel Struzyk, Mr. LaMar Jackson, Mr. Sean Dillon, Ms.

Maria Vargas Morales, and Ms. Bailey Groth for assistance in sample preparation and field

measurements; and Mr. Shyamal K. Mitra and Dr. Ananta Paine for providing feedback about this

manuscript.

Author contributions

P.D. designed the research, field assessment, and experiments. P.D., D.R.B., M.J.P, B.E.L., A.C.T., J.M.P.,

H.M.C., J.L.W., I.G., and D.G. conducted the field assessment; B.E.L. and A.C.T analyzed Pb in surface

dust, J. S., C.V.B., M.L.M., D. R. B., M.J.P., C.E.D. analyzed Pb in plants with the supervision of S.M.Z.

and P.D.; J.L.W., I.G., J.P.M., J.M.P., H.M.C., A.C.T. conducted phase II experiments with the supervision

of S.M.Z. and P.D; P.D. conducted the secondary analyses, isotherm modeling, statistical data analyses,

and developed the prediction models. M.C. created the GIS maps and D.R.B. created the Tableau maps.

P.D. wrote the manuscript with S.M.Z. and J.P.M. with contributions and feedback from all other authors.

Abbreviations:

Pb: lead; WTR: drinking water treatment residual; XRF: x-ray Fluorescence Analyzer; Al-WTR:

aluminum-based water treatment residual; SLL: soil lead level; mg/kg: milligram per kilogram;

WTR: water treatment residual; GIS: geographical information systems; LBP: lead based paint;

EBLL: elevated blood lead level; BLL: blood lead level; µg /dL : micrograms per deciliter;

USEPA: United States Environmental Protection Agency; USHUD: United States Department of

Housing and Urban Development; n: number of samples; HUD: Department of Housing and

Urban Development; TCLP: toxicity characteristic leaching procedures; SSR: solid to solution

25

690

691

692

693

694

695

696

697

698

699

700

701

702

703

704

705

706

707

708

709

710

711

712

713

714

715

716

717

718

719

720

25

ratio; KCl: potassium chloride; mg/L: milligrams per liter; HNO3: nitric acid; AAS: atomic

absorption spectrophotometer; RCRA metals: 8 metals (Arsenic, Barium, Cadmium, Chromium,

Lead, Mercury, Selenium, Silver) listed under Resource Conservation and Recovery Act; ICP-

OES: individually coupled plasma optical emission spectroscopy; Tukey-Kramer HSD: Tukey

Kramer honest significant difference.

References

1. Andra SS, Datta R, Sarkar D, Makris KC, Mullens CP, Sahi SV, et al. 2009. Induction of lead-binding phytochelatins in vetiver grass (Vetiveria zizanioides L.). J Environ Qual 38:868–877, PMID:19329675, https://doi.org10.2134/jeq2008.0316.

2. Bakulski KM, Rozek LS, Dolinoy DC, Paulson HL, Hu H. 2012. Alzheimer’s disease and environmental exposure to lead: the epidemiologic evidence and potential role of epigenetics. Current Alzheimer Research 9(5):563-573, PMID:22272628.

3. Bell T, Campbell S, Liverman DGE, Allison D, Sylvester P. 2010. Environmental and potential human health legacies of non-industrial sources of lead in a Canadian urban landscape – the case study of St John's, Newfoundland. International Geology Review 52(7-8):771-800, https://doi.org/10.1080/00206811003679786.

4. Bickel JM. 2010. Spatial and temporal relationships between blood lead and soil lead concentrations in Detroit, Michigan. Master’s Thesis, Wayne State University: Detroit, MI.

5. Bielinska EJ. 2009. Influence of the root layer on the content of cadmium and lead in soils andvegetable plants in regions with diverse anthropogenic impact. J Res Appl Agric Eng 54:16–20.

6. Bouchard MF, Bellinger DC, Weuve J, Matthews-Bellinger J, Gilman SE, Wright RO, et al. 2009. Blood lead levels and major depressive disorder, panic disorder, and generalized anxietydisorder in US young adults. Archives of General Psychiatry 66(12):1313-1319, https://doi.org/10.1001/archgenpsychiatry.2009.164.

7. Brown RW, Gonzales C, Hooper MJ, Bayat AC, Fornerette AM, Mcbride TJ, et al. 2008. Soil lead (Pb) in residential transects through Lubbock, Texas: a preliminary assessment. Environ Geochem Health 30:541–547, PMID:18546057, https://doi.org/10.1007/s10653-008-9180-y.

26

721

722

723

724

725

726

727

728

729730731732

733734735

736

737738739740741

742743744745746747748749750751752753754755756757

26

8. Canfield R, Henderson CR, Cory-Slechta DA, Cox C, Jusko TA, Lanphear BP. 2003. Intellectual impairment in children with blood lead concentrations below 10 mg/dL: the Rochester cohort study. N. Engl. J. Med 348:1517–1526, https://doi.org/10.1056/NEJMoa022848.

9. Caravanos J, Weiss AL, Blaise MJ, Jaeger RJ. 2006. A survey of spatially distributed exterior dust lead loadings in New York City. Environmental Research 100(2):165-172, https://doi.org/10.1016/j.envres.2005.05.001.

10. Castaldi P, Silvetti M, Garau G, Demurtas D, Deiana S. 2015. Copper(II) and lead(II) removal from aqueous solution by water treatment residues. Journal of Hazardous Materials 283:140-147, PMID: 25262486, https://doi.org/10.1016/j.jhazmat.2014.09.019.

11. CDC (Centers for Disease Control and Prevention). 2002. Managing Elevated Blood Lead Levels Among Young Children: Recommendations from the Advisory Committee on Childhood Lead Poisoning Prevention. https://www.cdc.gov/nceh/lead/casemanagement/casemanage_main.htm [Accessed 12 July 2020].

12. CDC (Centers for Disease Control and Prevention). 2005. Building Blocks for Primary Prevention: Protecting Children from Lead-Based Paint Hazards. https://www.cdc.gov/nceh/lead/publications/building_blocks_for_primary_prevention.pdf [Accessed 28 July 2020].

13. CDC (Centers for Disease Control and Prevention). 2018. Safe Blood Lead Testing Laws Requiring 5 ug/dL & CDC Reference Rule. https://www.cdc.gov/phlp/docs/laws-bll.pdf. [Accessed Jul. 28, 2020].

14. Clark JJ, Knudsen AC. 2013. Extent, characterization, and sources of soil lead contamination in small-urban residential neighborhoods. Journal of Environmental Quality 42:1498-1506, https://doi.org/10.2134/jeq2013.03.0100.

15. Crinnion WJ, Pizzorno JE. Clinical Environmental Medicine. Elsevier Health Sciences:Kindle Edition. 8535-8537.

16. Datko-Williams L, Wilkie A, Richmond-Bryant J. 2014. Analysis of U.S. soil lead (Pb) studies from 1970 to 2012. Science of the Total Environment 468-469:854-863, https://doi.org/10.1016/j.scitotenv.2013.08.089.

27

758759760761762763

764765766

767768769770771772773774775776777778779780781782783784785786787788789790791792793

794795796797798799

27

17. Dixon DL, Gaitens JM, Jacobs DE, Strauss W, Nagaraja J, Pivetz T, et al. 2009. Exposure of U.S. children to residential dust lead, 1999–2004: II. The contribution of lead-contaminated dust to children’s blood lead levels. Environmental Health Perspectives 117(3):468-474, https://doi.org/10.1289/ehp.11918.

18. EC (Commission Regulation EC). 2006. No 1881/2006 of 19 December 2006 Setting Maximum Levels for Certain Contaminants in Foodstuffs. Off J Eur Union L 364:5–24. https://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2006:364:0005:0024:EN:PDF.[Accessed 12 July 2020].

19. Filippelli GM, Laidlaw MAS, Latimer JC, Raftis R. 2005. Urban lead poisoning and medical geology: an unfinished story. Geological Society of America Today 15(1):4-11, https://doi.org/10.1130/1052-5173(2005)015<4:ULPAMG>2.0.CO;2.

20. Frank JJ, Poulakos AG, Tornero-Velez R, Xue J. 2019. Systematic review and meta-analyses of lead (Pb) concentrations in environmental media (soil, dust, water, food, and air) reported in the United States from 1996 to 2016. Science of The Total Environment 694:133489, https://doi.org/10.1016/j.scitotenv.2019.07.295.

21. Geltman PL, Brown MJ, Cochran J. 2001. Lead poisoning among refugee children resettled in Massachusetts, 1995 to 1999. Pediatrics 108(1):158–162, PMID: 11433069, https://doi.org/10.1542/peds.108.1.158.

22. Gulson BL, Mizon KJ, Korsch MJ, Palmer JM, Donnelly JB. 2003. Mobilization of lead from human bone tissue during pregnancy and lactation—a summary of long-term research. Science of the Total Environment 303(1–2):79–104, PMID: 12568766, https://doi.org/10.1016/s0048-9697(02)00355-8.

23. Hough RL, Breward N, Young SD, Crout NMJ, Tye AM, Moir AM, et al. 2004. Assessing potential risk of heavy metal exposure from consumption of home-produced vegetables by urban populations. Environ Health Perspect 112:215–221, PMID: 14754576, https://doi.org/10.1289/ehp.5589.

24. Hu H, Shih R, Rothenberg S, Schwartz BS. 2007. The epidemiology of lead toxicity in adults: measuring dose and consideration of other methodologic issues 115(3):455-462, PMID:17431499, https://doi.org/10.1289/ehp.9783.

25. Huang Z, Chen T, Yu J, Qin D, Chen L. 2012. Lead contamination and its potential sources invegetables and soils of Fujian, China. Environ Geochem Health 34:55–65, https://doi.org/10.1007/s10653-011-9390-6.

28

800801802803

804805806807808809810811812813814815816817818819820821822823

824825826827

828829830831832833834835836837

838839840

28

26. Inglezakis VJ, Stylianou MA, Gkantzou D, Loizidou MD. 2007. Removal of Pb(II) from aqueous solutions by using clinoptilolite and bentonite as adsorbents. Desalination 210(1-3):248-256, https://doi.org/10.1016/j.desal.2006.05.049

27. Jagetiya B, Kumar S. 2020. Phytoremediation of lead: a review. In: Lead in Plants and the Environment. Radionuclides and Heavy Metals in the Environment. (Gupta D, Chatterjee S, Walther C, eds). Springer. 179-202.

28. Johnson DL, Bretsch JK. 2002. Soil lead and children’s blood lead levels in Syracuse, FNY, USA. Environmental Geochemistry and Health 24:375-385, https://doi.org/10.1023/A:1020500504167.

29. Kastury F, Smith E, Lombi E, Donnelley MW, Cmielewski PL, Parsons DW, et al. 2019. Dynamics of lead bioavailability and speciation in indoor dust and x-ray spectroscopic investigation of the link between ingestion and inhalation pathways. Environmental Science and Technology 53:11486-11495, https://doi.org/10.1021/acs.est.9b03249.

30. Korfmacher KS, Kuholski K. 2007. Do the same houses poison many children? An investigation of lead poisoning in Rochester, New York, 1993–2004. Public Health Reports 122:482-487, https://doi.org/10.1177/003335490712200409.

31. LaBelle SJ, Lindahl PC, Hinchman RR, Ruskamp J, McHugh K. 1987. Pilot Study of the Relationship of Regional Road Traffic to Surface-soil Lead Levels in Illinois. Argonne, IL: Argonne National Laboratory. https://ntrl.ntis.gov/NTRL/dashboard/searchResults.xhtml?searchQuery=DE88013158. [Accessed 12 July 2020].

32. Laidlaw M. 2001. Distribution and sources of metals in soils of Marion County, Indiana. Master’s Thesis, Indiana University: Indianapolis, IN.

33. Laidlaw MA, Mielke HW, Filippelli GM, Johnson DL, Gonzales CR. 2005. Seasonality and children's blood lead levels: developing a predictive model using climatic variables and blood lead data from Indianapolis, Indiana, Syracuse, New York, and New Orleans, Louisiana (USA). Environmental Health Perspectives 113(6):783-800, PMID: 15929906, https://doi.org/10.1289/ehp.7759.

34. Lal R. 2020. Home gardening and urban agriculture for advancing food and nutritional security in response to the COVID-19 pandemic. Food Sec 2020. https://doi.org/10.1007/s12571-020-01058-3.

29

841

842843844845846847848849850851852853854855856857858859

860

861862863

864865866867868

869870

871872873874875876877878879880881

29

35. Linton RW, Farmer ME, Hopke PK, Natusch DFS. 1980. Determination of the sources of toxic elements in environmental particles using microscopic and statistical analysis techniques. Environmental International 4(5-6):453-461, https://doi.org/10.1016/0160-4120(80)90026-4.

36. Litt JS, Hynes P, Carroll P, Maxfield R, McLaine P, Kawecki C. 2002. Safe yards: improving urban health through lead-safe yards. J. Urban Technol 9:71–93, https://doi.org/10.1080/1063073022000016487.

37. Magavern S. 2018. Policies to reduce lead exposure: lessons from Buffalo and Rochester. International Journal of Environmental Research and Public Health 15(10):2197, PMID: 30304767, https://doi.org/10.3390/ijerph15102197.

38. Maskall JE, Thornton I. 1993. Metal contamination of soils at historical lead smelting sites. Land Contam Reclam 1:92–100.

39. McBride MB, Shayler HA, Spliethoff HM, Mitchell RG, Marquez-Bravo LG, Ferenz GS, et al. 2014. Concentrations of lead, cadmium and barium in urban garden grown vegetables: the impact of soil variables. Environ Pollut 194:254-261, PMID: 25163429, https://doi.org/10.1016/j.envpol.2014.07.036.

40. McKay G, Otterburn MS, Sweeney AG. 1980. The removal of colour from effluent using various adsorbents III. Silica: rate processes. Water Research 14(1):15–20, https://doi.org/10.1016/0043-1354(80)90037-8.

41. Menezes-Filho J, Carvalho CF, Rodrigues JLG, Araújo CFS, Dos Santos NR, Lima CS, et al.2018. Environmental co-exposure to lead and manganese and intellectual deficit in school-aged children. International Journal of Environmental Research and Public Health 15(11):2418, https://doi.org/10.3390/ijerph15112418.

42. Menke A, Muntner P, Batuman V, Silbergeld EK, Guallar E. 2006. Blood lead below 0.48 μmol/L (10 μg/dL) and mortality among US adults. Circulation 114(13):1388-1394, https://doi.org/10.1161/CIRCULATIONAHA.106.628321.

43. Meroufel B, Benali O, Benyahia M, Benmoussa Y, Zenasni MA. 2013. Adsorptive removal of anionic dye from aqueous solutions by Algerian kaolin: characteristics, isotherm, kinetic and thermodynamic studies B. J. Mater. Environ. Sci 4(3):482-491, ISSN: 2028-2508, CODEN: JMESCN.

44. Mielke HW. 1999. Lead in the inner cities: policies to reduce children's exposure to lead maybe overlooking a major source of lead in the environment. American Scientist 87(1): 62-73.

45. Mielke HW, Anderson JC, Berry KJ, Mielke PW, Chaney RL, Leech M. 1983. Lead concentrations in inner-city soils as a factor in the child lead problem. American Journal of Public Health 73(12):1366-1369, PMID: 6638229, https://doi.org/10.2105/ajph.73.12.1366.

30

882883884885

886887888889890891892893894

895896

897898899900901902903904905906907908909910911912913914915916917918919920921922923924925

30

46. Mielke HW, Covington TP, Mielke PW, Wolman FJ, Powell ET, Gonzales CR. 2011. Soil intervention as a strategy for lead exposure prevention: the New Orleans lead-safe childcare playground project. Environ Pollut 159:2071–2077, https://doi.org/10.1016/j.envpol.2010.11.008.

47. Mielke HW, Gonzales C, Mielke Jr. PW 2008. Urban soil-lead (Pb) footprint: retrospective comparison of public and private properties in New Orleans. Environmental Geochemistry and Health 30(3):231-242, PMID:17687626, https://doi.org/10.1007/s10653-007-9111-3.

48. Mielke HW, Gonzales CR, Powell E, Jartun M, Mielke Jr. PW. 2007. Nonlinear association between soil lead and blood lead of children in metropolitan New Orleans Louisiana: 2000-2005. Science of the Total Environment 388:43-53, PMID: 17884147, https://doi.org/10.1016/j.scitotenv.2007.08.012.

49. Mielke HW, Powell ET, Gonzales CR, Mielke Jr PW. 2006. Potential lead on play surfaces: evaluation of the “PLOPS” sampler as a new tool for primary lead prevention. Environmental Research 103(2):154-159, https://doi.org/10.1016/j.envres.2006.08.007.

50. Misenheimer J, Nelson C, Huertas E, Medina-Vera M, Prevatte A, Bradham K. 2018. Total and bioaccessible soil arsenic and lead levels and plant uptake in three urban community gardens in Puerto Rico. Geosciences 8(2):43-53, https://doi.org/10.3390/geosciences8020043.

51. Moir AM, Thornton I. 1989. Lead and cadmium in urban allotment and garden soils and vegetables in the United Kingdom. Environ Geochem Health 11:113-119, https://doi.org/10.1007/BF01758660.

52. Moyo M, Chikazaza L, Chomunorwa Nyamunda B, Guyo U. 2013. Adsorption batch studies on the removal of Pb(II) using maize tassel based activated carbon. Journal of Chemistry 2013, Article ID: 508934.

53. Murray H, Pinchin TA, Macfie SM. 2011. Compost application affects metal uptake in plantsgrown in urban garden soils and potential human health risk. J Soils Sediments 11:815–829, https://doi.org/10.1007/s11368-011-0359-y.

54. Nash D, Magder LS, Sherwin R, Rubin RJ, Silbergeld EK. 2004. Bone density-related predictors of blood lead level among peri- and postmenopausal women in the United States: the third national health and nutrition examination survey, 1988–1994. American Journal of Epidemiology 160(9):901-911, https://doi.org/10.1093/aje/kwh296.

31

926927928929930931932933934935936937938939940941

942943944

945

946947948949950

951952953

954955956957

958959960961962963964965966967

31

55. National Gardening Association. 2011. Impact of home and community gardening in America. [Accessed 12 July 2020]. https://garden.org/learn/articles/view/3126/.

56. Nevin R. 2007. Understanding international crime trends: the legacy of preschool lead exposure. Environmental Research 104(3):315–336, PMID: 17451672, https://doi.org/10.1016/j.envres.2007.02.008.

57. Olszowy H, Torr P, Imray P. 1995. Trace element concentrations in soil from rural and urban areas of Australia. In: Contaminated Sites Monograph Series 4. South Australian Health Commission. Glenelg Press: Glenelg, Australia.

58. Özer A, Pirinççi HB. 2006. The adsorption of Cd(II) ions on sulphuric acid-treated wheat bran. Journal of Hazardous Materials 137(2):849–855, https://doi.org/10.1016/j.jhazmat.2006.03.009.

59. Phouthavong-Murphy JC, Merrill AK, Zamule S, Giacherio D, Brown B, Roote C, et al.2020. Phytoremediation potential of switchgrass (Panicum virgatum), two United Statesnative varieties, to remove bisphenol-A (BPA) from aqueous media. Sci Rep 10:835,https://doi.org/10.1038/s41598-019-56655-w.

60. Putra RS, Tanaka S. 2011. Aluminum drinking water treatment residuals (Al-WTRs) as an entrapping zone for lead in soil by electrokinetic remediation. Separation and Purification Technology 79(2011):208-215, https://doi.org/10.1016/j.seppur.2011.02.015.

61. Reagan P, Silbergeld E. 1989. Establishing a health based standard for lead in residential soils. In: Trace Substances in Environmental Health (Hemphill and Cothern, eds). Suppl Environmental Geochemistry and Health 23:199-238.

62. Sall J, Creighton L, Lehman A. 2005. JMP Start Statistics, third ed. SAS Institute: Cary, NC and Pacific Grove, CA.

63. Schwarz K, Pickett ST, Lathrop RG, Weathers KC, Pouyat RV, Cadenasso ML. 2012. The effects of the urban built environment on the spatial distribution of lead in residential soils. Environ Pollut 163:32–39, PMID: 22325428, https://doi.org/10.1016/j.envpol.2011.12.003.

64. Soleimanifar H, Deng Y, Barrett K, Feng H, Li X, Sarkar D. 2019. Water treatment residual-coated wood mulch for addressing urban stormwater pollution. Water Environ Res 91(6):523-535, https://doi.org/10.1002/wer.1055.

65. Turkyilmaz H, Kartal T, Yigitarslan Yildiz S. 2014. Optimization of lead adsorption of mordenite by response surface methodology: characterization and modification. J Environ Health Sci Engineer 12(5), https://doi.org/10.1186/2052-336X-12-5.

32

968969970971972973974975976977978979980981982983984985986987988989990991992993994995996997998999

100010011002100310041005100610071008100910101011

32

66. USEPA (United States Environmental Protection Agency). 2000. America’s Children and the Environment: A First View of Available Measures. https://www.epa.gov/americaschildrenenvironment [Accessed 13 July 2020].

67. USEPA (United States Environmental Protection Agency). 2001. Lead;Identification of Dangerous Levels of Lead;Final Rule. Federal Register 66(4):1206-1240. https://www.govinfo.gov/content/pkg/FR-2001-01-05/pdf/01-84.pdf. [Accessed 13 July 2020].

68. USEPA (United States Environmental Protection Agency). 2011. National Emissions Inventory Data and Documentation. https://www.epa.gov/air-emissions-inventories/2011-national-emissions-inventory-nei-data. [Accessed 13 July 2020].

69. USFDA (United States Food and Drug Administration). 2010. Total Diet Study Statistics on Element Results Market Baskets 2006–1 through 2013–4. https://www.fda.gov/media/77948/download. [Accessed 13 July 2020].

70. USHUD (United States Department of Housing and Urban Development). 1990. Comprehensive and Workable Plan for the Abatement of Lead-based Paint in Privately OwnedHousing. Report to Congress. http://www.huduser.org/Publications/pdf/HUD-5716.pdf. [Accessed 13 July 2020].

71. USHUD (United States Department of Housing and Urban Development). 1995. Report on the National Survey of Lead-based Paint in Housing. http://www2.epa.gov/lead/report-national-survey-lead-based-paint-housing-base-report-june-1995. [Accessed 13 July 2020].

72. Wagner T, Langley-Turnbaugh S. 2008. Case study: examining the contribution of historical sources of lead in urban soils in Portland, Maine, USA. Journal of Environmental Planning and Management 51(4):525-541, https://doi.org/10.1080/09640560802117002.

73. Weisskopf MG, Weuve J, Nie H, Saint-Hilaire M, Sudarsky L, Simon DK, et al. 2010. Association of cumulative lead exposure with Parkinson’s Disease. Environmental Health Perspectives 118(11):1609-1613, https://doi.org/10.1289/ehp.1002339.

74. Wu J, Edwards R, He X, Liu Z, Kleinman M. 2010. Spatial analysis of bioavailable soil lead concentrations in Los Angeles, California. Environmental Research 110:309-317. PMID: 20219189, https://doi.org/10.1016/j.envres.2010.02.004.

75. Yaylah-Abanuz G. 2019. Application of multivariate statistics in the source identification of heavy-metal pollution in roadside soils of Bursa, Turkey. Arabian Journal of Geosciences 12(12):382, https://doi.org/10.1007/s12517-019-4545-3.

33

101210131014101510161017101810191020102110221023102410251026102710281029103010311032103310341035103610371038103910401041104210431044104510461047

1048

104910501051105210531054105510561057

33

76. Zahran S, Mielke HW, Gonzales CR, Powell ET, Weiler S. 2010. New Orleans before and after Hurricanes Katrina/Rita: a quasi-experiment of the association between soil lead and children’s blood lead. Environ. Sci. Technol 44:4433–4440. https://doi.org/10.1021/es100572s.

77. Zhang SQ, Hou WG. 2008. Adsorption behavior of Pb(II) on montmorillonite. Colloids and Surfaces A: Physicochemical and Engineering Aspects 320(1-3):92-97, https://doi.org/10.1016/j.colsurfa.2008.01.038.

78. Zhang H, Wang Z, Zhang Y, Ding M, Li L. 2015. Identification of traffic-related metals and the effects of different environments on their enrichment in roadside soils along the Qinghai–Tibet highway. Science of The Total Environment 521-522:160–172, PMID: 25835375, https://doi.org/10.1016/j.scitotenv.2015.03.054.

79. Zhou Y, Haynes RJ. 2011. Removal of Pb(II), Cr(III) and Cr(VI) from aqueous solutions using alum-derived water treatment sludge. Water, Air and Soil Pollution 215(1-4):631-643, https://doi.org/10.1007/s11270-010-0505-y.

34

1058105910601061

1062

10631064

1065106610671068106910701071107210731074

34

Table 1: Polynomial cubic fit of lead concentrations (mg/kg) in residential soil by distance from potential source of Pb based paint. The nonlinear correlation fit, regression equation, and parameter estimates were conducted using JMP (version 15). Parameters of the regression equations were estimated and the significance levels of the parameters were expressed as * (p < 0.05), ** (p < 0.01), *** (p < 0.001), **** (p < 0.0001).

