Environmental Geochemistry and Health

, Volume 35, Issue 4, pp 495–510 | Cite as

A comparison of three empirically based, spatially explicit predictive models of residential soil Pb concentrations in Baltimore, Maryland, USA: understanding the variability within cities

  • Kirsten SchwarzEmail author
  • Kathleen C. Weathers
  • Steward T. A. Pickett
  • Richard G. LathropJr.
  • Richard V. Pouyat
  • Mary L. Cadenasso
Original Paper


In many older US cities, lead (Pb) contamination of residential soil is widespread; however, contamination is not uniform. Empirically based, spatially explicit models can assist city agencies in addressing this important public health concern by identifying areas predicted to exceed public health targets for soil Pb contamination. Sampling of 61 residential properties in Baltimore City using field portable X-ray fluorescence revealed that 53 % had soil Pb that exceeded the USEPA reportable limit of 400 ppm. These data were used as the input to three different spatially explicit models: a traditional general linear model (GLM), and two machine learning techniques: classification and regression trees (CART) and Random Forests (RF). The GLM revealed that housing age, distance to road, distance to building, and the interactions between variables explained 38 % of the variation in the data. The CART model confirmed the importance of these variables, with housing age, distance to building, and distance to major road networks determining the terminal nodes of the CART model. Using the same three predictor variables, the RF model explained 42 % of the variation in the data. The overall accuracy, which is a measure of agreement between the model and an independent dataset, was 90 % for the GLM, 83 % for the CART model, and 72 % for the RF model. A range of spatially explicit models that can be adapted to changing soil Pb guidelines allows managers to select the most appropriate model based on public health targets.


Soil Pb Spatial modeling Classification and regression trees Random Forest Urban 



We are especially grateful to the homeowners for access to their property and to the University of California, Davis for use of the XRF. The building footprint dataset was used with permission under a license agreement with Baltimore City. Our understanding of GIS and spatial statistics greatly benefited from conversations with John Bognar, Dr. Adele Cutler, Amanda Elliot Lindsey, Dr. Elizabeth Freeman, Scott Haag, David Lewis, Dr. Zewei Maio, Dr. Samuel Simkin, Jim Trimble, and Dr. Weiqi Zhou. This work is a contribution to the long-term ecological research program (a program of the National Science Foundation) and the Cary Institute of Ecosystem Studies and was supported by NSF grants DEB 042376 and 0808418.


  1. Anderson, J. R., Hardy, E. E., Roach, J. T., & Witmer, R. E. (1976). Land use and land cover classification systems for use with remote sensor data. Washington, DC: US Geological Service. Professional Paper 964.Google Scholar
  2. Aschengrau, A., Beiser, A., Bellinger, D., Copenhafer, D., & Weitzman, M. (1994). The impact of soil Pb abatement on urban children’s blood Pb levels: Phase II results from the Boston Pb-in-soil demonstration project. Environmental Research, 67, 125–148.CrossRefGoogle Scholar
  3. Bellinger, D., & Dietrich, K. N. (1994). Low-level lead exposure and cognitive function in children. Pediatric Annals, 23, 600–605.Google Scholar
  4. Breiman, L. (2001). Random forests. Machine learning, 45(1), 5–32.CrossRefGoogle Scholar
  5. Breiman, L., Friedman, J., Stone, C. J., & Olshen, R. A. (1984). Classification and regression trees. London: Chapman and Hall.Google Scholar
  6. Carr, R., Zhang, C., Moles, N., & Harder, M. (2008). Identification and mapping of heavy metal pollution in soils of a sports ground in Galway City, Ireland, using a portable XRF analyser and GIS. Environmental Geochemistry and Health, 30(1), 45–52.CrossRefGoogle Scholar
  7. Cattle, J. A., McBratney, A. B., & Minasny, B. (2002). Kriging method evaluation for assessing the spatial distribution of urban soil Pb contamination. Journal of Environmental Quality, 31, 1576–1588.CrossRefGoogle Scholar
  8. Centers for Disease Control (CDC). (2012). CDC response to advisory committee on childhood lead poisoning prevention recommendations in “Low level lead exposure harms children: A renewed call of primary prevention.” Atlanta, GA. US Department of Health and Human Services. Available at pdf. Accessed November 1, 2012.
  9. Chaney, R. L., Sterrett, S. B., & Mielke, H. W. (1984). The potential for heavy metal exposure from urban gardens and soil. In J. R. Preer (Ed.), Proceedings of the symposium on heavy metals in urban gardens. Univ. Dist. Columbia Extension Service, Washington, DC.Google Scholar
  10. Chirenje, T., Ma, L. Q., Reeves, M., & Szulczewski, M. (2004). