June 1996
                                    EPA 747-R-98-OO1 b
                  FINAL REPORT
           SOURCES OF LEAD IN SOIL:
              A LITERATURE REVIEW
          VOLUME II: STUDY ABSTRACTS
                   Prepared By
             Battelle Memorial Institute
            Technical Programs Branch
           Chemical Management Division
       Office of Pollution Prevention and Toxics
Office of Prevention, Pesticides, and Toxic Substances
       U.S. Environmental Protection Agency
              Washington, DC 2O46O

-------
                                  DISCLAIMER

       The material in this document has been subject to Agency technical and policy review
and approved for publication as an EPA report. Mention of trade names, products, or services
does not convey, and should not be interpreted as conveying, official EPA approval,
endorsement, or recommendation.

-------
                     AUTHORS AND CONTRIBUTORS
       This study was funded and managed by the U.S. Environmental Protection Agency.  The
review was conducted by Battelle Memorial Institute under contract to the Environmental
Protection Agency. Each organization's responsibilities are listed below.
                          Battelle Memorial Institute (Battelle)

       Battelle was responsible for conducting the literature search, obtaining and reviewing the
identified articles and reports, developing the conclusions and recommendations derived from the
review, and preparing this report.
                      U.S. Environmental Protection Agency (EPA)

       The Environmental Protection Agency was responsible for managing the review,
providing guidance on the objectives for the review and report, contributing to the development
of conclusions and recommendations, and coordinating the EPA and peer reviews of the draft
report. In addition, EPA provided access to study results not yet available in the general
literature. The EPA Work Assignment Managers were Samuel Brown and John Schwemberger;
the EPA Project Officers were Jill Hacker and Sineta Wooten; the EPA Section Chief was Phil
Robinson; and the EPA Branch Chiefs were Cindy Stroup and Brion Cook.

-------
This page intentionally left blank.

-------
                        TABLE OF CONTENTS




                                                                  Page






EXECUTIVE SUMMARY 	 Mi



Abstract A-1   Three City Urban Soil Lead Abatement Demonstration Projectl



Abstract A-2   Helena Valley Lead Study	 7



Abstract A-3   Baltimore, Maryland Urban Garden Soil Study	11



Abstract A-4   Brigham and Women's Hospital Longitudinal Lead Study  ... 1 3



Abstract A-5   The Butte-Silver Bow Environmental Health Lead Study .... 15



Abstract A-6   Charleston Lead Study	19



Abstract A-7   The Cincinnati Longitudinal Lead Study	21



Abstract A-8   The Omaha Lead Study 	25



Abstract A-9   Leadville Metals Exposure Study	27



Abstract A-10 The HUD Abatement Demonstration Study 	31



Abstract A-11 The National Lead Survey	33



Abstract A-1 2 Silver Valley - Revisited Lead Study	37



Abstract A-1 3 Midvale Community Lead Study Final Report	41



Abstract A-1 4 Minnesota Soil Lead Study  	45



Abstract A-1 5 New Haven, Connecticut Lead Study 	49



Abstract A-1 6 New Orleans Lead Study	53



Abstract A-1 7 Honolulu Park Soil Lead and Mercury Study	57



Abstract A-1 8 Telluride Lead Study 	59



Abstract A-1 9 Mt. Pleasant Soil Lead Study  	63



Abstract A-2O Illinois Soil-Lead Study	67



Abstract A-21 Dallas Soil Lead Contamination Study	71



Abstract A-2 2 Aspen Garden Soil Lead Study	73



Abstract A-23 El Paso, Texas Lead Study	75

-------
                  TABLE OF CONTENTS (continued)

                                                                  Page

Abstract A-24 Corpus Christi Soil-Lead Study	77

Abstract A-25 Maine Urban Soil Study  	79

Abstract A-26 Beltsville Roadway Study 	81

Abstract A-27 Heavy Metal Exposure Smelter Study  	85

Abstract A-28 Survey of Lead Levels Along Interstate 88O  	87

Abstract A-29 Albuquerque Street Dirt Lead Study	89

Abstract A-3O Identification of Lead Sources through Stable Isotope
           Ratio Techniques: Case Studies  	93

Abstract A-31 California Lead Study: Three High-Risk Communities	95

Abstract A-32 Champaign-Urbana Lead Study  	99

Abstract A-33 Cincinnati Roadside Soil Study  	101

Abstract A-34 Granite City Lead Exposure Study	103

Abstract A-35 Rochester Side-by-Side Dust Collection Study	1O5

Abstract A-36 Washington, D.C. Soil Lead Study	107

-------
                             EXECUTIVE SUMMARY
       Title X of the Housing and Community Development Act, known as the Residential
Lead-Based Paint Hazard Reduction Act of 1992, contains legislation designed to evaluate and
reduce exposures to lead in paint, dust, and soil in the nation's housing.  As amended in Title X,
§403 of Title IV of the Toxic Substances Control Act (TSCA), EPA is required to "promulgate
regulations which shall identify, for the purposes of this title and the Residential Lead-Based
Paint Hazards Reduction Act of 1992, lead-based paint hazards, lead-contaminated dust, and
lead-contaminated soil."
       Integral to the development of the §403 mandated standards (especially for soil) is
information on the sources, extent, and geographic breadth of elevated lead contamination of soil
("elevated" because lead is naturally present in soil in many geographic regions).  Such
information provides perspective when considering what level of lead in soil will be defined as
hazardous, and is suggestive of the potential efficacy of some interventions prompted by
promulgation of the standards.
       The purpose of the study summarized in this report was to search and review the
scientific literature on the sources of elevated soil-lead concentrations. More importantly, the
basis upon which elevated soil-lead levels were attributed to a particular source was also
identified.  Literature searches were conducted to identify relevant articles and were
supplemented by studies previously uncovered by the authors of this report. In all, 36 relevant
studies were identified, forming the basis for this report.
       The results of the literature search indicate that studies assessing soil-lead concentrations
and sources have been conducted in a wide variety of communities across the United States.  The
scientific literature, however, contains a preponderance of urban and smelter community studies.
Rural studies were relatively rare, their soil-lead levels usually used only as a measure of
background lead when examining results from urban environments.
       Consistent with what might be expected, three sources of elevated soil-lead levels were
identified in the literature: (1) lead-based paint; (2) point source emitters; and (3) leaded gasoline
emissions.  Eight types of supporting evidence, commonly reported in the literature as
justification for asserting that a particular source contributes to elevated soil-lead levels, were
                                            in

-------
identified: (1) residential area pattern; (2) paint-lead loading on exterior walls of residence; (3)
age of residence; (4) type and condition of housing; (5) distance from a hypothesized source of
elevated soil-lead levels; (6) ambient air-lead levels; (7) traffic volume on roadways in the
vicinity of areas being examined; and (8) community area pattern.
       The implications of the reviewed information concerning questions of source
apportionment were investigated. No definitive evidence was found within the literature,
however, suggesting a particular source can be regularly identified as responsible for elevated
soil-lead concentrations at a residence. In fact, many studies cite more than one source as
commonly responsible for elevated soil-lead levels. Moreover, labor- and cost-intensive
techniques for carefully apportioning the sources of lead exposure to soil suggest varying relative
contributions from candidate sources. It may be possible on a case-by-case basis to apportion the
responsible sources, but no generalizations are possible based on readily obtained categorical
factors (e.g., urban verus rural, northeast versus southwest). It is worth noting that within the
literature lead-based paint is often cited as the source responsible for higher concentrations of
lead in the surrounding soil; homes with extreme lead levels in their soil were often found to be
coated with lead-based paint.
       Although the results of this study suggest that a single source cannot be universally
associated with elevated soil-lead levels, the results do confirm the suspected pairwise
associations between elevated soil-lead levels and lead-based paint, leaded gasoline emissions, or
point source emissions.  As such, interventions targeting these sources should prove at least
partially beneficial in reducing lead contamination of soil. In particular, lead-based paint
interventions, such as those prompted by the promulgation of the §403 standards, should have an
additional benefit of removing a source of lead in soil above and beyond any benefit seen in
reduced indirect exposure to elevated dust-lead levels and direct exposure to  paint chips.
                                             IV

-------
ARTICLE ABSTRACTS

-------

-------
                      Abstract of Soil - Lead Studies
                                  Abstract A-1

Study Name: Urban Soil Lead Abatement Demonstration Project
             (Three-City Soil Lead Abatement Demonstration Project)

Study Dates: 1988-1992

Study Locations: Boston, Massachusetts; Baltimore, Maryland; Cincinnati, Ohio

References:

      Elias, R., Marcus, A., and Grant, L. (1996) "Urban Soil Lead Abatement Demonstration
      Project, Volume I: EPA Integrated Report," U.S. Environmental Protection Agency,
      Report No. EPA/600/P-93/001AF.

      Farrell, K., Chisolm, I, Rohde, C., Lim, B., Brophy, M., Strauss, W. (1992) "Baltimore
      Soil Lead Abatement Demonstration Project," U.S. Environmental Protection Agency
      Draft Report.

      United States Environmental Protection Agency. (1991) "Three City Urban Soil-Lead
      Demonstration Project: Midterm Project Update," Final Report.

      Weitzman, M., Aschengrau, A., Bellinger, D., Jones, R., Hamlin, J.  S.,  and Beiser,
      A. (1993) "Lead-Contaminated Soil Abatement and Urban Children's Blood Lead
      Levels," Journal of the American Medical Association. 269(13): 1647-1654.

      Aschengrau, A., Beiser, A., Belinger, D., Copenhafer, D., and Weitzman, M. (1994) "The
      Impact of Soil Lead Abatement on Urban Children's Blood Lead Levels: Phase II Results
      from the Boston Lead-in-Soil Demonstration Project," Environmental Research.
      67:125-148.

      Van Leeuwen, P., Bornschein, R., and Clark, S. (1992) "Cincinnati Lead Soil
      Demonstration Project," Presented at the Hazardous Materials Control/Superfund 92:
      13th Annual Conference and Exhibition, p 280-284.

      Mclntyre, D. and Fletcher, B. (1992) "Boston Lead-in-Soil Demonstration Project,"
      Presented at the Hazardous Materials Control/Superfund 92: 13th Annual Conference and
      Exhibition, p 274-277.

Objectives:

      Primary Objective: examine whether a reduction of environmental lead levels in dust and
      soil  will result in decreased blood-lead levels in children

-------
       Secondary Objectives:

          Boston: conduct the experiment so as to more clearly separate the beneficial results of
          abating lead in soil and in dust

          Baltimore: examine if a reduction in residential soil lead will result in a corresponding
          reduction in the amount of lead in interior household dust

          Cincinnati: test whether interim interior dust abatement, in conjunction with exterior
          dust and soil abatement, will result in a greater reduction in blood lead than either
          abatement method alone
Sampling Frame:
       Boston: Children aged 6 to 48 months old, with blood-lead levels between 7 and 24
       |ig/dL, residing in one of the study areas where there is a history of high incidence of lead
       poisoning. Each child enrolled was randomly assigned to one of three experimental
       groups: Study (52 children, 34 properties), Control A (51 children, 36 properties), or
       Control B (47 children, 30 properties).  The Study group received interior paint
       stabilization, interior dust abatement, and soil abatement in the first year of the study and
       no further treatment.  Control Group A received interior paint stabilization and interior
       dust abatement in the first year. Only interior paint stabilization was performed for
       Control Group B in the first year of the study. At the beginning of the second year of the
       study, soil abatement treatments were done for both control groups. Preliminary soil
       samples were collected to determine eligibility.  Detailed soil sampling was conducted
       before, immediately following, and nine months following the soil intervention.

       Baltimore:  Study areas were chosen for comparable demographic, soil-lead, and housing
       characteristics. The 63  Study Group properties received soil abatement and exterior paint
       stabilization, and 51 Control Group properties received  exterior paint stabilization (this
       includes six properties in the study area that did not receive soil abatement).  Soil-lead
       concentrations were sampled before and immediately following (for the study group) the
       intervention.

       Cincinnati:  Three study areas (A, B, and C) were chosen based upon a set of eight
       criteria designed to ensure comparable demographic and housing characteristics. The
       chosen areas contained  housing that had previously undergone extensive rehabilitation to
       remove or encapsulate most of the lead-based paint. A total of 215 land parcels were
       sampled (Area A: 55 land parcels; Area B:74 parcels; Area C: 86 parcels). There were
       nine phases of environmental and biological monitoring, which took place before and
       after each abatement cycle. Phase 00 included the project design and initial
       measurements. Area A received soil lead, exterior and interior dust abatement treatments
       between Phases 01 and  02. Area B received interior dust abatement between phases 01
       and 02 and exterior dust and soil abatement between Phases 05 and 06. Area C served as

-------
       a control group during the study, receiving all abatement treatments after Phase 09 was
       completed.  Soil samples were collected at seven points throughout the study.

Sampling Method:

       Boston: Soil samples were taken throughout the property using line source, targeted, or
       small area patterns. Both top and bottom of each 15 cm deep core sample were taken for
       analysis.  The most commonly used line source soil sampling method consisted of
       parallel lines 0.5 m from the foundation and boundary areas. More lines were added for
       larger properties. Each line was split into 7 meter segments, and a 2 foot by 2 foot
       composite soil sample was taken from a random point in each segment. An average of
       eight top and bottom samples were taken from each household.

       Baltimore: The soil area was measured to determine the sampling scheme (different
       schemes were used for small, large, and very large areas). The entire soil region
       surrounding the residence was sampled. The area was partitioned into front, back, etc.
       For each core sample, the top 2" and bottom 2" were retained.  For small areas, less than
       two meters in either direction, a single core sample was taken; areas less than  10 feet
       wide had a core sample at the house foundation and one at the boundary; for larger
       regions 16-20 feet long,  the region was divided in half and core samples were  taken at the
       foundation, boundary, and mid-yard lines.

       Cincinnati:  Soil samples were taken throughout a property using line source, targeted, or
       small area patterns. Both surface scrapings and core samples were collected for analysis.
       Line source sampling consisted of taking samples from lines that were 0.5 meters, 10 feet,
       and 15-20 feet away from a boundary, such as  a building or sidewalk, depending on the
       size of the yard. Areas too small for a line pattern were randomly sampled, and areas
       such as play equipment were targeted.

Analysis Method:  X-Ray Fluorescence

-------
Results for Soil:
Soil-Lead Concentrations (ug/g) for each Study Group by City and Round

City Study Group Statistic

Boston Study N3
Group 25th %tile
Median4
75th %tile
A. Mean4
Control A N
25th %tile
Median
75th %tile
A. Mean
Control B N
25th %tile
Median
75th %tile
A. Mean
Cincinnati Area A N
25th %tile
Median
75th %tile
A. Mean
Area B-B4 N
25th %tile
Median
75th %tile
A. Mean
Area B-D N
25th %tile
Median
75th %tile
A. Mean
Area B-F N
25th %tile
Median
75th %tile
A. Mean
Area C-G N
25th %tile
Median
75th %tile
A. Mean
Area C-M N
25th %tile
Median
75th %tile
A. Mean


1
35
1678
2413
3367
2625
36
1813
2477
3300
2831
30
1611
2268
3890
2728
112
79
273
1190
991
26
42
89
107
122





46
104
608
1421
1045
118
8
69
221
176
44
100
349
1179
809


2
26
98
125
160
139










104
0
0
88
166
26
50
87
115
117
92
219
758
1561
1141
48
125
760
1740
1256
120
17
114
268
202
55
139
637
1376
1013


3
35
70
113
192
234
35
1480
2148
3286
2502
30
1572
2155
3880
2679
104
18
28
64
132
26
58
93
131
153
88
228
667
1400
966
49
108
294
1159
817
120
46
124
308
573
49
109
338
795
654
Round2

4
21
100
174
284
206
22
161
278
505
429
17
110
204
240
307
100
24
41
78
140
26
62
99
126
277
86
307
768
1424
1084
48
93
379
1109
814
119
41
97
180
187
49
109
277
509
525


5















100
25
38
112
163
26
49
64
93
67
84
17
29
155
351
48
21
37
605
511
120
43
99
216
192
48
110
349
673
613


6















101
24
42
122
198
26
41
52
79
59
82
22
41
156
334
48
19
70
1406
929
119
45
109
202
179
47
102
363
848
501


7















103
25
37
117
167
26
37
55
77
68
88
24
68
465
508
47
32
50
713
835
121
38
111
197
169
48
132
416
860
530

-------
Soil-Lead Concentrations (ug/g) for each Study Group by City and Round (continued)
City
Baltimore














Study G rou p Statistic
Study Group N
25th %tile
Median
75th %tile
A. Mean
Control Group N
25th %tile
Median
75th %tile
A. Mean
Control Group N
in Study Area 25th %tile
Median
75th %tile
A. Mean
1 2
56
374
511
674
532
45
372
515
650
568
6
167
182
214
189
Round2
34567
56
12
29
73
69










3.
4.
5.
This table was abstracted from Table A-l of the Integrated Report. Dashed lines indicate when the soil intervention was performed.
Round is defined in the Integrated Report to be "a distinct period of time when one or more measurement were made...There is no
consistent pattern for when abatement occurred for the different individual cities." [Integrated Report p. 2-10]
N equals the number of properties or land parcels.
This is the median or mean of the average soil-lead levels for each property/parcel.
In the Integrated Report, Areas B and C were split into neighborhoods and analyzed separately.
Comments:
       There have been numerous publications on the Urban Soil Lead Abatement
       Demonstration Project.  In particular, review draft reports detailing the results for each
       city have been made available to the public.  However, these reports have not been
       formally released by EPA. Therefore, only results published in the Integrated Report
       were included.

       Complex statistical methods such as cross-sectional structural equations models, repeated
       measures analysis of variance, and longitudinal structural equations models were used to
       model changes in blood-lead concentrations that occur in response to changes in
       environmental lead.
Conclusions:
       "Soil abatement reduced soil-concentrations in all three studies, and there was no
       evidence of soil recontamination in either Boston or Cincinnati. There were no follow-up
       measures of soil in Baltimore that would detect recontamination." In Boston, the
       reductions of lead in the soil persisted throughout the two year follow-up period.
       [Integrated Report p. 1-18, 1-19]

-------
"There was some evidence for exterior dust recontamination in Cincinnati. The
Cincinnati group suggests that this might be caused by chipping and peeling lead-based
paint from the exterior surfaces of nearby buildings not included in the project."
[Integrated Report p. 1-18]

-------
                      Abstract of Soil - Lead Studies
                                  Abstract A-2

Study Name: Helena Valley Lead Study

Study Dates: 1983      Study Location: East Helena, Montana

References:

      Lewis and Clark County Health Department, Montana Department of Health and
      Environmental Sciences,  Centers for Disease Control, U.S. Department of Health and
      Human Services, and U.S. EPA. (1986) "East Helena, Montana: Child Lead Study,
      Summer 1983," Final Report.

      Centers for Disease Control. (1983) "East Helena, Montana Child Lead Study," Report by
      the CDC, Public Health Services, U.S.Department of Public Health and Human Services,
      Atlanta, GA.

Objectives:

      Evaluate residential exposures to smelter-associated lead and other heavy metals in
      neighborhoods near an operating lead smelter in East Helena, Montana.

Sampling Frame:

      Three study areas were identified as follows: Area 1: within 1  mile of the smelter; Area
      2: 1-2.25 miles from the smelter, Area 3: more than 5 miles from the smelter. All
      households with children aged  1 to 5 years and having resided  in the area at least 3
      months were sought for the study. In all, 396 children aged 1 to 5 years were sampled.

Sampling Method:

      Composite core samples one to three inches in depth were taken.  The first sample was a
      composite of 8 subsamples, four from the front of the house and 4 from the rear.
      Subsamples from the front of the were collected at least .5 meters from the curb and no
      more than 2.5 away from the curb. Subsamples from the rear were collected at least 6
      meters from painted surfaces. The second sample was a composite of 4 subsamples taken
      from the side of the house within 1 meter of the foundation. The third sample was a
      composite from the child's play area and the fourth sample was a composite from the
      garden if such was present.

Analysis Method: Atomic Absorption Spectrometry

Results for Soil:

                                          7

-------
                  Geometric Means
                                          Ranges
Soil
Levels
Comp
Side
Play
Garden
Correlation

Comp
Area 1
720
796
365
539
Between


Area 2
217
169
121
179
Soil Lead:
Side
.78
Area 3
86
92
73
95
Area 1
81 -
41 -
3-
70-

Play
.50
3414
7964
5770
2038



Area 2
58 - 1252
3- 883
3 - 6030
50- 599

Garden
.76
Area 3
54 - 237
47 - 500
28 - 373
58 - 162

Dust
.70
      Eq#

       1)
      Eq#

       1)
Indep. Var.        Coefficient

  Ln(soil)               0

             Type

 Log-Linear Multiple Regression
  Dependent Variable:

        Ln(PbB)

 Add. Indep. Variables:

Air, Dust, Home Location
   Mouthing Variables
Comments:
       All statistical analysis was done on log-transformed variables.

       The multiple (backwards stepwise) log-linear regression was performed to estimate the
       independent contribution to blood lead from lead in soil or in dust, or in soil and dust
       together.

       Principal component analysis was done to reduce the dimensionality of the mouthing
       behavior related variables.

       Paired t-tests comparisons were made between the  1 inch and the 3 inch core samples. (18
       pairs)

-------
Conclusions:
       "when the variables designating home location are added to the model, soil lead is no
       longer a statistically significant contributor to the variance in children's blood lead levels"
       [Reference p. 31]

       Soil lead contamination is associated with children's blood lead levels as evidenced by
       highly significant differences among the three  study areas in the lead levels of all four
       types of soil tested and these differences are mirrored by highly significant area
       differences in children's blood levels.  In addition, due to strong similarities in the age of
       the children, the positive association between blood and soil lead levels is not likely to be
       due to a confounding effect of age or age-related behavioral characteristics.

       "Children living closer to the smelter had higher blood lead levels than children living
       farther away."  [Reference p.  33]

       "Distribution of soils with high lead levels is not uniform throughout the valley."
       [Reference p. 21]

       "No statistically significant difference was found between the lead levels in 1 and 3 inch
       soils cores collected in front and side yards." [Reference p. 21]

       Significant correlations exist between environmental lead levels.

       No trends of association between erythrocyte protoporphyrin level and soil lead level are
       apparent.

-------
10

-------
                      Abstract of Soil - Lead Studies
                                 Abstract A-3

Study Name: Baltimore, Maryland Urban Garden Soil Study

Study Dates: 1982      Study Location: Baltimore, Maryland

References:

      Mielke, H. W., Anderson, J. C., Berry, K. I, Mielke, P. W., Chaney, R. L, and 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.

Objectives:

      Measure  and survey the distribution of soil lead within metropolitan Baltimore via
      measuring and surveying vegetable garden  soils.

Sampling Frame: All urban residences in Baltimore.

Sampling Method:

      Soil samples were randomly collected from 422 vegetable gardens within an area defined
      by a 30 mile radius  from a designated center point in downtown Baltimore.

Analysis Method:   Varian atomic absorption spectrophotometer with deuterium background
                   correction.

Results for Soil:

Lead (N= 422, p-value < *10'23)
Percentiles (p
Max
10900
90
777.5
80
421
70
258.5
60
167
50
100
3m)
40
55.5
30
35
20
24.5
10
14.5
Min
1
Comments:
       * proportionate measure of having more extreme result (i.e. more extreme clustering of
       the high group) by chance alone.
                                         11

-------
       Statistical analysis was completed using an approach termed Multi-Response Permutation
       Procedures or MRPP. The 422 soil samples were partitioned at the median value into
       high-low groups of 211 each.  The test statistic is based on the average distance between
       all pairs of sites within the high group.
Conclusions:
       "Although some literature proposes that house paint is the major source of lead
       contamination, the urban patterns of soil lead in Baltimore suggests that the inner-city
       lead contamination is due to another source." [p. 1367]  These sources include activities
       such as emissions from industries and incinerators, paints, solders, insecticides, rubbish
       and emissions from vehicular traffic.

       High levels of traffic are the reason for elevated lead levels of garden soils near unpainted
       brick structures in the inner city of Baltimore.

       Residences with elevated garden soil lead levels (those above the median) were more
       tightly clustered toward the center of Baltimore as to be explained by chance error in 1 to
       the trillion raised to the second power, (i.e. p-value < 10"23)
                                            12

-------
                       Abstract of Soil - Lead Studies
                                  Abstract A-4

Study Name: Brigham and Women's Hospital Longitudinal Lead Study

Study Dates: 1979-1983  Study Location: Boston, Massachusetts

References:

      Rabinowitz, M. B. and Bellinger, D. C. (1988) "Soil Lead - Blood Lead Relationship
      among Boston Children," Bulletin of Environmental Contamination and Toxicology.
      41:791-797.

      Rabinowitz, M., Leviton, A., Needleman, H., Bellinger, D., and Waternaux, C. (1985)
      "Environmental Correlates of Infant Blood Lead Levels in Boston," Environmental
      Research. 38:96-107.

Objectives:

      Identify sources and magnitude of lead exposure in early childhood (pregnancy through
      two years).

      Examine the association between soil lead and blood lead levels.

Sampling Frame:

      Births at Brigham and Women's Hospital between April  1979 and April 1981 were
      categorized into the highest, lowest or middle deciles of umbilical cord blood lead.
      Families resided within a 12-mile radius of the hospital, spoke English as their primary
      language, and the infants had no serious illness.  589 infants were eligible, 249 enrolled
      and 202 children completed the 2 year follow-up. Soil lead values were taken at the
      homes of 195 infants.

Sampling Method:

      Soil samples were taken from the top centimeter of soil from each of 3 points that were
      one meter apart and at least 3 meters from any road or structure. Soil samples were
      collected at the 18 and  24 month visits and preference was given to any obvious play area.

Analysis Method: Flame atomic absorption
                                         13

-------

N
195
Mean
(ug/g)
702
Median
(ug/g)
365

Range
7 - 13240
Results for Soil:


       PbS
Correlation Between Soil Lead with Blood and Environmental Lead:

                   PbA       Road Score       Pb Paint        PbB          PbD
     PbS           .18            .01             -.07           .30           .40

       Eq #           Indep. Var.          Coefficient         Dependent Variable:
        1)             Ln(PbS)         8.9 ng/L/ng/kg             Ln(PbB)
        2)             Ln(PbS)         .8 In ng/g/Hg/kg             Ln(PbB)
       Eq #                        Type                     Add. Indep. Variables:
        1)                  Simple Log linear reg.                      None
        2)                 Stepwise Log linear reg.            Ln(PbD), season, refinish
Comments:
       The stepwise regression used both forward and backward elimination.  Results shown are
       for 18 month olds (N = 212). For all other age groups (6, 12, 24) PbS had a coefficient
       ofO.
       The simple log-linear regression uses all of the data, not just the data up to 18 months.
       Correlations were done using Spearman's nonparametric rank correlation procedure.
Conclusions:
       There is a connection between PbS and PbB.
       Lead levels in soil and dust rose across the levels of blood lead.
       Lead levels in soil, air and dust were strongly correlated.
                                          14

-------
                       Abstract of Soil - Lead Studies
                                  Abstract A-5

Study Name: The Butte-Silver Bow Environmental Health Lead Study

Study Dates: 1990         Study Location: Butte, Montana

References:

      Butte-Silver Bow Department of Health and Department of Environmental Health,
      University of Cincinnati. (1991) "The Butte-Silver Bow Environmental Health Lead
      Study," Draft Final Report.

Objectives:

      "ascertain whether or not the children of Butte... are currently exhibiting elevated blood
      lead concentrations"

      "identify and quantify accessible lead...in the environment"

      "establish the extent of association between sources of environmental lead and blood lead
      and to estimate the relative contribution of these sources of lead to the children's blood
      lead"

      "provide residence-specific data to guide future remediation efforts"

                               [Quotes from reference p. 6]

Sampling Frame:

      "Butte historically has been an important mining,  milling and smelting district."
      [Reference p. 3] Children less than 72 months of age residing for at least 3 months prior
      to survey, in one of seven study neighborhoods (A-F) in the environs of Butte. The seven
      neighborhoods vary in their proximity to known mining wastes and age of neighborhood.
      Areas A and G are close to old mine sites while Areas C and D are near a copper mill and
      tailings.  Areas E and F are homes built after WWI not near any mine or tailings. Finally,
      Area B was mostly mobile  homes close to mine tailings. Blood samples were taken from
      430 individuals with 294 less than 72 months of age. There were  650  composite soil
      samples collected.

Sampling Method:

      Soil core samples were taken at 2 cm depth and exterior surface dust samples were
      collected. A composite sample of soil  cores from  perimeter (three feet from building

                                          15

-------
       wall) of the residence was collected.  Each composite consisted of 8 to 12 soil core
       samples. Composite cores were also taken from garden, bare yard area, and play
       areas/sand box if such were present.  A composite exterior surface dust sample was
       collected by vacuum over paved areas and other hard surfaces near the residence's
       entrance.

Analysis Method:  Atomic Absorption Spectrometry

Results for Soil:

House Perimeter
Garden
Play Area
Bare Area
Area A
AreaB
AreaC
AreaD
AreaE
AreaF
Area G
N
215
82
169
184
145
10
7
9
21
12
11
Geometric
Mean (mini)
515.51
317.64
254.63
430.67
750.24
249.75
139.45
230.31
151.02
178.17
1030.56
Geometric Std.
Dev. (ppm)
2.89
3.11
3.80
3.09
2.45
1.70
2.70
2.33
2.14
1.89
1.46
Range
5% - 95%
71.7-2356
49.9- 1399
19.5- 1639
62.3 - 2460







Correlation Between Soil Lead with Blood and Environmental Lead:
b Sgard
.83
.53
.10
PbSbare
.86
.60
.26
PbSplay
.62
.51
.24
Ext. Pt
.59
.54
.19
Int. Pt
.50
.43
.19
Rd. Dst
.64
.51
.19
    PbSpeim
   Int. Dust

  Blood Lead
 Corr (PbSpeim, Blood Lead) = .24

Eq #  Indep. Var.   Coefficient   Dependent Variable:
 1)    Ln(PbSperm)    0
 2)    Ln(PbSperm)  22.485
Ln(PbB)
Ln(PbDinterior)
                                          16

-------
     Eq#                   Type                       Add. Indep. Variables:

      1)       Log-linear Least Squares         Ln(PbDint), Work, Child's age, Mouth,
               Backwards Elimination          House age Socio-economic Status

      2)       Structural Equations Analysis     Ln(PbDint), Ln(PbPext), Child's age, Mouth,
               Log-Linear Equation             Work

Comments:

      Field duplicates were collected at 10% of the sampling sites.

      Regression analysis was based on only 192 families. One child was randomly selected
      per family. Regression was performed to examine those factors most related to blood
      lead levels.

      The Structural Equations Analysis had four dependence relations and was performed to
      investigate the pathways  of lead contamination of children's blood.

Conclusions:

      "Lead based house paint  contributes to lead contaminated soil which in turn contributes
      to lead contaminated house dust. Only lead in house dust was directly related to blood
      lead."

      "40% of the variability in soil lead is attributable to lead based paint.  The remaining 60%
      is due to the heterogenous distribution of lead in soil, and lead from other sources."

      "Locations of residence and age of housing were found to be strong predictors of soil lead
      and dust lead concentrations.  Environmental lead concentrations were found to be the
      highest in older neighborhoods."

                            [Quotes from reference pp. 125-126]
                                           17

-------
18

-------
                       Abstract of Soil - Lead Studies
                                  Abstract A-6

Study Name: Charleston Lead Study

Study Dates: 1973   Study Location: Charleston, South Carolina

References:

       Galke, W. A., Hammer, D. I, Keil, J. E., and Lawrence, S. W. (1975) "Environmental
       Determinants of Lead Burdens in Children," In:  International Conference on Heavy
       Metals in the Environment: Symposium Proceedings, T. C. Hutchinson, S. Epstein, A. I.
       Page, J. VanLoon and T. Davey (eds.), Institute for Environmental Studies, Toronto, ON,
       Canada, 3:53-74.

       See Also: US Environmental Protection Agency Report No. EPA-600/J-78-022.
       Washington, D.C.: Available from: NTIS, PB-283567. Springfield, VA.

Objectives:

       Examine two hypothesis:

       (1) Blood lead levels are positively related to soil lead
         levels.

       (2) Blood lead levels are positively related to automobile
         traffic, independent of soil lead exposure.

Sampling Frame:

       Four recruitment areas:  high traffic/high soil lead; high traffic/low soil lead; low
       traffic/high soil lead; and low traffic/low soil lead, were established based on prior
       knowledge of population characteristics. African American pre-school aged children
       were sampled from each of these areas.

       In all,  194 children from 170 families were sampled, but because of incomplete data, only
       187 children from 164 families were used in subsequent analysis.

Sampling Method:

       A soil sample was taken from child's chief play area. The exact methodology is not
       available in reference.

Analysis Method: Atomic absorption spectrophotometry

                                          19

-------
Results for Soil:


                             Median
                              (ppm)              75th %tile              Range
         PbS                  585                   1400                9-7890


                                                  Median
                                              Traffic Volume
                                                 (cars/day)               Range
 Facing Street           PbS<585 ppm               100                100 - 12275
                        PbS>585ppm               100                100-15575

 All Streets              PbS<585 ppm              1100               100-35300
 Within 76 Meters       PbS>585 ppm              3200               100-35400

Comments:

       Several regressions were performed, regressing soil lead levels against each paint lead
       variable and traffic volume.  Analytical results were not available in the reference.  A
       separate analysis was completed for each traffic volume representation. Additional
       regressions were performed regressing PbB on soil, traffic, and paint lead levels.

       Additional descriptive statistics were presented for various soil, traffic, age, and blood
       levels.
Conclusions:

       Soil lead exposure was the best index to environmental lead in this study.

       "Regressing soil lead levels against each paint lead variable and traffic volume variable
       independently found significant relations between soil lead and exterior siding paint,
       window sill paint, and traffic (as a dichotomous variable). Still, none of these
       relationships explained much of the soil lead variability."  [Reference p. 61]

       Children's blood lead level was positively related to soil, traffic, and paint (determined
       using multiple regression analysis [i.e. a positive coefficient])

       Blood lead levels were related to exposure to traffic, independent of soil exposure
       (determined using multiple regression analysis).
                                           20

-------
                       Abstract of Soil - Lead Studies
                                  Abstract A-7

Study Name: The Cincinnati Longitudinal Lead Study

Study Dates: 1980-1987    Study Location: Cincinnati, Ohio

References:

      Bornschein, R. L., Hammond, P. B., Dietrich, K. N., Succop, P., Krafft, K., Clark, S.,
      Berger, O., Pearson, D., and Que Hee, S. (1985) "The Cincinnati Prospective Study of
      Low-Level Lead Exposure and Its Effects on Child Development: Protocol and Status
      Report," Environmental Research. 38:4-18.

      Que Hee, S .S., Peace, B., Clark, S., Boyle, J. R., Bornschein, R. L., and Hammond, P. B.
      (1985) "Evolution of Efficient Methods to Sample Lead Sources, Such as House Dust and
      Hand Dust, in the Homes of Children," Environmental Research. 38:77-95.

      Bornschein, R. L., Succop, P. A., Krafft, K. M., Clark, C. S., Peace, B., and Hammond, P.
      B. (1986) "Exterior Surface Dust Lead, Interior House Dust Lead and Childhood Lead
      Exposure in an Urban Environment," Conference in Trace Metals in Environmental
      Health, Columbia, MO.

      Bornschein, R. L., Succop, P., Dietrich, K. N.,  Clark, S., Que Hee, S., and Hammond, P.
      B. (1985) "The Influence of Social and Environmental Factors on Dust Lead, Hand Lead,
      and Blood Lead Levels in Young Children," Environmental Research. 38:108-118.

Objectives:

      Integrate information on exposure history (from birth to 5 years), cognitive and
      behavioral development, and health and social  functioning in order to delineate the
      association between chronic, low level lead exposure and behavioral development.

Sampling Frame:

      Enrolled expectant mothers residing in one of a group of census tracts identified as
      having a long history of producing children with elevated blood lead levels. The mothers
      were patients at one of three prenatal clinics in  the area intending to deliver at Cincinnati
      General Hospital.
                                         21

-------
Sampling Method:
      Surface scrapings were collected from play areas and/or from immediately outside
      dwelling unit entry.  Samples were taken from 80 houses.
Analysis Method: Atomic Absorption Spectroscopy
Results for Soil:
 PbS
N    Geometric Mean
           (ppm)
80         1360.32
Geometric Std. Dev.
      (ppm)
       4.67
                                                                      Range
 76-54519
        Age/Type-condition of Home
            20th Century/Public
         19th Century/Rehabilitated
          19th Century/Satisfactory
          19th Century/Deteriorated
                                   N
                                   14
                                   18
                                    7
                                   13
                Geometric Mean (ppm)
                         572
                         804
                        2540
                        2670
Correlation Between Soil Lead with Blood and Environmental Lead:
 Ln(PbS)
           Ln(XRF/Hazard)
                 .41
      Ln(PbD)
         .57
Ln(PbB)
   .30
   Eq #       Independent Variable
    1)              Ln(PbS)

   Eq #               Type
    1)          Structural Equations
                              Coefficient
                                 .268
                  Dependent Variable:
                        Ln(PbD)
                                Additional Independent Variables:
                                  Ln(XRFHAZ), Ln(PbH),Ln(PbD)
                                        22

-------
Comments:
       There were three equations in the structural analysis. The results presented are reduced
       structural model for children 18 months old.
Conclusions:
       "There is evidence to support the hypothesis that exterior environmental lead can result in
       blood lead elevations through the path PbS—>PbD—>PbH—>PbB. An increase in PbS
       from 0 to 1000 ppm results in an indirectly mediated increase in PbB of 6.2 |ig/dl."
       [Reference 3 p. 331]

       "PbS is low outside public housing units, while being considerably higher and more
       variable outside other home types."  [Reference 3 p. 324]
                                           23

-------
24

-------
                       Abstract of Soil - Lead Studies
                                  Abstract A-8

Study Name: The Omaha Lead Study

Study Dates: 1971-1977    Study Location: Omaha, Nebraska

References:

       Angle, C. R. and Mclntire, M. S. (1979) "Environmental Lead and Children: The Omaha
       Study," Journal of Toxicological and Environmental Health. 5:855-870.

       Angle, C. R., Mclntire M. S., and Colucci, A. V. (1974) "Lead in Air, Dustfall, Soil,
       House Dust, Milk and Water: Correlation with Blood of Urban and Suburban School
       Children," Trace Substances in Environmental Health - VIII, Ed. D. D. Hemphill, pp 23-
       29.

       Angle, C. R., Marcus, A. H., Cheng, E. H., and Mclntire, M. S. (1984) "Omaha
       Childhood Blood Lead and Environmental Lead: A Linear Total Exposure Model,"
       Environmental Research. 35:160-170.

Objectives:

       Obtain data to predict and understand the relationship between air lead levels, dustfall
       rates and the resulting soil accumulation as it pertains to the blood lead concentrations of
       children.

Sampling Frame:

       Children were recruited from three areas of interest: urban commercial (C), urban mixed
       (M) which is a residential area contiguous with downtown, and a suburban (S) area. No
       attempt was made to randomize (all subjects were volunteers) nor obtain equal
       distribution among the three areas.  In total, 1074 children were sampled.  Of those
       sampled, 242 were 1-5 years old and 832 6-18 years old.

Sampling Method:

       Two inch core samples were taken halfway between the building and lot line.
       Samples were collected on all four sides of the building. The values reported
       were the arithmetic means for each  site. Soil samples were collected from 37
       individual houses and 148 samples were also taken at each child's school. A total
       of 20 sites were sampled 1 to 5 times.

Analysis Method: Not available in reference

                                          25

-------
Results for Soil:
 SiteC

 SiteM

 SiteS
N        Geometric Mean (ug/g)

69                 262

56                 339

51                 81
       Range 5%-95%

           53-1615

           20-4792

            16-341
Correlation Between Soil Lead with Blood and Environmental Lead:
             Ln(PbA)    Ln(PbDF)    Ln(PbHD)    Ln(PbB)    Ln(PbB) 1-5 yr, 6-18 yr

  Ln(PbS)       .37          .35          .27         .29               .29
 Eq #     Independent Variable

    1)               Ln(PbS)

    2)               Ln(PbS)

 Eq #                Type

    1)     Multiple log-linear regression

    2)     Non-linear regression
              Coefficient

                    .1253

                    .0046
Dependent Variable

          Ln(PbB)

          Ln(PbB)
                     Additional Independent Variables

                              House dust, air

                              House dust, air
Conclusions:
       The re-analysis of the data concluded that the regression model is better than the power
       model in explaining the relationship between environmental lead and blood levels.

       Continued control of PbA will likely result in the existing Pb in the environment, such as
       soil lead, becoming a more significant contributor of Pb to blood than PbA.

       Community-wide changes in PbB are multifactorial. Air, soil, water, housing, and
       socioeconomic shifts all have an additive or possibly even synergistic effect on blood
       lead.
                                         26

-------
                       Abstract of Soil - Lead Studies
                                  Abstract A-9

Study Name: Leadville Metals Exposure Study

Study Dates: 1987 - 1988      Study Location: Leadville, Colorado

References:

       Colorado Department of Health, University of Colorado at Denver, and U.S. Department
       of Health and Human Services. (1990) "Leadville Metals Exposure Study," Final Report.

       Cook, M., Chappell, W., Hoffman, R., and Mangione, E. (1993) "Assessment of Blood
       Lead Levels in Children Living in a Historic Mining and Smelting Community,"
       American Journal of Epidemiology.  137(4):447-455.

Objectives:

       "Characterize the levels of heavy metals in the residential environment and the
       relationships of human exposure to environmental concentrations."

       "Determine the extent to which environmental, behavioral and socio-economic factors are
       predictive of heavy metal  exposure."

       "Determine the levels of heavy metals and other indicators of metal toxicity in the blood
       and urine of individuals, principally young children, residing in Leadville."

                            [Quotes from first reference p. 1]

       Compare the results to national averages and the results from similar mining, milling, and
       smelting communities.

Sampling Frame:

       All households with children aged 6 to 71  months residing in Leadville census tract for at
       least three months preceding the study; an additional sample consisted of households with
       children  aged 6 years and  older. A total of 2631 eligible individuals were identified with
       239 children between 6 and  71 months of age.

       The initial sample size was 300 individuals of which 150 were children 6 to 71 months
       old. Twenty-five children from each sex were randomly selected in each of three age
       groups:  6-14,  15-44, 45-65 years.
                                          27

-------
          The final sample size was 233 individuals of which 150 were children 6 to 71 months
          of age; 29, 6 - 14 years old; 28, 15 - 44 years old; 26, 45 - 65 years old.

       Soil and dust samples were collected from 105 households.

Sampling Method:

       A composite sample of four 1 inch core samples (taken along roof edge drip lines) were
       taken at the front, rear and in reported play areas. In addition, surface scrape samples were
       collected at the entry-way and in reported play areas.
Analysis Method:
       Described in "EPA User's Guide to the Contract Laboratory Program" December, 1986
       and "Contract Laboratory Statement of Work" July, 1985.
Results for Soil:
Correlation Between Soil Lead:

                D. Fir       D.  Sill

   C. Rear        .24          .26


 Eq #     Independent Variable

    1)             Ln(C. Rear)

 Eq #                Type

    1)      Linear Regression, Stepwise
              Forward Elimination
Geometric
Mean (ppm)
1108.3
914.7
868.1
C. Frnt
.52
Coefficient
.13
Geometric Std
Dev. (ppm)
2.8
3.1
3.7
C. Ply S.
.64
Range
5% - 95%
49- 15100
10 - 27800
2.7 - 8620
Ply S. Entry
37 .35
Dependent Variable

Ln(PbB)
Additional Independent Variables
C. Frnt, C. Ply, Scrapes,
  Behavior variables
                                          28

-------
Sample
Type
Children
<6 years
Core Rear

Contaminant
Level (ppm)
<500
>500
Blood Lead
Level <10
(ug/dl)
33
56
Blood Lead
Level >10
(ug/dl)
4
57
                                                                      X2         Odds
                                                                                 Ratio
                                                                     18.1         8.40
                                                                                 (2.8-
                                                                                25.26)*
Comments:

       * 95 % confidence interval.

       All statistical analyses were performed with log-transformed blood lead values.

       Values below the detection limit were replaced with /^ of the detection limit for statistical
       analysis.

       All children were used in the analysis including siblings.

       Stepwise (forward elimination) regression was used to develop models explaining the
       variation in blood lead levels.

       Odds ratios were calculated for behavioral characteristics and blood lead levels.
       Significant associations were tested using a Chi-Square test.


Conclusions:

       "Core samples collected at the rear of the house were significantly correlated with all
       other dust and soil samples collected." [Reference p. 32]

       "Significant associations exist between blood lead levels greater than or equal to 10 |ig/dl
       and soil lead levels greater than 500 ppm for core samples collected taken at the rear of
       the house...and scrape samples collected in the play area." [Reference p. 33]

       "Soils in Leadville were found to have much higher levels of lead than soils in other parts
       of the United States." [Reference p. 51]
                                            29

-------
30

-------
                      Abstract of Soil - Lead Studies
                                Abstract A-1O

Study Name: The HUD Abatement Demonstration Study

Study Dates: 1989-1990

Study Location:  Baltimore, MD; Washington, D.C.; Seattle, WA; Tacoma, WA; Indianapolis,
                IN; Denver, CO; Birmingham, AL

References:

      U.S. Department of Housing and Urban Development. (1991) "The HUD Lead-Based
      Paint  Abatement Demonstration (FHA)," Office of Policy Development and Research,
      Washington, D.C.

Objectives:

      Develop reliable estimates of the cost of lead-based paint abatement.

      Examine the extent of exposure to hazards experienced by workers during abatement.

      Examine the extent of post-abatement hazards presented to new residents.

Sampling Frame:

      HUD-owned, FHA foreclosed, single family homes in target cities.

      There were 455 paired (pre- and post-abatement) soil samples taken corresponding to 130
      dwelling units.

Sampling Method:

      A series of samples was collected approximately one to three feet from the base of each
      exterior wall. Each series consisted of 5 samples, which were collected at evenly spaced
      intervals and then combined to make a composite which was then analyzed for lead
      content.

Analysis Method: Flame atomic absorption
                                        31

-------
Results for Soil:

                                      N                 Arithmetic Mean (ppm)

 Pre-Abatement                       455                         755.0
 Post-Abatement                       455                         867.5


Comments:

       A paired t-test was used to determine significance between before and after abatement
       means.

       Additional tests were also completed. The methodology of these tests is unavailable in
       the reference. These additional tests examined the relationship between abatement
       strategy and increases in soil lead greater than 250 ppm.

Conclusions:

       Post-abatement soil lead levels were significantly higher, at the .01 level, than pre-
       abatement levels.

       There was some evidence of a statistical relationship between abatement strategy and
       increases in soil lead greater than 250 ppm. Units abated under the Hand-Scraping and
       Chemical Strategies were most likely to experience soil lead increases of over 250 ppm.
                                           32

-------
                       Abstract of Soil - Lead Studies
                                 Abstract A-11

Study Name: The National Lead Survey

Study Dates: 1989-1990

Study Location: Various urban areas throughout the US; 30 counties in the 48 contiguous
                states

References:

      Weitz, S., Clickner, R. P., Blackburn, A., Buches, D., et al. (1990) "Comprehensive and
      Workable Plan for the Abatement of Lead-Based Paint in Privately Owned Housing:
      Report to Congress," U.S. Department of Housing and Urban Development, Washington,
      D.C.

      Rogers, J., Clickner, R., Vendetti, M., and Rinehart, R. (1993) "Data Analysis of Lead in
      Soil," U.S. Environmental Protection Agency, Office of Pollution Prevention and Toxics,
      Report Number EPA 747-R-93-011.

Objectives:

      Estimate the number of dwelling units in the U.S. with interior and exterior lead based
      paint, by year built, type of housing, threshold level of lead concentration, and census
      region.

      Estimate the number of multifamily residences with lead-based paint in common areas,
      by the year built, threshold level of lead concentration, and census region.

      Estimate the costs of abating lead-based paint in public and privately owned housing.

      Investigate the associations among soil, dust, and paint measurements.

      Identify the variables that predict soil and dust lead levels.

Sampling Frame:

      Multi-stage stratified sampling design by dwelling unit age and type, of housing in
      the U.S. built before 1980.

      284 privately owned dwelling units and 97 public housing units were sampled.
                                          33

-------
Sampling Method:
       Data were collected from three soil sample sites: along the drip line of the XRF wall (less
       than 1 foot from the foundation); at a remote location (l/2 the distance from the property
       line and the foundation, between 5-30 feet from the foundation); at the most-used
       entryway. Three soil samples were also taken from playgrounds. Three subsamples were
       collected at each site and composited into a single sample. The top  2-3 cm were sampled
       using a corer.
Analysis Method:    Inductively coupled plasma-atomic emission spectrometry.
Results for Soil:
                                                   Geometric Std.  Dev.
                     N     Geometric Mean (ppm)         (PPm)           95% CI
 Soil Entry          260             83                    4.35            70-100
 Soil Remote        253             47                    4.14             40-56
 Soil Drip           249             72                    5.37             58-89

 Correlation With Soil Lead:
                                              Soil Entry    Soil Drip    Soil Remote
 Soil Entry (ppm)                                 -            .715         .609
 Soil Drip (ppm)                                 .715           --           .678
 Soil Remote (ppm)                              .609          .678
 Exterior Paint Lead (mg/sq cm)                    .277          .274         .274
 Exterior Painted Area (sq ft)                      .134          .156         .151
 Dwelling Unit Age                              .584          .590         .534
 Traffic Vehicles (miles/day)                      .202          .238         .281
                                          34

-------
Multiple Log Regression Results:
                                                  Parameter Estimates for
                                                 Dependent Variable Model
                                          Drip Line
                                               Entry
Remote
Independent Variable
 Paint Lead
      Dry Rooms
      Wet Rooms
      Exterior

 Proportion Damaged Paint
      Dry Rooms
      Wet Rooms
      Exterior

 Painted Surface Area
      Dry Rooms
      Wet Rooms
      Exterior

 Number of Wet Rooms

 Number of Dry Rooms

 Local Traffic

 Unit Age

Comments:
       *   Significant at the 5 percent level.

       Correlations were calculated on log-transformed data.

       Other regressions were also done with principal components and using linear
       combinations representing differences in dust concentrations between wet and dry rooms.

Conclusions:

       "...the probability of excessive soil lead somewhere on the property...is four to five times
       larger when exterior lead-based paint is present than when it is not."  [First Reference
       p. 3-17]
.02
.08
.07*
-4.1*
-1.7
-0.3
-.16
-.06
.03
-.57*
0.00
.08
1.55*
.07
.07
.05
-8.5
1.1
0.1
-.09
-.07
.09
-.54*
-.13
-.10
1.20*
-.03
.09*
.05
-4.6
-1.6
0.4
-.21*
.02
.02
-.35
.18
.26*
.95*

-------
"There is also strong statistical evidence that exterior lead-based paint, especially
defective paint, is an important source of lead in soil; and lead in the soil, as well as
interior lead-based paint, is a source of interior dust lead on the floors." [First Reference
p. 3-32]

The entryway and drip line soil concentrations were not significantly different. However,
the remote soil concentrations was, on the average, significantly lower than either the drip
line or entryway soil concentrations.

"The strongest predictors of soil-lead are dwelling unit age and county of residence, for
all three sample locations." [Second Reference p.89]

"The parameter estimates suggest that paint lead from dwelling surfaces contribute more
to the entrance and drip line soil lead samples than to the remote sample." [Second
Reference p.89]

Local traffic volumes were only significant for the remote soil samples. However,
"...significant positive estimates for the square of the traffic volume for the drip line and
entrance samples suggests that a contribution of lead from traffic may be significant at
higher traffic volumes." [Second Reference p.89]

"...a combination of increased exterior paint damage in conjunction with higher exterior
lead loadings is associated with increased drip line soil lead concentrations..." [Second
Reference p.93]

Soil lead concentrations generally decreased with increasing age of the the dwelling.
                                     36

-------
                       Abstract of Soil - Lead Studies
                                 Abstract A-12

Study Name: Silver Valley - Revisited Lead Study

Study Dates: 1983         Study Location: Kellogg, Idaho

References:

      Panhandle District Health Department, Idaho Department of Health and Welfare, Centers
      for Disease Control, and U.S. Environmental Protection Agency. (1986) "Kellogg
      RevisitedSl983: Childhood Blood Lead and Environmental Status Report," Final Report
      of the U.S. Public Health Service.

      Yankel, A. 1, von Lindern, I. H., and Walter, S. D. (1977) "The Silver Valley Lead
      Study: The Relationship Between Childhood Blood Lead Levels and Environmental
      Exposure," Journal of the Air Pollution Control Association.  27(8):763-767.

Objectives:

      "To determine the current blood levels of children aged 1-9 in the Silver Valley area."

      "To compare the blood levels of that population to national norms."

      "To study the relationship between children's blood levels and the environmental
      exposures under current conditions."

      "Identify sources,  transport mechanisms, and factors important to lead absorption under
      current conditions."

                              [Quotes from reference p. 2]

      Document the health and environmental improvements since the initial Silver Valley
      Lead study in 1974 and the closure of the smelter in 1981.

Sampling Frame:

      Three study areas were identified; Area 1: all residences within 1 mile of the smelter;
      Area 2: all residences 1-2.25 miles from the smelter, Area 3: all residences 2.5 - 6 miles
      from the smelter.  All households in Area 1 were sampled while every other residence
      was sampled in Areas 2 and 3. A total of 364 children were tested.

Sampling Method:

                                          37

-------
       Composite core samples were taken from the top one inch of soil. The first sample was a
       composite of 8 subsamples, 4 from the front of the house and 4 from the rear.  These
       subsamples were collected at least 1.5 meters from the curb, street, sidewalk or the house.
       The second sample was a composite of 4 subsamples taken from the side of the house
       within 1 meter of the foundation. The third and fourth samples were composites from the
       child's play area and the garden if such was present. In addition, at every 7th residence in
       the survey, duplicate soil core samples and special soil samples were collected via a 5-
       inch hand trowel.

Analysis Method: Atomic Absorption Spectrometry

Results for Soil:
 Soil
 Levels

 Comp*

 Side

 Play

 Garden
     Geometric Means

Area 1     Area 2    Area 3
Area 1
 Correlation Between Soil Lead:

                       Side

 Comp                  .75
Ranges

 Area 2
                            Play

                            .22
          Garden

            .59
Area 3
3474
5163
3616
507
2632
2512
996
978
481
541
431
318
322- 18400
83-17550
258-15585
95-2705
53 - 20700
108-41200
80-34475
141-5160
151 -2915
97-7375
37-6370v
118-1102
              Dust

               .57
     Eq#           Indep. Var.           Coefficient

       1)             Ln(comp)               .0616

     Eq#                        Type

       1)        Log-Linear Least Squares
                                                Dependent Variable:

                                                     Ln(PbB)

                                               Add. Indep. Variables:

                                             House dust, child's age
                                             behavioral variables
Comments:
                                         38

-------
       The blood and environmental measurements were log transformed for statistical analysis.

       Multiple log-linear regression models were used to determine which variables were
       significantly related to blood lead.

       Stepwise regression models (backwards, MAXR) were also constructed to assess the
       independent significance of soil lead and dust lead in predicting blood lead levels.
Conclusions:
       "Soil lead contamination was associated strongly with children's PbB levels in Shoshone
       County." [Reference p. 34]

       "The results of this study support the conclusion that, in absence of significant air lead
       contamination, children who are exposed to heavily leaded soils may develop lead
       toxicity." [Reference p. 38]

       "Very little lead was found in other environmental media, indicating that the positive
       association between house dust lead contamination and children's PbB levels was likely
       to have occurred as a result of the soil lead contamination." [Reference p. 38]

       "Children who played  on non-grassy surfaces in the most heavily contaminated
       areas...had statistically significantly higher mean PbB levels than children who played on
       grassy areas."  [Reference p. 36]
                                           39

-------
40

-------
                       Abstract of Soil - Lead Studies
                                 Abstract A-13

Study Name: Midvale Community Lead Study Final Report

Study Dates: 1989             Study Location: Midvale, Utah

References:

       Bornschein, R. L., Clark, S., Pan, W., and Succop, P. (1990) "Midvale Community Lead
       Study," Final Report, University of Cincinnati.
Objectives:

       1.  "ascertain whether or not the children of Midvale... are currently exhibiting
          elevated blood lead concentrations"

       2.  "identify and quantify accessible lead and arsenic... in the environment e.g.
          lead in soil, dust, paint and water or arsenic in soil and dust"

       3.  "establish the extent of association between certain sources of environmental
          lead and blood lead"

                              [Quotes from reference p. 6]

Sampling Frame:

       Children aged 6 to 72 months living in a neighborhood of residential and
       commercial buildings parallel to the Sharon Steel Mill tailings and former smelter
       site. Blood lead samples were obtained for 291 individuals, 181 from children
       less than 6 years of age. Complete blood lead, interview, exterior and interior
       environmental data were collected from a random sample of 112 of the 249
       eligible children in the area.

Sampling Method:

       Soil cores at 2 cm depth, and exterior surface dust samples were collected. A
       composite sample of soil cores from the perimeter of the residence was collected.
       The perimeter was defined to be three feet from building wall.  A maximum of 12
       soil  core samples could be taken per composite.  Composite cores were also taken
       from garden,  bare yard  area, and play areas/sand box if such were present.  A
       composite exterior surface dust sample was collected by vacuum over paved areas
       and  other hard surfaces near the residence's entrance.

Analysis Method: Atomic Absorption  Spectrometry

                                          41

-------
Results for Soil:
 Building Perimeter
 Garden
 Play Area
 Bare Area
           N

          112
           46
           42
           88
Geometric
Mean (ppm)
341.81
294.59
77.95
313.20
Geometric Std.
Dev. (ppm)
2.54
2.65
5.52
2.60
Range
Min - Max
58 - 6665
57 - 2746
1 - 6665
24 - 2920
Correlation Between Soil Lead with Blood and Environmental Lead:
 PbSn
 PbSn
N-S, E-W = North-South, East-West grid coordinates resp.
DM = Distance to mill building;  DS = Distance to smelter stack
DMPL = Distance to mill property line; DMSPL = Min (DM, DS}
PbSp
.96
N-S
.68
PbSg
.05
E-W
-.74
PbSply
.16
DM
-.68
PbSb
.39
DS
-.60
PbDext
.77
DMP1
-.07
PbDint
.74
DMSPL
-.45
XRFext
.43


XRFint
.36


     Eq#

      1
     Indep. Var.
Ln(XRFE)
  E-W
  N-S
Housing Age
  19th C
  Post WWII

Soilfill (Y)
Ln(PbSp)

Ln(PbSmax)
Ln(XRFE)
  E-W
  N-S
Coefficient
   .179
  -.0009
   .0001

   .040
   -.567

   -.552
   .748

   .144

   .3010
  -.0009
   .0001
Dependent Variable:
Ln (Pbsperm)
                                                             Ln(PbDext)


                                                             Ln(PbB)

                                                             Ln(PbSmax)
Eq #      Type                 Add. Indep. Variables:
 1,2)  Log-Linear Reg.            XRFint, DS, DM, Age
          Backward Elimination

 2,3)  Structural Equations Analysis   House Age,SES,Age
                                         42

-------
          (two dependence relations)       Mouthing Behaviors
Comments:

       Field duplicates were collected at 10% of the sample sites.

       In the correlations all lead variables were log transformed.


Conclusions:

       "Lead based house paint and lead contaminated soil were identified as the principal
       contributors to blood lead."

       "Location of residence was found to be a strong predictor of soil lead concentrations."

       "The effect of soil lead on blood lead is both small and weak. Blood lead was found to
       increase 1.25  |ig/dl/1000 ppm increase in lead in soil. About 3.0% - 12% of the variance
       in blood lead is attributable to lead in soil."

                           [Quotes from reference pp. 127-128]

       "Environmental soil and dust lead concentrations were generally moderately
       correlated...correlations between environmental lead and blood lead concentrations were
       considerably weaker."

                               [Quote from reference p. 32]
                                           43

-------
44

-------
                       Abstract of Soil - Lead Studies
                                 Abstract A-14

Study Name: Minnesota Soil Lead Study

Study Dates: 1986-1987

Study Location: 27 counties in Minnesota, concentrating in Duluth, St. Cloud, Minneapolis, St.
                Paul, and Rochester

References:

       Schmitt, M. D. C., Trippler, D. I, Wachtler, J. N., and Lund, G. V. (1988) "Soil Lead
       Concentrations in Residential Minnesota as Measured by ICP-AES," Water, Air, and Soil
       Pollution. 39:157-168.

       Trippler, D. J., Schmitt, M. D. C., and Lund, G. V. (1989) "Soil Lead in Minnesota," In:
       Lead in Soil: Issues and Guidelines, Supplement to Volume 9 of Environmental
       Geochemistry and Health.  Edited by Davis, B.E. and Wixson, E.G., pp 273-280.

       Mielke, H. W., Adams, J. L., Reagan, P. L., and Mielke, P. W., Jr. (1989) "Soil-Dust
       Lead and Childhood Lead Exposure as a Function of City Size and Community Traffic
       Flow: The Case for Lead Abatement in Minnesota," In: Lead in Soil: Issues and
       Guidelines, Supplement to Volume 9 of Environmental Geochemistry and Health. Edited
       by Davis, B. E. and Wixson, B. G., pp 253-271.

Objectives:

       Describe "...the extent of lead contamination in the soil, the lead concentrations in the
       blood of populations at contaminated sites, the size of the population at risk from
       exposure to lead in the soil...." [Reference p. 254]

       Evaluate the quantitative relationship between soil lead and childhood blood lead as
       observed among cities in the study.

Sampling Frame:

       Cross-section (27) of all counties in Minnesota, based on socio-economic data from 1980
       census.

       Reports of > 1000 mg Pb/kg in the soil resulted in MDH protocol blood testing on a
       sample of children within 5 blocks.
                                          45

-------
       Two separate soil datasets were collected corresponding to paired (187) and unpaired
       (1266) blood samples. In all, 2454 soil samples were taken.

Sampling Method:

       Foundation samples were collected within  1.5 meters of the building.  Yard samples
       (front, back, and side) were taken at the midpoint of the yard and at least 1.5 m from a
       foundation. Street samples were collected  within 1.5m of the curb. In addition samples
       were collected in the following sites: garden, downspout, industrial, open, play area, and
       side yard.  Soils were sampled from the top 2 cm.

Analysis Method: Inductively coupled plasma-atomic emission spectroscopy.

Results for Soil:
 Minneapolis
       Foundation
       Backyard

 St. Paul
       Foundation
       Backyard
 Duluth
       Foundation
       Backyard
 Rochester
       Foundation
       Backyard

 St. Cloud
       Foundation
       Backyard

 Outstate
       Foundation
       Backyard
                              N
199
119


127
114


 32
 32


 19
 15


 13
 18


 67
 34
 Geometric
Mean (ppni)

     665
     186


     472
     119


     455
     106


       65
       23


       85
       25


     105
       18
  Geometric
Std. Dev. (ppm)

      3.5
      2.6


      4.5
      3.8


      5.2
      4.1


      8.4
      4.1


      7.5
      4.9


      9.2
      2.7
                                          46

-------
Frequency table of yard soil lead data:

                        Percent of samples
Soil Lead
(mg/kg)
< = 50
51 - 150
151 -300
301 -600
601 - 1200
>=1201
Minneapolis
6.4
26.0
36.2
21.9
8.7
0.8
St. Paul
20.5
28.5
30.0
13.5
5.0
2.5
Duluth
13.5
40.4
19.2
21.2
3.8
1.9
St. Cloud
76.2
19.0
4.8
0.0
0.0
0.0
Rochester
68.2
27.3
0.0
0.0
4.5
0.0
Comments:
       Comparisons made between yard, street-side, foundation, blood and other related tests
       were completed using Fisher's exact test (usually comparing frequencies for <=150 mg/kg
       and>=151 mg/kg)
Conclusions:
       "...soil lead concentrations in cities and the blood lead concentrations of the childhood
       population generally vary in a lock-step manner with each other." [Reference p. 257]

       "The state soil data support earlier findings that environments of the three largest cities of
       Minnesota are burdened with the highest concentrations of soil lead."  [reference p. 260]
       In fact, there is a consistent trend of soil or dust lead and blood lead that corresponds with
       city size: that is, the largest cities have the highest lead concentration and the smallest
       cities have the lowest.

       There is a strong association between blood lead and soil lead.

       "There is no discernible general pattern between the age of the dwellings within a
       community and the lead concentrations of either the soil or the blood of the childhood
       population." [Reference p. 264]

       Generally, the highest contaminated soil levels are found in inner cities with the most
       severely contaminated soils being located near the foundations of private residences. In
       addition, the soil levels tended to decrease with increasing distance from the city center.
                                            47

-------
48

-------
                      Abstract of Soil - Lead Studies
                                 Abstract A-15

Study Name: New Haven, Connecticut Lead Study

Study Dates: 1974-1977 Study Location: New Haven, Connecticut

References:

       Stark, A. D., Quah R. F., Meigs, J. W., and DeLouise, E. R. (1982) "The Relationship of
       Environmental Lead to Blood-Lead Levels in Children," Environmental Research.
       27:372-383.

       Stark, A. D., Meigs, J. W., Fitch, R. A., and Delousie, E. R. (1987) "Family Operational
       Cofactors in the Epidemiology of Childhood Lead Poisoning," Archives of Environmental
       Health. 33:222-226.

Objectives:

       1)  Determine the important environmental sources of lead in New Haven.

       2)  Determine how these sources are distributed throughout the city.

       3)  Examine the relationship between these sources and observed blood lead levels of
          children in New Haven.

Sampling Frame:

       A screening of 80%, 8289  individuals, of the population of 1 to 6 year old children
       residing in New Haven. From this screening a subset was sampled. A child was included
       in the subset if the child had lived at the same address for at least one year and if the child
       had at least two blood tests during that time both of which were <=29 |ig% or 30-39 |ig%
       or >=40 |ig%.  The final sample consisted of 377 children. Each of the 377 children had
       environmental measures taken.

Sampling Method:

       Surface scrape samples were taken close to the house and close to the street. A sample of
       5-10 g of soil was taken from the top l/2 inch of soil.

Analysis Method: Delves Cup Atomic Absorption Spectrophotometry
                                         49

-------
Results for Soil:
                              Level of Socioeconomic Status

1
Near Soil Level *233.3

**(398.1)
Far Soil Level 209.5

(79.4)
2
756.5
(125.9)
700.1
(39.8)
3 4
1327.3 830.5
(251.2) (316.2)
660.1 665.2
(63.1) (63.1)
5
703.5
(199.5)
599
(199.5)
Construction Year


NEAR
Soil
(ppm)
FAR
Soil
(ppm)
1910- 1920-
1919 1929
1200.1 * 1273.3
63.1 ** 79.4
4j *** 42
798.2 770.1
39.8 39.8
41 42
1930-
1939
1299
251.2
29
917.6
39.8
42
1940- 1950- 1960-
1949 1959 1969
444 929.6 309.7
1258.9 398.1 501.2
86 29 29
507.4 479.3 390.2
316.2 100 50.1
86 29 30
1970-
1977
131.3
50.1
O
310.9
63.1
3
* Geometric Means
** Geometric Standard Deviations
*** Sample Size
Correlation Between Soil Lead with


Far Soil
Near Soil
Near Soil Air

.3044 -.15
	 _ flfiS

Blood and
Dust

.154
.188
Environmental Lead:
Kitchen Age of
Paint Building
.192 -.2632
.202 -.2333

Exterior
Paint
.2806
.4332
                                           50

-------
                                        Coefficient

                                             .14247

                                             .14324

                                              0.0
                                              0.0
Dependent Variable

          Ln(PbB)

          Ln(PbB)

          Ln(PbB)
                                               Additional Independent Variables

                                                            None

                                                 Porch and kitchen paint levels

                                          Exterior Paint Levels, socioeconomic variables
 Eq #     Independent Variable

    1)             Ln(near soil)

    2)              Ln(far soil)

    3)       Ln(near soil), Ln(far soil)


 Eq #                Type

    1)      Simple log-linear regression

    2)      Simple log-linear regression

    3)        Multiple log regression
              model: best N variable
                    model)

Comments:
       The simple linear regression was performed to determine the proportion of variation in
       children's blood lead levels associated with each of the environmental variables.

       Multiple regression was performed in an attempt to further explain the above variation as
       a function of environmental, personal, and economic variables.

Conclusions:

       "Substantial amounts of lead were present in the soil, paint, and house dust throughout
       New Haven." [Reference p. 382]

       Based on regression results, the most important variables on blood lead levels were soil
       lead near home and street and exterior paint. (Log of near home soil lead, r=.22; log of
       near street lead, r=.197; log of exterior house paint lead,  r=.143).

       Elevated levels of lead in the proximate environment do  not explain most of the variation
       in blood lead across the population (11.7% via multiple regression).
                                           51

-------
52

-------
                      Abstract of Soil - Lead Studies
                                Abstract A-16

Study Name: New Orleans Lead Study

Study Dates: 1983     Study Location: New Orleans, Louisiana

References:

      Mielke, H. W. (1991) "Lead in Residential Soils: Background and Preliminary Results of
      New Orleans," Water, Air, and Soil Pollution. 57-58:111-119.

      Mielke, H. (1995) "Lead in New Orleans Soils: New Images of an Urban Environment,"
      Environmental Geochemistry and Health. 16(3-4): 123-128.

Objectives:

      Map community patterns of lead in New Orleans.

Sampling Frame:

      Residential neighborhoods within 283 census tracts in the New Orleans metropolitan
      area.

Sampling Method:

      Ten to fifteen samples were collected from the top 2.5 cm of the soil using a stainless
      steel garden trowel.  In each census tract, ten streetside samples (within 1 m of the street),
      2 houseside (within 1 m of the house) and 2 samples were collected from open areas such
      as vacant lots or parks. A total of 3,704 soil samples were collected.

Analysis Method:   Flame atomic absorption spectrometers and deuterium background
                   correction

Results for Soil:

          Location                N        Median     Min (ug/g)        Max
Inner-city
Foundation
Streetside

201
723

840
342

8
4

69000
9450
                                         53

-------
           Location                 N         Median      Min (ug/g)        Max
Open Area
Mid-City
Foundation
Streets! de
Open Area
Suburban
Foundation
Streets! de
Open Area
74

220
765
80

332
1195
114
212

110
110
40

50
86
28
10

1
1
2

2
2
4
10600

24400
6340
3960

5650
2150
540
       Lead (N= 60, p-value < *1Q-6)

Comments:

       * proportionate measure of having more extreme result (i.e. more extreme clustering of
       the high group) by chance alone.

       Statistical analysis was completed using an approach termed Multi-Response Permutation
       Procedures or MRPP. The 114 census tracts were sorted according to their median.  The
       top, middle, and bottom 20 tracts were used in the analysis. Additional analysis was
       performed by visually inspecting a topographical map of soil-lead levels.

Conclusions:

       There were extreme differences between inner-city and non inner-city census tracts as
       evidenced by the small p-value of the above test.

       "The most extreme lead levels are found next to foundations of homes located in the
       inner-city." [Second reference p. 124]

       "Lead in soil of vacant lots and open spaces away from streets and households in the
       inner-city is still higher that any lead levels in mid-city and suburban locations." [Second
       reference p. 124]
                                           54

-------
An exponential decrease in soil-lead levels was observed with increasing distance from
the center of the city.  The authors suggest that this may be due to an increased presence
of lead-based paint in the inner-city and a history of higher traffic congestion.
                                      55

-------
56

-------
                      Abstract of Soil - Lead Studies
                                Abstract A-17

Study Name: Honolulu Park Soil Lead and Mercury Study

Study Dates: 1972, 1987       Study Location: Honolulu, Hawaii

References:

      Fu, S., Hashimoto, H., Siegel, B. Z., and Siegel, S. M. (1989) "Variations in Plant and
      Soil Lead and Mercury Content in a Major Honolulu Park, 1972 to 1987, a Period of
      Significant Source Reduction," Water, Air, and Soil Pollution. 43:109-118.

Objectives:

      To compare lead and mercury levels from 1972 to 1987 to determine if significant
      reductions in soil lead and soil mercury levels exist. Additionally, to examine the effect
      of traffic volume on lead and mercury soil levels.

Sampling Frame:

      A 150 meter transect extending N to S for 150 meters from a grassy median along Ala
      Moana Boulevard.

Sampling Method:

      Soil samples were taken from a strip one meter in width running the length of the
      transect.  Loose debris was removed and surface soil scrapings, 5 to 6 cm, were taken.
      Each  soil sample had a net weight of over 50 grams.

Analysis Method: Flame Atomic Absorption Spectrophotometry

Results for Soil:

                       N     Arithmetic Mean            Arithmetic Std. Dev.
1972 Survey
1987 Survey
14
18
467
367
93
37
Conclusions:
                                         57

-------
The relationship between soil-lead levels and distance from automotive sources such as a
roadway was reaffirmed as soil levels fell as the distance from the roadway increased only
to increase as the beach road was reached.
                                    58

-------
                      Abstract of Soil - Lead Studies
                                 Abstract A-18

Study Name: Telluride Lead Study

Study Dates: 1986         Study Location: Telluride, Colorado

References:

      Bornschein, R. L., Clark, S., Grote, I, Peace, B., Roda, S., and Succop, P. (1988) "Soil
      Lead-Blood Lead Relationship in a Former Lead Mining Town," In: Lead in Soil: Issues
      and Guidelines, Supplement to Volume 9 of Environmental Geochemistry and Health.
      Edited by Davis, B.E. and Wixson, E.G., pp 149-160.

Objectives:

      To further examine and investigate the association between lead in soil and blood lead
      levels.

      Investigate the threat of lead poisoning due to residual lead.

Sampling Frame:

      Residents of a town containing former lead mining and milling operations. Ninety-four
      young children (<= 72 months) were sampled.  In addition, older children and young
      adults, pregnant women and occupationally exposed adults were also sampled. Total
      sample size: 258 individuals and 45 residences, 5 daycare centers or schools and
      numerous parks.

Sampling Method:

      Composite core (2.5 cm deep) samples were taken from grassy areas around the perimeter
      of the residence and along the street or sidewalk.

      Soil surface scrapings were taken from exposed soil in play areas, paths through the yard
      or playground, and  from paved areas outside entryway.

Analysis Method: Not available in Reference
                                         59

-------
Results for Soil:
 PbS,


N
45
45
Geometric
Mean
(ppm)
145
178
Geometric
Std. Dev.
(ppm)
3.2
2.5

Range
5% - 95%
17 - 804
16- 1895
Correlation Between Soil Lead with Blood and Environmental Lead:
 PbD
 XRFext
 PbScore
 Dist 1
 Dist2
PbB
-.30
NS
NS
NS
NS
NS
PbD
—
NS
.51
.46
NS
-.38
XRFext

—
.40
.49
NS
NS
PbSscri


—
.67
NS
-.35
                                                                          PbS,
                                                                  .43

                                                                  NS
PbScore
XRFext
Dist 1
Dist 2
= median surface scraping in |ig/g
= median 1 inch soil core in |ig/g
= maximum exterior XRF
= distance of house from tailing pond in feet
= distance of house from railway and creek in feet
Eq#
1)
2)
Eq#
1,2)
Indep. Var. Coefficient
Ln(PbSscrape) .40
Ln(XRFext) .288
Dist 2 .001
Type
Structural Equation analysis
                                                              Dependent Variable
                                                                    Ln(PbD)

                                                                  Ln(PbSscrape)


                                                             Add. Indep. Variables:
                                                                  Age, PbHhands
                                         60

-------
Comments:

       Environmental lead measures were log transformed.

       Only 45 children had complete environmental, social and blood lead data available.

       Structural Equations Analysis was used to investigate the lead-blood pathways and the
       associated correlations. Equations for blood, hand dust and soil lead were developed. The
       resulting equations from a system of log-linear equations.

Conclusions:

       Accessibility of the soil, particle size, chemical speciation and bioavailability of lead
       affect the risk of soil to blood lead.

       "Soil lead and PbSscrape immediately around the residence...were low and comparable to
       that seen in lead abated housing in inner-city Cincinnati." [Reference p. 152]

       "No simple, direct association was found to be significant between PbSscrape or PbScore and
       PbHhand or PbB."  [Reference p. 154] However, the exposure model of an indirect
       pathway from soil lead to blood lead was confirmed.

          PbSscrape -> PbD -> PbHhand -> PbB

       "PbScore was not associated with PbD after adjusting for the influence of PbSscrape. This
       again suggests that although PbScore is a reservoir of lead with potential for future child
       exposure, it is the available surface lead (PbSscrape) which is most associated with other
       current Pb levels." [Reference p. 155].
                                           61

-------
62

-------
                      Abstract of Soil - Lead Studies
                                 Abstract A-19

Study Name: Mt. Pleasant Soil Lead Study

Study Dates: 1990      Study Location: Mt. Pleasant, Michigan

References:

      Francek, M. A. (1992) "Soil Lead Levels in a Small Town Environment: A Case Study
      from Mt. Pleasant, Michigan," Environmental Pollution.  76:251-257'.

      Francek, M. A., Makimaa, B., Pan, V., and Hanko, J. H. (1994) "Small Town Lead
      Levels: A Case Study from the Homes of Pre-schoolers in Mt. Pleasant, Michigan,"
      Environmental Pollution. 84(2): 159-166.

Objectives:

      Examine the relationship between traffic variables, wind direction, home age, home
      condition, wall orientation and soil-lead levels.  In addition, evaluate soil-lead levels for
      schools, an abandoned dump, and a salvage yard.

      In particular, the relationship between household lead levels and home age, distance to
      road, traffic volume adjacent to the home, and the amount of exposed soil was
      investigated.

Sampling Frame:

      Roadsides, schools, and homes in Mt. Pleasant, Michigan.

Sampling Method:

      Soil samples were collected from 42 homes at 189 locations 0-5 cm in depth, using a
      stainless steel trowel.  Roadside samples (n=83) were collected from three traffic volume
      categories: low (Average Daily Traffic or ADT < 8000), moderate (800020000).  Home samples were collected from within one meter of the
      wall avoiding sites with flaking paint (n=55). School samples were collected at
      entrances, playgrounds,  and adjacent parking lots (n=27). In addition, three samples were
      taken at the city dump and three at a salvage yard. Eighteen background samples were
      also collected along a 500 m2 grid in and around city limits.
                                         63

-------
Analysis Method:  Flame Atomic Absorption Spectrophotometry




Results for Soil:

Roadside Soils
Homes
School
City Dump
Salvage Yard
Play Area Soil
Background
N
83
55
27
3
3
42
18
Arithmetic
Mean
(ppm)
320
1176
191
167
893
53
172
Arithmetic
Std. Dev.
(ppm)
125
2419
48
44
195
205
42
Median
(ppm)
280
460
200
200
940
7
200
Range
100-840
100-16839
100-260
100-200
600-1140
0-594
100-220
Roadside soils by Traffic Volumes:

Heavy
Moderate
Low
Soil-Lead Levels by
Condition of Home
Excellent
Good
Fair
Poor
N
33
14
26
Condition





Arithmetic
Mean
(ppm)
343
345
286
of Home:
N
6
18
13
18
Arithmetic
Std. Dev.
(ppm)
106
170
126

Arithmetic
Mean
(ppm)
203
347
2537
1346
Median
(ppm)
320
320
250






Range
180-540
200-840
100-620

Arithmetic
Std. Dev.
(ppm)
60
446
4631
858
                                        64

-------
 Play Area Soil Soil-Lead Levels by Age of Home:

                                        Arithmetic        Arithmetic
                                          Mean            Std. Dev.          Range
 Age of Home               N             (PPm)             (PPm)           (PPm)
 <30 years old               16               3                  3               1-9

 30-60 years                 12              28                54             0-187

 >60 years                  14              132               149            0-594


Correlation Between Soil Lead and Related Variables:

                           Soil Lead
Traffic Volume          .27 (p=.019)
Home age               .59 (p<.001)


Comments:

       One-way analysis of variance, correlation and t-tests were used to examine the data.
       Various groups such as high, moderate and low traffic volume, were compared using the
       F-test associated with the analysis of variance.  Variables were log transformed for all
       analysis.

Conclusions:

       There exists a positive relationship, though weak, between roadside soil-lead levels and
       increasing traffic.

       Strong changes in soil-lead levels are not due to prevailing wind patterns nor automobile
       idling behavior.

       There was no significant difference in the soil-lead levels found near houses in good or
       excellent condition from those found near homes in fair or poor condition.

       There is a positive relationship between home age and soil-lead levels. In addition,
       "former use of lead based paints on older homes elevated soil-lead levels." [First
       Reference p.225]

       In general soil levels were inversely related to distance from the roadway.
                                          65

-------
66

-------
                             Abstract of Soil - Lead Studies
                                     Abstract A-20

Study Name: Illinois Soil-Lead Study

Study Dates: 1985

Study Location:  City of Chicago; six suburban counties surrounding Chicago; rest of Illinois
                 ("downstate").

References:

       LaBelle, S. 1, Lindahl, P. C., Hinchman, R. R., Ruskamp, 1, and McHugh, K. (1987)
       "Pilot Study of the Relationship of Regional Road Traffic to Surface-Soil Lead Levels in
       Illinois," Published Report of the Argonne National Laboratory, ANL/ES-154.

Objectives:

       This study was conducted to evaluate soil-lead levels as a function of proximity to
       roadways and overall traffic density. In addition, questions concerning lead levels in
       relationship to traffic volume were also of concern.

Sampling Frame:

       Playgrounds, parks, schools, and day care centers in the city of Chicago and its suburbs.
       Similar areas throughout the rest of Illinois.

Sampling Method:

       Four to eight soil samples were taken at 158 locations. Each location was on the property
       of a park, playground, school, or day care and was near a "well traveled" road.  In
       addition, most soil samples were collected from bare play areas used by children under 7
       years of age. All but one of the soil  samples taken at each location were from the top 5
       cm of soil. One sub-surface, 25-30 cm, soil sample was taken at each location. Surface
       samples were taken 10 and 100 feet from the roadway edge and near play equipment.  In
       addition field replicates were collected at five locations.
                                           67

-------
Analysis Method: Atomic Absorption Spectrophotometry
 Results for Soil:
        Region
 City of Chicago
 Suburbs
 Downstate



N
256
244
167
Surface
Arithmetic
Mean
(ppm)
157
83
44
            N
            50
            61
            48
Subsurface
Arithmetic
  Mean
  (ppm)
    118
    49
    27
                     Samples other than those near play equipment
 Total Traffic Volume
 <5000
 5000 - 9999
 10000 -  19999
 2000 - 49999
 >50000
N
96
30
77
87
63
Arithmetic
(ppm)
90
141
187
265
236
Arithmetic
Std. Dev.
(ppm)
13
33
23
26
41
   Eq#       Indep. Var.       Coefficient
     1)         Ln(PbS)          -0.141
                                 -0.629
                                   .756
                                   .377
                                   .114
     2)           PbS              .055
                                 2.910
* Only for roads within .25 miles of the sampling.
         Dependent Variable:
Distance from nearest Road
*Traffic Vol. Busiest Road
*Ttl. Traffic Vol all roads
Depth cone, soil lead
Annual local traffic density
*Traffic Vol. Busiest Rd.
Annual Local traffic density
                                         68

-------
           Eq#                        Type                  Add. Indep. Variables:

            1)             Log-Log                                   None

            2)             * Physical Model (exponential)               None
Comments:
       Individual regression analysis was also completed without including depth concentration
       of lead in soil as an independent variable.

       Additional regression analyses were completed for each region separately. Also
       examined via regression was a subset of the data including only surface soil  samples
       where the surface soil-lead levels were greater than the subsurface-lead level.
Conclusions:
       While elevated soil-lead levels were found, they were generally less, 25-300 ppm, than
       soil-lead levels found near industrial sources or leaded paint.

       There is a relationship between the surface soil-lead concentration near road and the
       traffic on and distance from the road.

       The physical model was not as powerful as the log-log regression model in explaining the
       patterns of surface soil-lead levels.

       The lead levels in Chicago suburbs were similar to the levels found in Chicago itself. In
       addition, both of these levels were higher than those found in downstate Illinois.

       The patterns of soil-lead concentrations support the hypothesis that leaded-gasoline
       emissions are a source of lead in the soil.
                                            69

-------
70

-------
                      Abstract of Soil - Lead Studies
                                 Abstract A-21

Study Name: Dallas Soil Lead Contamination Study

Study Dates: 1982         Study Location: Dallas, Texas

References:

      Brown, K. W., Mullins, J. W., Richitt, E. P., Jr., Flatman, G. T., Black, S. C., and Simon,
      S. J. (1985) "Assessing Soil Lead Contamination in Dallas, Texas," Environmental
      Monitoring and Assessment. 5:137-154.

Objectives:

      "To determine and identify distribution patterns of soil lead concentrations within three
      designated study areas." [Reference p. 139]

Sampling Frame:

      A one mile radius from each of two smelters owned by RSR Corporation (RSR) and
      Dixie Metal Company (DMC) and operated inside city limits.  A reference area (REF)
      inside city limits was also sampled.

Sampling Method:

      A total of 177 sampling locations within a one mile radius of a smelter were marked for
      sampling. In addition, 80 locations were identified for sampling in a half-mile reference
      area. A total of 2795 soil core samples were collected. Soil core samples were taken at a
      depth of 7.5 cm and were two cm in diameter. Four samples were taken at each location
      and composited into a single sample. If possible,  soil  samples were collected greater than
      20 feet from painted buildings and as far as possible from any vehicle activity.

Analysis Method: Atomic  Absorption Spectroscopy

Results for Soil:

                           Maximum                                 Maximum
                             Inside                                    Outside
                            Isopleth              68% CI              Isopleth
                             (ppm)                (ppm)                (ppm)
                                         71

-------
 DMC Smelter                3000              1666-5400               300
 RSR Smelter                 2500              1389-5000               300
 Reference                     500                                        200

Comments:

       Statistical analysis was performed using the geostatistical methodology of kriging. This
       permitted evaluation of both spatial variability but also a means to identify the estimated
       variance for each point within the sampling area.

Conclusions:

       "Among a few patterns of high values both DMC and RSR have a single dominant
       pattern that includes the smelter.... The closure of the pattern implies the source is inside
       and there is a steep gradient demarking the polluted area therefore indicating that the
       smelters are probably the source of lead." [Reference p. 153]

       The soil-lead values between the two smelters were similar and both differed from the
       levels identified in the reference area.
                                           72

-------
                      Abstract of Soil - Lead Studies
                                Abstract A-22

Study Name: Aspen Garden Soil Lead Study

Study Dates: 1983            Study Location: Aspen, Colorado

References:

      Boon, D. Y. and Soltanpour, P. N. (1992) "Lead, Cadmium, and Zinc Contamination of
      Aspen Garden Soils and Vegetation," Journal of Environmental Quality. 21:82-86.

Objectives:

      "The  objective of this study, conducted in 1983, was to determine the source and extent
      of Pb, Cd, and Zn contamination of Aspen garden soils and metal concentrations in
      vegetables grown in contaminated gardens." [Reference p.82]

Sampling Frame:

      Household gardens in Aspen. In addition, areas that were visible mine dumps or areas
      suspected of being mine dumps.

Sampling Method:

      Surface soil samples, 0-15 cm depth, were collected from 65 gardens. Of the gardens
      sampled, 50 were located directly on top  of mine dump materials. Additionally, between
      5 and 10 samples were taken from each of 18 different dump locations.  These samples
      were then combined into a single composite sample for each location.

Analysis Method: TCP Emission Spectrometry

Results for Soil:



Mine Dump
Garden Soil
Background


N
65
65

Arithmetic
Mean
(mg/kg)
6375
172
10
Arithmetic
Std. Dev.
(mg/kg)
5973
155



Range
135-21700
9.2 - 808

                                        73

-------
Comments:

       Results were also presented in the article for other heavy metals and for vegetation lead
       uptake.

       Regression analysis was performed to relate soil-lead levels and vegetation uptake.

Conclusions:

       Garden soils situated on mine dump materials in Aspen are contaminated with lead.
                                           74

-------
                       Abstract of Soil - Lead Studies
                                 Abstract A-23

Study Name: El Paso, Texas Lead Study

Study Dates: 1971 - 1973          Study Location: El Paso, Texas

References:

      Landrigan, P., Gehlbach, S., Rosenblum, B., Shoults, J., Candelaria, R., Barthel, W.,
      Liddle, J., Smrek, A., Staehling, N., and Sanders, J. (1975) "Epidemic Lead Absorption
      Near an Ore Smelter: The Role of Particulate Lead," New England Journal of Medicine.
      292(3): 123-129.

      Rosenblum, B. F., Shoults, J. M., and Candelaria, R. (1976) "Lead Health Hazards from
      Smelter Emissions," Texas Medicine. 72(l):44-56.

Objectives:

      To ascertain "the prevalence and severity of lead absorption in this locale, and to evaluate
      the role here of particulate lead in lead uptake." [Reference p. 123]

Sampling Frame:

      The survey area was divided along  1970 U.S. census tract boundaries into three
      roughly concentric sections 1.6 to 2.6 km in diameter with the smelter at the
      center.

      All families in Area I were selected and a random  sample of households was sampled in
      Areas II and III. In all, 670 households participated and netted 758 individuals (1 to 19
      years old) as part of the random sample survey.

      Additional monitoring of environmental variables  occurred before and after the above
      survey (monitoring of El Paso was the original project).

      The review that follows concentrates on the random sample survey.

Sampling Method:

      Soil samples were collected at 99 sites in El Paso and at three remote sites at depths of
      2.5, 5.0 and 7.5 cm.  These sites were distinct from the homes sampled in the random
      sample survey).