Site R2 Fratio

p value Regression EquationsSignificantParameters

A1 0.9994 632.9 0.0292 y=3684 – 575.9∗x+89.45∗(x−4.321)2 –2.277∗(x−4.321)3 Intercept**, x2**

A3 0.9366 14.78 0.0266 y=2141 –1307∗x+8370∗(x−1.045)2 – 5519∗(x−1.045)3 x2*

A4 0.9868 50.02 0.0197 y=2226 – 282.6∗x+126.5∗( x−3.569 )2 – 28.10∗(x−3.569)3 x2*

L1 0.9567 22.13 0.0151 y=388 .3 –249.5∗x+497.1∗(x−2.199)2 – 108.1∗( x−2.199)3 x2*, x3*

L2 0.8593 18.32 0.0004 y=420.3 – 170.1∗x+277.5∗(x−2.110)2 – 61.83∗(x−2.110 )3 x2***, x3*

M 0.8402 21.03 <0.0001 y=394.3 –68.31∗x+135.8∗(x−3.029)2 –28.03∗(x−3.029)3 x2***, x3**

H 0.9820 218.5 <0.0001 y=752.9 –220.8∗x+127.8∗(x−2.286)2 – 22.73∗(x−2.286)3 Intercept****, x****,x2****, x3*

C1 0.9607 32.63 0.0029 y=232.8 –183.5∗x+486.1∗(x−1.181)2 – 246.8∗(x−1.181)3 x2**

C2 0.9868 150.4 <0.0001 y=153.3 –71.01∗x+362.7∗(x−1.371)2 –268.6∗(x−1.371)3 x2****, x3**

B 0.9865 97.77 0.0003 y=1175 – 713.6∗x+105.9∗(x−1.124 )2 –150.5∗(x−1.124)3 Intercept***, x***,x3*

G1 0.8887 13.31 0.0081 y=361.8 –71.36∗x+97.78∗(x−1.439)2 – 41.25∗(x−1.439)3 Intercept*, x2*

G2 0.9848 86.41 0.0004 y=508.9 – 71.63∗x+13.23∗( x−3.458 )2 – 0.4656∗(x−3.458)3 Intercept***, x**,x2**

Table 2. Pb concentrations in produce or parts of plants (mg/kg) grown in the gardens of the tested residential properties. Data are expressed as mean (n=3) + one standard deviation. Mean comparison is conducted by one-way ANOVA (F ratio = 121.4; p <0.0001) followed by Tukey-Kramer honest significant difference test. Mean comparison letters are expressed as superscript and levels not connected by same letters are significantly different from one another.

35

10751076

1077

1078

10791080

10811082

1083

1084

1085

1086

35

Produce/PlantsParts of plants

analyzedSite

Pbconcentration in soil(mg/kg)

Pbconcentration

in plants by dryweight (mg/kg)

Pb concentrationin plants byfresh weight

(mg/kg)

Guidance Value(EC, 2006)

Tomato(Solanum lycopersicum)

Tomato (fruit) Garden bed in C 8.5 0 + 0G 0.02 + 0.009I 0.1

Arugula(Eruca vesicaria)

Leaves Garden bed 1 in A 11.2 3.52 + 0.22 DE 0.19 + 0.016H 0.3

Lettuce(Lactuca sativa) Leaves Garden bed 8 in A 28.3 2.98 + 0.10 DE 0.01 + 0.005I 0.3

Radish(Raphanus sativus)

Radish (root) Garden bed 8 in A 28.3 4.67 + 0.33 DE 0.08 + 0.014I 0.1

Garlic(Allium sativum)

Garlic (bulb) Garden bed 7 in A 30.4 1.25 + 0.02 F 0.34 + 0.004G 0.3

Tomato(Solanum lycopersicum)

Stem Garden bed 7 in A 30.4 1.97 + 0.04 F 0.37 + 0.006G 0.3

Potato(Solanum tuberosum)

Root Garden bed 2 in A 32.6 1.88 + 0.07 F 0.23 + 0.008H 0.1

Broccoli(Brassica oleracea var. italica)

Leaves Garden bed 5 in A 52.8 1.79 + 0.07 F 0.48 + 0.012EF 0.3

Ground Ivy(Glechoma hederacea)

Whole plant Front yard in A 7400 16.07 + 0.37 A 3.35 + 0.59A N/A

Elephant Foot Yam(Amorphophallus paeoniifolius)

Whole plant Front yard garden in H 358.7 3.42 + 0.13 E 0.34 + 0.013G 0.1

Elephant Foot Yam(Amorphophallus paeoniifolius)

Whole plant Front yard garden in H 511.4 4.42 + 0.01 DE 1.05 + 0.003D 0.1

Elephant Foot Yam(Amorphophallus paeoniifolius)

Whole plant Front yard garden in H 663.3 4.89 + 0.08D 0.55 + 0.008E 0.1

Elephant Foot Yam(Amorphophallus paeoniifolius)

Whole plant Front yard garden in H 998.9 4.54 + 0.14DE 0.40 + 0.013FG 0.1

Elephant Foot Yam(Amorphophallus paeoniifolius)

Whole plant Front yard garden in H 1042 10.89 + 0.56B 1.22 + 0.054C 0.1

Tomato(Solanum lycopersicum)

Tomato (fruit) Back yard garden in M 985.1 8.40 + 0.30C 1.59 + 0.056B 0.1

36

10871088

36

Table 3. Physicochemical characteristics and TCLP of WTR collected from lagoon 1 in 2017 and lagoon 2 in 2018. Data are expressed as mean (n = 3) + one standard deviation. Mean comparison is conducted between WTR collected from different space and time.

TCLP RCRA Metals (mg/L)

MoistureContent

(%)pH

ElectricalConductivity (� s/cm)

Organicmattercontent

(%)

Ba Ag As Cr Pb Cd Se Hg

WTR fromLagoon 1

collected in2017

48.32 +1.997

6.99 +0.064

371.8 +38.09

2.195 +0.166

1.157 +0.0102

0.001 +0.000

0.086 +0.023

0.033 +0.006

0 +0

0.003 +0.001

0.036 +0.006

0 + 0

WTR fromLagoon 2

collected in2018

47.99 +1.962

6.98 +0.035

450.3 +52.50

2.259 +0.157

1.047 +0.021

0.001 +0.000

0.098 +0.035

0.04 +0.001

0 +0

0.002 +0.001

0.036 +0.006

0 + 0

EPApermissible

Limit100 5 5 5 5 1 1 0.2

F ratio 0.042 0.155 4.398 0.235 3.381 NA 0.273 3.811 NA 0.500 0.018 NA

p value 0.848 0.714 0.104 0.653 0.139 NA 0.629 0.123 NA 0.519 0.899 NA

Table 4. Percentile, mean, standard deviation (SD) of soil lead level (SLL) (mg/kg) and potential blood lead level (BLL) (g/dL) in

children and potential lead on play surfaces (PLOPS) (g /ft2) using three prediction models developed by Mielke et al., (2007),

Mielke et al., (2006), and Johnson and Bretsch (2002) respectively.

37

1089

1090

109110921093

109410951096

1097

1098

1099

37

Percentile SLL (mg/kg)SLL Statistics

(mg/kg)Potential BLL (g/dL)

PLOPS(g /ft2)

Max 75% Median 25% Min Mean SD N Equation 6 Equation 7 Equation 8

UrbanA after

Renovation173500 7346 1665 455 57 9619 25045 79 9.1 12 4107

A beforeRenovation

50000 6811 963 257 82 6857 11768 42 7.4 10.8 2800

C 1553 992 578 144 11 619 469 58 6.2 9.7 1956

M 3743 990 442 253 58 707 759 36 5.7 9.2 1617

H 2378 530 357 181 112 502 490 45 5.3 8.7 1389

G 1200 427 276 198 75 328 206 54 4.9 8.2 1156

B 1077 504 187 131 65 329 326 16 4.4 7.4 874

K 847.7 248 109 82 32 190 180 43 3.8 6.2 589

Suburban

L 3088 1099 558 201 34 753 720 45 6.1 9.7 1909

I 1295 150 116 80 65 202 301 16 3.9 6.4 615

J 115.7 97 82 47 37 76 28 8 3.6 5.6 478

E 248.4 63 45 35 25 66 61 15 3.2 4.4 301

Overall

Urban 173500 1073 413 185 11 3178 12768 373 5.5 9.0 1542

Suburban 3088 711 155 66 25 461 627 84 4.2 7.0 763

38

1100

1101

1102110311041105

38

Figure 1. Field assessment site maps with respect to number of pre-1978 built structures (a), racial diversity (b), education (c), and annual household income (d) of the sub-counties. These maps were created using Tableau software, which developed a geographical map of Monroe and Ontario County utilizing latitude and longitude coordinates for each sub-county boundary. In subfigure 1a, the size of the bubble depicts the number of pre-1978 houses found within that sub-county, and the color of the bubble denotes the respective sub-county. In subfigure 1b, 1c, and 1d, the distribution of racial diversity, education, and annual income data in their respective categories was represented by overlaying pie-charts into each sub-county.

Figure 2. Effect of distance (m) from potential source of lead-based paint on the lead concentrations (mg/kg) of residential soils in the urban or suburban neighborhoods of Rochester, NY; Data are expressed as mean (n = 20) + one standard deviation. Each subfigure expressed data for one residential property except site A, which was expressed in two subfigures (a, b). The subfigures are arranged left to right in rows based on their decreasing Y axis values, indicating the magnitude of soil Pb, of which the maximum was recorded in the front yard of site A after renovation (a) followed by the back yard of site A (b) before and after renovation; site L (c); site M (d); site H (e); site C (f), site B (g); site G (h); and site K (i).

Figure 3. SLL measurements in the backyard of site G showing two assessment lines; G1 (pink line), which was measured from the corner of the garage at site G, showed a decrease in soil Pb with distance except on one location (X), which was found to be horizontally aligned to the corner of the neighbor’s house, indicating it to be an additional source of LBP (fig. S.2). The line G2 was measured from a point adjacent to the property line between these two houses and was horizontally aligned to the corner of the neighbor’s house. G2 also showed a continual decrease with distance in the soil Pb concentration, except the increase at the last measured location point, which is the closest to the main house in site G, suggesting an additional source of LBP. The photograph is included in the manuscript with the permission of the homeowner.

Figure 4. Multivariate correlation among Pb in outdoor dust, paint and the depth index indicating how deep Pb is in the paint from a window (a), door (b), and front stairs (c) of site G; discarded old house materials from site A (d); pillars from the front porch in site H (e); and discarded old house materials left in a barn at site E (f).

Figure 5. Kinetic sorption (%) of Pb in presence or absence of WTR. Data are expressed as mean (n = 3) + one standard deviation. Mean comparison is conducted by two-way ANOVA followed by Tukey-Kramer honest significant difference test. It is conducted separately for each initial Pb concentration. Levels not connected by same letters are significantly different.

39

1106

1107

1108

1109

1110

11111112

1113

1114

1115

1116

1117

11181119

1120

1121

1122

1123

1124

1125

11261127

1128

1129

11301131

1132

1133

1134

39

Figure 6. Equilibrium sorption and desorption of Pb in 5% WTR. Linear (a), Freundlich (b), and Langmuir (C) isotherms of Pb-sorption by WTR in 24h. Data are expressed as mean (n = 3) + one standard deviation. % sorption (d) and % total desorption (e) of Pb by WTR in three desorption cycles of 24h each. Desorption of Al (mg/kg) from WTR during sorption of Pb (f). Mean comparison is conducted by one-way ANOVA followed by Tukey-Kramer honest significant difference test. Levels not connected by same letters aresignificantly different.

Figure 7. Effect of WTR concentrations on the leaching of lead (µg/L) from the paint chips collected from site A. Data are expressed as mean (n = 3). Mean comparison is conducted by two-way ANOVA followed by Tukey-Kramer honest significant difference test. Levels not connected by same letters are significantly different.

Figure 8: GIS maps showing distribution of SLL (mg/kg) in site A before (8a) and after (8b) the restoration/repainting of the built structure. These maps were created using ArcGIS Pro 2.5.2 by plotting the SLL (mg/kg) measurements into their respective latitude and longitude. An approximate property-line (dotted line) was overlaid to show the boundaries of the property segregating it from the side walk in south and west and three built structures of neighboring property in north and east. Based on different guidelines for safety, 7 sizes and color categories were used, each representing a range of SLL.