      Surface soil scrapings were taken. (466 samples)

                                         75

-------
Analysis Method: Atomic Absorption Spectrophotometry
Results for Soil:
 Area I
 Area II
 Area III
 Within 200m of
 Smelter
 N
 82
184
200
 54
 Geometric
Mean (ppm)
    1791
    684
    370
    3457
  Geometric
Std. Dev. (ppm)
  Not Avail.
  Not Avail.
  Not Avail.
  Not Avail.
  Range
Not Avail.
Not Avail.
Not Avail.
560- 11450
Comments:
       Comparisons done in this study were completed using ANOVA and F-tests.
       Only trace amounts of lead (< 50 ppm) were found at the remote sites. High levels were
       found within 200 meters of the smelter, and lead content was consistently highest at the
       surface.  (From soil monitoring March 1972  - June 1973)
Conclusions:
       Soil lead levels in Areas II and III were not significantly different, but both were
       significantly lower than the soil lead level in Area I.
       "In Area I, a significant relation (p <.05) was found between soil lead content and blood
       lead levels in subjects one to 19 years old." [Reference p. 127] This relation was not
       found in Areas II or III.
       Particulate lead from soil contributed relatively less to lead uptake than did dust.
       "Soil lead, although valuable as an index of environmental  contamination, appears to be
       relatively immobile and thus less accessible for absorption." [Reference p. 128]
                                          76

-------
                      Abstract of Soil - Lead Studies
                                Abstract A-24

Study Name: Corpus Christ! Soil-Lead Study

Study Dates: 1984     Study Location: Corpus Christi, Texas

References:

      Harrison, G. (1987) "A Survey of the Lead Distribution in the Soil of Corpus Christi,
      Texas," Texas Journal of Science.  39(1): 15-22.

Objectives:

      "To present a detailed overview of lead distribution in a major city based on numerous
      closely-spaced sampling sites, and to look for anomalies or trends, or lack of."
      [Reference p. 16]

Sampling Frame:

      Parks, schools, and roadside embankments within the city limits of Corpus Christi.  There
      were 485 locations sampled, 94 in parks, 12 at schools, and the remainder in various
      locations.  The only limitation on sampling was that only vegetated non sandy soil was
      sampled.

Sampling Method:

      Soil samples were taken from the top two centimeters with a Teflon knife.

Analysis Method: Flameless Atomic Absorption Spectrophotometry

Results for Soil:
Arithmetic Mean
N (ppm)
All Samples
Parks
Schools
All others
485
94
57
379
208
55
57
250
Arithmetic
Std. Dev.(ppm)
236
66
77
250
Range
(ppm)
8
8
11
8
-2969
-318
-258
-2969
                                         77

-------
Conclusions:
       The hypothesis that leaded gasoline emissions is concentrated around roadways and is
       directly proportional to traffic volume/flow was reaffirmed.
                                           78

-------
                      Abstract of Soil - Lead Studies
                                Abstract A-25

Study Name: Maine Urban Soil Study

Study Dates: 1988         Study Location: Portland, Maine

References:

      Krueger, J. A. and Duguay, K. M. (1989) "Comparative Analysis of Lead in Maine Urban
      Soils," Bulletin of Environmental Contamination and Toxicology. 42:574-581.

Objectives:  To survey and document soil-lead levels in Portland, Maine.

Sampling Frame:

      Homes at least 30 years old, parks, and playgrounds in Portland, Maine. Homes were
      considered to be high risk locations while parks and playground were thought to be low
      risk locations. There were 100 samples taken; 75 from high risk areas, 25 from parks or
      playgrounds.

Sampling Method:

      Samples were collected randomly from the top .5 cm of soil in parks and playgrounds.
      Four samples were collected at random along a line approximately two feet from the
      foundation.

Analysis Method:  X-Ray Fluorescence, Atomic Absorption Analysis

Results for Soil:

                          N         Geometric Mean (ppm)        Range (ppm)

 High Site Risk             75                  1275                  50-10900
 Low Site Risk             25                  205                   50-700

                                Lead Level Ranges

                 % Low 0 - 499       % Moderate 500 - 999        % High >1000
                     (ppm)                  (ppm)                   (ppm)

 High Risk              50                      13                      37
 Low Risk              92                       8                        0
                                        79

-------
Conclusions:
       Soil located adjacent to the foundation of homes older than 30 years are significantly
       contaminated with lead.

       "The lead in paint chips remains relatively intact, close to the original location with little
       leaching." [Reference p. 581]
                                            80

-------
                      Abstract of Soil - Lead Studies
                                 Abstract A-26

Study Name: Beltsville Roadway Study

Study Dates: 1971 - 1977          Study Location: Beltsville, MD

References:

      Milberg, R. P., Lagerwerff, J. V., Brower, D. L., and Biersdorf, G. T. (1980) "Soil Lead
      Accumulation Alongside a Newly Constructed Roadway," Journal of Environmental
      Quality. 9:6-8.

Objectives:

      Determine yearly rates of roadside soil-lead accumulation. In addition, predict soil-lead
      accumulation based on traffic volume.

Sampling Frame:

      Sides of Interstate 1-95 near Beltsville, Maryland. "The highway runs in a north-south
      direction and consists of six lanes divided by a 30-m-wide grass median strip."
      [Reference p. 6]


Sampling Method:

      A single soil sample was collected yearly on both sides of the roadway at distances of 8,
      25, and 50 meters from the edge of the roadway.  Samples were collected from the sides
      of a pit, 30cm X 60cm, at depths of 0-5,5-10,and 10-15 cm.  Additionally, a series of four
      samples was collected in 1977 from pits dug at 5 meter increments. These additional pits
      ran parallel to the highway approximately 25 meters from the edge of the road.
                                         81

-------
Analysis Method: Atomic Absorption Spectrophotometry

Results (ppm) for Soil, 0-5 cm depth, West of 1-95:
Distance (meters)
8
25
50
Year
1971
16.8
18.4
16.8
1972
19.4
20.0
12.8
1973
49.4
21.2
21.2
1974
100
35.6
23.6
1975
120
19.2
19.6
1976
176
22.8
20.0
1977
130
40.8
22.8
Results (ppm) for Soil, 5-10 cm depth, West of 1-95:
Distance (meters)
8
25
50
Year
1971
11.2
18.4
16.4
1972
8.4
10.4
18.0
1973
21.4
14.4
15.0
1974
22.0
8.0
15.6
1975
16.7
10.8
5.2
1976
36.7
10.8
9.6
1977
28.0
21.6
20.0
Correlation Between Soil Lead Accumulation Rates with Distance from West edge of
Roadway:
Distance
(Meters)
8
25
50
Correlation
Coefficient, r
.92
.53
.19
Comments:

       Correlations analysis was performed via simple linear regression.

       Similar results were presented for the area East of 1-95.

Conclusions:

       "In general, after 1971, soil-lead levels decreased with distance from the roadway and
       with depth in the soil profile." [Reference p.7.]
                                          82

-------
The accumulation rate fell rapidly as the distance from the roadway increased.




Accumulation rate was correlated with traffic volume.
                                    83

-------
84

-------
                      Abstract of Soil - Lead Studies
                                 Abstract A-2 7

Study Name: Heavy Metal Exposure Smelter Study

Study Dates: 1978-1979

Study Location: Bartlesville, Oklahoma; Palmerton, Pennsylvania; Ajo, Arizona; Anaconda,
                Montana

References:

      Hartwell, T. D., Handy, R. W., Harris, B. S., Williams, S. R., and Gehlbach, S. H. (1983)
      "Heavy Metal Exposure in Populations Living Around Zinc and Copper Smelters,"
      Archives of Environmental Health. 38(5):284-295.

      Handy, R. W., Hariss, B. S. H., Hartwell, T. D., and Williams, S. R. (1986)
      "Epidemiologic Study Conducted in Populations Living Around Non-Ferrous Smelters,
      Vol. I," U.S. Environmental Protection Agency Report Number EPA/600/1-81/070A.

Objectives:

      Estimate the levels of arsenic, cadmium, lead, zinc, copper, and magnesium in biologic
      and environmental media in the various smelter communities. Also, examine the levels
      of these chemicals as a function of distance from a  smelter.

Sampling Frame:

      Each of the four communities were stratified into seven or eight exposure regions.
      Within each of these regions, five sampling segments (e.g. city blocks) were selected at
      random.  Finally, households within the sampling segment were selected at random and
      screened for eligible participants (individuals who had no occupational exposure and had
      lived in the study area for the preceding 12 months).

Sampling Method:  Soil samples were collected from random  areas in the yard.

Analysis Method: Flame atomic absorption spectrophotometry
                                         85

-------
Results for Soil:
Median Soil-Lead Levels by Distance from the Smelter
Bartlesville (n=38)
Distance
from Smelter
(km)
3.5-24.0
1.3-3.7
0.8-4.3
0.8-1.5
Median
(Mg/g)
34.8
243
829
821
Palmerton (n=42)
Distance
from Smelter
(km)
11.0-26.0
5.4-14.5
3.3-9.9
0.3-2.8
Median
(Mg/g)
532
117
326
331
Ajo (n=53)
Distance
from
Smelter
(km)
3.4-68.0
1.0-6.4
0.5-2.3
0.5-1.3
Median
(Mg/g)
57.8
64.5
76.5
94.8
Anaconda (n=49)
Distance
from Smelter
(km)
10.0-26.0
3.5-21.0
2.0-11.0
2.0-3.5
Median
(Mg/g)
75
115
294
424
Correlation Between Distance from the Smelter and Soil-Lead Levels

Bartlesville
Palmerton
Ajo
Anaconda
Correlation*
-0.40
0.13
-0.13
-0.19
Sample Size
165
65
90
74
Comments:

* A negative correlation indicates that metal levels decrease as distance from the smelter
increases. The tests for significance were calculated using Spearman's correlation coefficient.

Tests of significance between the four sampling strata at each site were conducted using the
Kruskal-Wallis test statistic.

Conclusions:

There was a "...general trend toward increasing levels of environmental metal burdens with
proximity to the smelter." [First reference p. 287]
                                           86

-------
                      Abstract of Soil - Lead Studies
                                Abstract A-28
Study Name: Survey of Lead Levels Along Interstate 880

Study Dates: 1993

Study Location: Alameda County, California

References:

       Teichman, I, Coltrin, D., Prouty, K., and Bir, W. A. (1993) "A Survey of Lead
       Contamination in Soil Along Interstate 880, Alameda County, California," Journal of the
       American Industrial Hygiene Association. 54(9):557-559.

Objectives:

       "Determine the levels of lead in soils taken from yards in close proximity to a major
       freeway." [reference p. 557]

Sampling Frame:

       Private homes, parks and playgrounds (21 samples), and public housing developments in
       Alameda County along Interstate 880 or within a 1-mile radius.  There were 138 samples
       collected from residences (116 east of the interstate, 22 west).

Sampling Method:

       Both surface (top 0.5 to 0.75 inches) and subsurface samples were collected.  The
       subsurface samples were collected from 3 to 8 inches in depth. In all, approximately 200
       samples were collected.

Analysis Method: Flame atomic absorption spectroscopy

Results for Soil:

Comparison of Subsurface and Surface Soil-Lead Levels (n=19)
Surface (ppm)
Mean
567.7
Min
195.3
Max
2026.6
Subsurface (ppm)
Mean
618.3
Min
369.8
Max
1405.7
                                         87

-------
Comparison of Downwind (East) and Upwind (West) Surface Soil-Lead Levels
East(ppm)n=116
Mean
594.3
Min
22.3
Max
3187.4
West (ppm) n=22
Mean
263.3
Min
89.7
Max
862.0
Conclusions:
       The subsurface soil-lead levels are higher than the surface lead levels.
       The soil-lead levels downwind of the interstate were higher than those found on the
       upwind side of the interstate.

-------
                       Abstract of Soil - Lead Studies
                                 Abstract A-29

Study Name: Albuquerque Street Dirt Lead Study

Study Dates: 1981

Study Location: Albuquerque, New Mexico

References:

       Franz, D. A. and Hadley, W. M. (1981) "Lead in Albuquerque Street Dirt and the Effect
       of Curb Paint," Bulletin of Environmental Contamination and Toxicology. 27(3):353-
       358.

Objectives:

       The purpose of this study is to investigate the source of lead in street-side soil samples.
       In particular, to determine the possible contribution of curb paint.

Sampling Frame:

       The Albuquerque urban area. Samples were collected from residential and school yards
       (16) and in three areas adjacent to city streets: residence side of walk (7), curbs and
       gutters (13), and central medians (7). Additional samples were collected from the
       medians of three city streets to examine the effects of curb paint particles.

Sampling Method:

       All samples were collected using a plastic scoop and brush. For the median, samples were
       taken in close proximity to a painted curb and at least 30 meters from any painted surface.
       Residential and school yard samples were collected at least 30 meters from the street.

Analysis Method: Flame atomic absorption spectrophotometry
                                          89

-------
Results for Soil:
 Description

 Residential and school Yards

 City Streets, residence side of walks*

 City Streets, curbs and gutters*

 City Streets, central medians*

Relationship Between Mean Soil-Lead Levels and Proximity to Painted Curb
Lead




Levels (|ig/g)
3-110
270-1640
950-3540
980-5280
Sample Size
16
7
13
7
Site**
A
B
C
Close (<2 meters from Painted
curb)
Mean (ppm)
1770
3720
4860
SD
240
820
430
Distant (> 30 meters from painted
curb)
Mean (ppm)
1120
2600
3660
SD
190
180
220
Comments:

       * Average traffic volume > 14,000 vehicles/day

       ** Average traffic volume 23,200 - 30,000 vehicles/day, sample sizes unavailable in
       reference.

       Significance tests (95% confidence) were made using the Student's t-test.

Conclusions:

       In all three of the median sites, the mean lead levels for the close samples were
       significantly higher than those of the distant samples collected from the same median.

       "The erosion of leaded roadway paint constitutes a hitherto unrecognized contribution to
       high lead levels in urban street dirt in Albuquerque, and possibly in other cities."
       [reference p.357] However, the authors conclude that "automotive emissions constitute
       the most significant source of lead." [reference p.3 57]
                                            90

-------
"The highest lead concentrations were found in dirt on and adjacent to busy streets, while
near-background levels were found in soils from residential streets and schoolyards."
[reference p.354]

"At any given street collection site, lead levels were lowest in the soil on the residence
side  of the sidewalk, higher in the dirt from curb and gutter, and highest in the soil from a
median strip." [reference p.354]
                                     91

-------
92

-------
                       Abstract of Soil - Lead Studies
                                 Abstract A-3O

Study Name: Identification of Lead Sources through Stable Isotope Ratio Techniques: Case
             Studies

Study Dates: 1978-1979

Study Location:  Oakland, California

References:

       Yaffe, Y., Flessel, C. P., Wesolowski, J. J., del Rosario, A., Guiruis, G., Matias, V.,
       Gramlich, J. W., Kelly, W. R., DeGarmo, T. E., and Coleman, G. C. (1983)
       "Identification of Lead Sources in California Children Using the Stable Isotope Ratio
       Technique," Archives of Environmental Health.  38(4):237-245.

Objectives:

       The objective of this  study was to examine the feasibility of using the isotopic ratio
       method to identify sources of lead in children.

Sampling Frame:

       Two different case control studies were conducted. In both cases, the children resided in
       Alameda county. For the first case, the sources of lead for 8 children from an extended
       family living in the same house were investigated.  For the second case study, sources of
       lead for twins of a previously lead-burdened mother were examined.

Sampling Method:

       In the first case study, four surface soil samples (<1 inch in depth) were collected around
       the home (2 backyard and curbside, 2 alongside house) and near the street curb.

       For the second case study, eight surface soil samples were collected from the front  (4),
       side (1) and backyard (2). One sample was collected from the backyard of a neighboring
       yard where the twins frequently played.