Figure 1:

40

1135

1136

1137

1138

1139

11401141

1142

1143

11441145

1146

1147

1148

1149

115011511152115311541155115611571158115911601161116211631164

40

Figure 2:

41

a b

c d

1165

11661167116811691170

41

0 0.5 1 1.5 2 2.50

5000

10000

15000

20000

25000

30000

35000

40000A5 (after renovation)

Distance from potential source of Pb based Paint (m)

Le

ad

Co

nc

en

tra

tio

n in

so

il (m

g/k

g)

0 1 2 3 4 5 6 7 8 9 100

2000400060008000

100001200014000160001800020000 A1

(before renovation)A2 (after renovation)

Distance from potential source of Pb based Paint (m)

Le

ad

Co

nc

en

tra

tio

n in

so

il (m

g/k

g)

0 1 2 3 4 5 60

500

1000

1500

2000

2500

3000L1 L2

Distance from potential source of Pb based Paint (m )

Lead C

oncentr

ation in

soil

(mg/k

g)

0 1 2 3 4 5 6 7 80

500

1000

1500

2000

2500

3000

3500 M

Distance from potential source of Pb based paint (m)

Le

ad

Co

nc

en

tra

tio

n in

so

il (m

g/k

g)

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50

500

1000

1500

2000H

Distance from potential source of Pb based Paint (m)

Le

ad

Co

nc

en

tra

tio

n in

so

il (m

g/k

g)

0 0.5 1 1.5 2 2.5 30

200400600800

10001200140016001800

C1

Distance from potential source of Pb based Paint (m )

Le

ad

Co

nc

en

tra

tio

n in

so

il (m

g/k

g)

42

ab

c

dD

e f

g h i

1171

1172

42

0 0.5 1 1.5 2 2.5 30

200

400

600

800

1000

1200B

Distance from potential source of Pb based Paint (m)

Le

ad

Co

nc

en

tra

tio

n in

so

il (m

g/k

g)

0 1 2 3 4 5 6 7 8

G1 G2

Distance from potential source of Pb based Paint (m)

Le

ad

Co

nc

en

tra

tio

n in

so

il (m

g/k

g)

0 1 2 3 4 5 6 7 80

100

200

300

400

500

600K

Distance from potential source of Pb based Paint (m )

Le

ad

Co

nc

en

tra

tio

n in

so

il (m

g/k

g)

43

dD e f

11731174

43

Figure 3:

44

G1

G2

X

11751176117711781179118011811182118311841185118611871188118911901191119211931194119511961197119811991200120112021203120412051206

44

Figure 4:

45

500

750

1000

1250

1500

1750

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

1

1.2

1.4

1.6

1.8

Pb in dust

r=0.8202

r=-0.6845

500 1000 1500

r=0.8202

Pb in paint

r=-0.9757

0.1 0.3 0.5 0.7

r=-0.6845

r=-0.9757

Depth index

1 1.2 1.4 1.6 1.8

50

100

150

200

250

300

350

0.15

0.20.25

0.3

0.35

0.4

0.45

0.5

1.5

2

2.5

3

Pb in dust

r=-0.6914

r=-0.7013

50 150 250 350

r=-0.6914

Pb in paint

r=0.7785

0.15 0.25 0.35 0.45

r=-0.7013

r=0.7785

Depth index

1.5 2 2.5 3

30

35

40

45

0.15

0.2

0.25

0.3

0.35

0.4

0.45

2

2.5

3

Pb in dust

r=0.8736

r=0.7256

30 35 40 45

r=0.8736

Pb in paint

r=0.9368

0.15 0.25 0.35 0.45

r=0.7256

r=0.9368

Depth index

2 2.5 3

0

5001000

1500

20002500

30003500

0

5

10

15

20

1

1.5

2

2.5

Pb in dust

r=0.8894

r=-0.2848

0 1000 2000 3000

r=0.8894

Pb in paint

r=-0.3090

0 5 10 15 20

r=-0.2848

r=-0.3090

Depth index

1 1.5 2 2.5

0

1000

2000

3000

4000

5000

6000

0

5

10

15

0.9

0.95

1

1.05

1.1

Pb in dust

r=0.9128

r=0.0000

0 2000 4000 6000

r=0.9128

Pb in paint

r=0.0000

0 5 10 15

r=0.0000

r=0.0000

Depth index

0.9 0.95 1 1.05 1.1

a b c

de

f

12071208120912101211121212131214121512161217121812191220122112221223122412251226122712281229123012311232123312341235123612371238

45

Figure 5.

0 5 10 15 20 250

20

40

60

80

100

120

0 mg/L Pb 5% WTR25 mg/L Pb 5%WTR200 mg/L Pb 5%WTR1000 mg/L Pb 5%WTR0 mg/L Pb 0%WTR25 mg/L Pb 0%WTR

Contact Time (h)

% S

orp

tio

n o

f P

b

46

Tukey-Kramer HSD % WTR*Contact time (h)

InitialPb

(mg/L)

Contacttime (h)

WTR(%)

MeanComparison

Letters25 0.13 5 A25 1.13 5 A25 2.13 5 A25 5.13 5 A25 10.13 5 A25 24.13 5 A25 0.13 0 B25 1.13 0 B25 2.13 0 B25 5.13 0 B25 10.13 0 B25 24.13 0 B

220 0.13 5 D220 1.13 5 C220 2.13 5 B220 5.13 5 A220 10.13 5 A220 24.13 5 A220 0.13 0 E220 1.13 0 E220 2.13 0 E220 5.13 0 E220 10.13 0 E220 24.13 0 E

1025 0.13 5 D1025 1.13 5 C1025 2.13 5 B1025 5.13 5 A1025 10.13 5 A1025 24.13 5 A1025 0.13 0 FG1025 1.13 0 G1025 2.13 0 EFG1025 5.13 0 E1025 10.13 0 EFG1025 24.13 0 EF

123912401241

1242

12431244124512461247

46

Figure 6.

0 10 20 30 40 50 60 70 80 90 100

02000400060008000

100001200014000160001800020000

Equilibrium Pb Concentration in Solution Ceq (mg/L)

Sorb

ed P

b b

y W

TR

Q (

mg/k

g)

0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2

3.2

3.4

3.6

3.8

4.0

4.2

4.4

f(x) = 0.55x + 3.21R² = 1

log Ceq

log

Q

0 10 20 30 40 50 60 70 80 90 100

0.0000

0.0010

0.0020

0.0030

0.0040

0.0050

0.0060

f(x) = 0x + 0R² = 0.98

Ceq

Ce

q/Q

47

a b c

d e f

12481249125012511252

1253125412551256125712581259126012611262126312641265126612671268

47

Figure 7.

48

Tukey-Kramer HSD(WTR % * Time in

days)

WTR%

Time

(days)

MeanCompari

sonLetters

0 1 A10 1 AB20 1 BC50 1 DE100 1 DE0 7 A10 7 CD20 7 CDE50 7 E100 7 E

1269127012711272

48

010

2050

100

0

1000

2000

3000

4000

5000

6000

7000

8000

7 Days 1 Day

% WTR

Le

ac

he

d L

ea

d in

so

luti

on

(µ

g/L

)

49

127412751276

49

Figure 8:

Figure 8 Footnote: Table S.4 includes a guideline created for the homeowners to explain each color category used in these maps, associated risk, and the recommended use of that location in their yards.

50

ba

127712781279

12801281

1282

1283

1284

1285

1286

1287

1288

1289

1290

50

Supplementary Information

Table S.1: Overview of outdoor Pb exposure in the urban and suburban residential properties in Rochester, NY.

ResidentialProperties

Neighborhood YearPb in Paint(mg/cm2)

Depth IndexMaximumPb in soil(mg/kg)

Maximum Pb inoutdoor dust

(g/ft2)A (Before

Renovation)Urban Unknown 33.7 1.0 50000 Not Measured

A (AfterRenovation)

Urban Unknown 20.3 6.3 173500 3635

B Urban 1923 2.3 1.0 1077 Not MeasuredC Urban 1930 5.9 10 1553 43.39E Suburban barn 1830 14.1 1.6 NA 8612F Suburban 2001 0 1.0 0 0G Urban 1910 6.4 4.4 1200 1969H Urban 1883 27.7 5.5 2378 14791I Suburban 1900 23.0 1.9 390.5 290.2J Suburban 1954 8.4 6.5 115.7 83.86K Urban 1930 19.4 9.0 847.7 156.4L Suburban 1923 33.8 10.0 3088 80.42M Urban 1910 38.2 4.7 3743 Not Measured

Table S.2: Parameters of linear and nonlinear bivariate regression fit of Pb concentrations (mg/kg) in residential soil by distance (m) from potential source of Pb based paint using data collected from all properties.

51

12911292

12931294

1295129612971298129913001301130213031304

51

Degrees of Fit R2 F ratio p value

1 0.074 11.13 0.00112 0.138 11.04 <0.00013 0.179 9.992 <0.00014 0.209 9.017 <0.00015 0.229 8.016 <0.00016 0.238 6.986 <0.0001

Table S.3: Distribution of soil Pb and associated risks in three yards of site A. Percentile, mean, standard deviation (SD) of soil lead level (SLL) (mg/kg) and potential blood lead level (BLL) (g/dL) in children and potential lead on play surfaces (PLOPS) (g /ft2) using three prediction

models developed by Mielke et al., (2007), Johnson and Bretsch (2002), and Mielke et al., (2006), respectively.

52

1305

130613071308130913101311131213131314131513161317131813191320132113221323132413251326

132713281329

52

Percentile SLL (mg/kg)

SLL Statistics(mg/kg)

Potential BLL (g/dL) PLOPS(g/ft2)

Equation 8Max 75% Median 25% Min Mean Std Dev N Equation 6 Equation 7

Frontyard

BeforeRenovation

4130 3516 854 392 211 1667 1568 7 7.1 10.6 2575

AfterRenovation

173500 26350 12050 3660 373 26167 39086 26 20.9 16.1 16220

Side yard

BeforeRenovation

1044 621.3 345.4 201 82 410 286 11 5.2 8.7 1358

AfterRenovation

740.6 550.95 367.5 121 61 362 237 9 5.3 8.8 1420

Back yard

BeforeRenovation

50000 19625 2899.5 645 232 11326 14045 24 11.3 13.1 6042

AfterRenovation

19000 2534.75 1644.5 469 68 2340 3211 42 9.0 11.9 4071

Table S.4: A guideline created for the homeowners to explain each color category used in the GIS map (figure 8), associated risk, and the recommended use of that location in their yards.

53

1330133113321333133413351336133713381339134013411342134313441345

53

Figure S.1: In situ measurements of soil Pb by XRF and marking the assessment with colored flags, a tool that was used to communicate with the homeowners to show them the distribution of soil Pb and risk hotspots instantly. Red, green, and blue flags were used to mark SLL > 400, 80 –

54

Colorcategories

Based on SLLRisk

CategoriesRecommended use

Dark Green58 mg/kg, which associates with the European Union’s

advisory limit for Pb in vegetables, calculated according tothe prediction models developed in this study.

Minimal toNo Risk

Safer locations to set up yardgardens

Light Green100 mg/kg, which is Minnesota’s current soil Pb advisory

and recommended federal standard by scientistsVery Low

Risk

Safer area for the children toplay

Not safe for home gardens

Yellow 400 mg/kg, current federal limit for children’s play area Low RiskLow exposure risk for the

children to playNot safe for home gardens

Orange 1200 mg/kg, current federal limit for remaining yardModerate

Risk

Not safe for the children toplay

Not safe for home gardensRed

Three extents of high concentrations exceeding allregulatory threshold

Risk HotspotsShould be completely avoided.Remedial actions are required.

MaroonBlack

13461347134813491350135113521353135413551356135713581359136013611362136313641365136613671368136913701371137213731374137513761377

54

400, and < 80 mg/kg, respectively. The photograph is be included in the manuscript with the permission of the homeowner. Photo courtesy: Michael Fisher.

55

13781379138013811382

13831384138513861387138813891390

55

Figure S.2: Lead concentration (mg/kg) in soil at Location A before (a) and after (b) renovation in 2018 and 2019 respectively. Data are expressed as mean (n=20) and + one standard deviation.

1

2

3

4

5

0 50000 100000 150000 200000 250000

Soil with burried paint chips

Bulk Soil

Pb concentration of soil (mg/kg)

1

2

3

4

5

6

7

8

9

0 50000 100000 150000 200000 250000

Soil with burried paint chips

Bulk Soil

Pb concentration of soil (mg/kg)

56

ba

1391139213931394

13951396139713981399140014011402140314041405140614071408

56

Figure S.3: Multivariate correlation among Pb in outdoor dust, paint and the depth index on discarded painted pieces of wood from site A (a), a window in site G (b), and a window in site H (c).

57

10

20

30

40

50

60

70

80

0.51

1.52

2.53

3.54

4.5

1.5

2

2.5

Pb in dust

r=0.1586

r=0.0527

10 30 50 70

r=0.1586

Pb in paint

r=0.9484

0.5 1.5 2.5 3.5 4.5

r=0.0527

r=0.9484

Depth index

1.5 2 2.5

1409141014111412141314141415141614171418141914201421142214231424142514261427142814291430143114321433

57

Figure S.4: Effect of soil Pb (mg/kg) on Pb concentration in plants (mg/kg) collected from all four home gardens of the tested residential properties and its bivariate regression with a polynomial square fit.

58

143414351436

143714381439

1440

1441

1442

58

download fileview on ChemRxivChemRxiv Final Pb Manuscript_Das.docx (9.98 MiB)