Analysis Method: Flame Atomic Absorption Spectrophotometry
                                          93

-------
Results for Soil:
Description
No. of
Samples
Average Pb
Ratio
Median Pb
Conc.(ppm)
Range (ppm)
Case I:
Backyard and Curbside
Alongside house
2
2
2.084
2.068
1160
1300
1050-1260
1220-1370
Case II:
Front Yard
Sideyard
Backyard
Neighbor's Backyard
4
1
2
1
2.126
2.085
2.080
2.102
2430
1420
1100
990
480-7130
*
940-1300
*
Comments:
       As only a very limited number of samples were collected the results of this study should
       be interpreted with caution.
Conclusions:
       The authors concluded that in both cases, the results of the isotopic ratio analysis suggests
       that the lead in the backyard soil samples derived primarily from the weathering of nearby
       exterior paint.
                                            94

-------
                       Abstract of Soil - Lead Studies
                                 Abstract A-31

Study Name: California Lead Study: Three High-Risk Communities

Study Dates: 1987-1991

Study Location: Alameda, Sacramento, Los Angeles Counties

References:

       Sutton, P., Athanasoulis, M., Flessel, P., Guirguis, G., Haan, M., Schlag, R., and
       Goldman, L. (1995) "Lead Levels in the Household Environment of Children in Three
       High-Risk Communities in California," Environmental Research.  68:45-57.

Objectives:

       Examine the levels of lead in these three high-risk communities. In addition, determine
       how well environmental lead levels are predicted by age of housing.

Sampling Frame:

       Within Alameda, Sacramento, and Los Angeles counties, communities with a high risk
       for childhood lead poisoning were selected for sampling based upon: the number of
       children between I and 6 years, older housing, and populations having cultural  and/or
       ethnic risk factors. Census blocks were used to determine the number of children
       between I-6 years living in a particular community. Any households in the selected
       census tracts with a child between 1-6 years were eligible for the study. In Los Angeles
       county, a random sample of blocks within two census tracts were selected and households
       with a child between I and 6 years were eligible. In all, 933 households (358 in Oakland,
       343 in Los Angeles, 232 in Sacramento) were surveyed.

Sampling Method:

       Surface soil samples (
-------
Analysis Method: Flame Atomic Absorption Spectrophotometry




Results for Soil:
Oakland:
Front-yard
Rear-yard
Side-yard
Los Angeles:
Front-yard
Rear-yard
Side-yard
Sacramento:
Front-yard
Rear-yard
Side-yard
No. Houses
Sampled

231
141
147

290
236
245

221
197
198
Geometric
Mean (ppm)

716
889
942

181
215
203

225
217
290
+- 1 Std.
Deviation

313-1639
377-2096
339-2221

88-372
102-454
88-467

90-563
92-513
99-846
Range (ppm)

56-5827
78-7175
57-6985

17-1481
31-8269
28-4554

17-3795
32-7770
11-12094
                                         96

-------
Logistic Regression Results*
Independent Variable
Reported exterior paint change in last year
Exterior paint >= 5000 ppm
Home built before 1920 compared to post 1950
Home built 1920-1950 compared to post 1950
Home located in Oakland compared to Los
Angeles
Home located in Sacramento compared to Los
Angeles
Adjusted Odds
Ratio
1.77
2.11
10.43
2.07
46.39
3.90
95% Confidence
Interval
0.73-4.28
1.03-4.33
3.11-35.03
0.64-6.65
15.90-135.33
1.42-10.66
       Dependant variable was soil concentration categorized into above or below 500 ppm
Comments:
       All analysis was completed using log-transformed values for the environmental samples.

       Multiple logistic regression analysis was used to examine the relationship between soil
       lead concentration and related variables.  T-tests, chi-square analysis, correlations, and
       analysis of variance were also used to analyze soil and the other environmental samples.
Conclusions:
       "Age of housing is highly predictive of the presence of environmental lead-levels and
       paint deterioration. In this survey age of housing was the best measured predictor of lead
       in soil and dust" [reference p.56] However, the authors also state that "for soil lead levels,
       construction year is probably a measure of historical paint lead levels, the deterioration of
       paint, and the deposition of atmospheric lead in soil over time." [reference p.55]

       "Soil lead levels measured next to homes (i.e., in the side yards) in Oakland and
       Sacramento were significantly higher that the front and rear yards, presumably due to lead
       paint and rain run-off." [reference p.55]
                                            97

-------
98

-------
                             Abstract of Soil - Lead Studies
                                    Abstract A-32

Study Name: Champaign-Urbana Lead Study

Study Dates: 1976

Study Location: Champaign-Urbana, Illinois

References:

       Solomon, R. L. and Hartford, J. W. (1976) "Lead and Cadmium in Dusts and Soils in a
       Small Urban Community," Enviromental Science and Technology.  10(8):773-777.

Objectives:

       Survey the lead levels on settled dusts and soils in a small urban community. In addition,
       it was desired to examine the lead levels in areas where the homes were in good
       condition, painted with non-lead or low-lead paints, and in low traffic areas.

Sampling Frame:

       Residential and nonresidential homes in the Champaign-Urbana community. In all, 288
       soil  samples were collected in residential homes that were in good condition, painted with
       non-lead or low-lead paints, and in low traffic areas. Additionally, soil samples were
       collected from 7 homes coated with leaded paint, and located in high traffic areas.
       Finally, 183 soil samples were collected from 20 brick or stone buildings designated as
       nonresidential.

Sampling Method:

       The soil samples were collected using a standard soil boring tool. The top 2.5 cm were
       extracted from the  15-cm core and analyzed for lead.

       Exterior dust samples were collected with a specially modified vacuum using a 0.5 x 0.5
       meter template.

Analysis Method: Atomic absorption spectrometry
                                          99

-------
Results for Soil:
                Description*

 Side Lawn Near (<1 meter) House

 Rear Lawn Near (<1 meter) House

 Far Front Lawn

 Far Lawn
Median (ppm)

      50

      100

      70

      40
Comments:
       * Results for 10 homes in good condition, painted with non-lead or low-lead paints, and
       in low traffic areas. A total of 288 samples were collected, however the exact number of
       samples at each location is unavailable in the reference.
Conclusions:
       Lead-based paint alone is not an adequate indicator of the presence of high soil lead
       concentration.

       "Soil lead is high near the road and falls off but then increases to the highest level
       adjacent to the stone buildings. The latter increase is possibly due to leaching of window
       trim paint, but more likely is due to wash-off of non-paint source  settled lead from roofs
       or ledges." [reference p.776]

       The soil lead levels in the brick and stone non-residential buildings were "...higher than
       that found adjacent to the frame houses painted with high lead paint. This strongly
       suggests that a wash-off of settled lead of airborne origin occurs." [reference p.776]. In
       particular the soil-lead levels are "...several times higher that normal residential values
       due to higher vehicular traffic  in the area." [reference p.776]
                                           100

-------
                      Abstract of Soil - Lead Studies
                                Abstract A-33

Study Name: Cincinnati Roadside Soil Study

Study Dates: 1990

Study Location: Cincinnati, Ohio

References:

       long, S.  T. (1990) "Roadside Dusts and Soils Contamination in Cincinnati, Ohio,"
       Environmental Management. 14(1):107-114.

Objectives:

       Examine the current levels of lead in roadside soil. In addition, to investigate the
       relationship between lead-levels in the roadside soil, housing age, and traffic volume.

Sampling Frame:

       The sampling in this study was performed in selected areas of The Greater Cincinnati
       Metropolitan District. Sampling areas were identified by road system (defined by average
       daily  traffic volume, ADT) and neighborhood (defined by age of housing). In all,
       samples were collected from three road systems: highways (ADT >20,000), boulevards
       (8000
-------
Results for Soil:

Average Daily Traffic
Volume
>20,000
8,000 - 20,000
<8,000
Housing Age
Pre-1950
Post- 1960
0-5 cm Soil Sample (n=60)
Mean (ppm)

1125.7
999.7
886.9
1256.2
752.0
SD

1282.8
1043.5
623.5
1254.3
557.4
15-20 cm Soil Sample (n=60)
Mean (ppm)

1318.0
1045.2
1540.0
1602.4
999.7
SD

1311.9
957.1
1423.8
1563.8
744.7
Comments:
       Streetside dust was also collected using a brush and dustpan.  Dust samples were also
       collected 30 meters from the roadway.

       One-way analysis of variance was used to investigate relationships between soil-lead
       levels and traffic flow, housing age, rainfall levels, and other factors.
Conclusions:
       "The amount of lead in the surficial soil was lower than the underlying horizons."
       [reference p. 109]

       "The mean values of lead in the topsoil, wet and dry dust samples at the curb, and dry
       dust samples 30 meters from the curb, were higher in places with higher vehicular
       traffic." [reference p. 109] However, a statistical difference was not observed.

       Older neighborhoods had higher levels of lead.
                                           102

-------
                       Abstract of Soil - Lead Studies
                                 Abstract A-34

Study Name: Granite City Lead Exposure Study

Study Dates: August-September 1991

Study Location: Granite City, Illinois

References:

      Kimbrough, R., Levois, M., and Webb, D. (1995) "Survey of Lead Exposure Around a
      Closed Lead Smelter," Pediatrics. 95(4):550-554.

      Kimbrough, R. D., LeVois, M., and Webb, D. R. (1994) "Management of Children with
      Slightly Elevated Blood Levels," Pediatrics.  93(2):188-191.

Objectives:

      Test the hypothesis that elevated soil-lead is related to blood-lead levels in children living
      in the vicinity of a closed smelter.

Sampling Frame:

      A population census was conducted in an area extending 4 km from the smelter to
      identify households with children under 6 years of age. All households with children
      under 6 were asked to participate in the study. Environmental samples were collected
      from four concentric circular regions centered by the smelter. Each region was
      approximately 1 km wide.

Sampling Method:

      Two soil (core, 1 inch in depth) samples were collected from 10 play area locations and
      composited into a single sample for each location.  Unless the play area was near the
      building, an effort was made to avoid sampling from the building dripline.

Analysis Method:   EPA Method 6010 Using Inductively  Coupled Argon Plasma Emission
                    Spectroscopy
                                         103

-------
Results:

                       N         A. Mean        Min          Max           S.D

 Soil(ppm)           338           449            37           3010           420

Variables Positively Correlated With Soil Lead (P-value <.01):

       Dust lead loading
       Indoor lead paint
       Age of home
       Poor rating of the "Condition of the house"

Comments:

       Data analysis was performed through two-tailed t-tests and chi-square analyses for
       categorical variables.

       Older  homes were generally closer to the smelter than newer homes which could be a
       confounding factor  in the relationship between distance from the smelter and soil lead
       levels.

Conclusions:

       "Condition of the house, lead in paint, lead in dust, lead in soil, smoking of the parents,
       proximity to the closed smelter, education and income of the parents, and behavioral
       factors for the children predicted PbB in young children." [reference p553.]

       Many  of the environmental factors were related to each other.  In particular, soil lead
       levels  were related to  age and condition of the home.
                                          104

-------
                            Abstract of Soil - Lead Studies
                                    Abstract A-35

Study Name: Rochester Side-by-Side Dust Collection Study

Study Dates: March 1991 - September 1992

Study Location: Rochester, New York

References:

       Lanphear, B. P., Emond, M., Jacobs, D. E., Weitzman, M., Tanner, M., Winter, N. L.,
       Yakir, B., and Eberly, S. (1994) "A  Side-By-Side Comparison of Dust Collection
       Methods for Sampling Lead-Contaminated House Dust," Environmental Research.
       68(2): 114-123.

       Department of Pediatrics, Biostatistics, and Environmental Medicine, The University of
       Rochester School of Medicine, New York, and The National Center for Lead-Safe
       Housing, Columbia, Maryland. (1995) "The Relation of Lead-Contaminated House Dust
       and Blood Lead Levels among Urban Children," Final Report, Volume II, #MLDP
       TOGO 1-93.

Objectives:

       There were three objectives of this study: 1) to determine if dust-lead loading were a
       better predictor of blood-lead than dust lead-concentrations; 2) to compare dust sampling
       techniques for efficiency and performance; 3) to identify interior surfaces for routine
       sampling.

Sampling Frame:

       Children aged 12 to 30 months who had resided in the same residence since 6 months of
       age.  In addition, the children had to live in the city of Rochester and spend only a limited
       time away from their house. Children were identified for the study from randomized
       hospital records.

Sampling Method:

       A composite soil sample (of three samples) was collected from core (top !/2 inch) samples
       taken on each side of the child's  house. These samples were collected from bare soil
       areas around the perimeter of the foundation. In addition, composite soil samples were
       collected from bare soil play areas.

Analysis Method:    Flame atomic absorption spectroscopy

                                         105

-------
Results for Soil:
Type of Sample
Foundation Coarse Soil
Foundation Fine Soil
Play Area Coarse Soil
Play Area Fine Soil
Geometric Mean
n (ug/g)
182
182
82
82
981
732
299
271
± 2 Standard
Deviations
52-18565
54-9994
30-2961
35-2104
Correlation Between Soil and Paint Samples
 Sample Type
                                              Correlation with Foundation Coarse Soil
 Foundation Fine Soil

 Exterior Paint
.84

.37
Comments:
       Fine and coarse soil samples were created in the laboratory by sieving the soil samples
       through a 2 mm mesh sieve followed by a 250 |im mesh sieve.
Conclusions:
       "...foundation soil lead appeared to be higher than play area soil lead and the coarse-
       sieved foundation soil fraction appeared to have a higher lead concentration than the fine-
       sieved soil fraction." [reference  p. 118]
                                          106

-------
                      Abstract of Soil - Lead Studies
                                Abstract A-36

Study Name: Washington, D.C. Soil Lead Study

Study Dates: Unavailable in Reference

Study Location: Washington, D.C.

References:

      Elhelu, M. A., Caldwell, D., and Hirpassa, W. (1995) "Lead in Inner-City Soil and Its
      Possible Contribution to Children's Blood Lead," Archives of Environmental Health.
      50(2): 165-169.

Objectives:

      The objective of this study was to assess the source of lead in inner-city soils in
      Washington, D.C.

Sampling Frame:

      Soil samples were randomly collected from 239 homes throughout the eight wards of
      Washington, D.C.


Sampling Method:

      Soil samples were collected from unpaved front yards approximately 1 meter from the
      dwelling. Each soil sample was collected at a depth of 15 cm. On average, the dwellings
      were 4.5 m from the road.
Analysis Method:   perkin Elmer 2100 Atomic Absorption Spectrophotometer

Results for Soil:
Ward
%-tile
Max
75
Median
25
Min
n(# of Sites)
1
4905
1145
444.2
228.
36.4
30
2
4520
975
471.4
344.8
48.3
30
3
815
105.7
53.7
25.1
10.2
30
4
4575
294.9
198.9
95.5
32.7
30
5
5056
380.4
221.9
101.3
12
30
6
1720
427.9
260.4
125
13.8
30
7
3740
274.9
144.4
70.3
36.2
30
8
6015
307.9
129.7
68.45
22.2
30
      Extracted from Table 1. Reference p. 167
                                        107

-------
Comments:
          Only 24% of the houses in Ward 3 had a painted exterior.  Approximately 74% of the
          houses in the other seven wards had painted exteriors.
Conclusions:
       The authors conclude that paint, rather than vehicular emission, was the main source of
       lead in Washington, D.C. soils.

       "Concentrations of lead in soil may be high near a road, but may be highest in areas
       adjacent to buildings."  [reference 166]
                                           108

-------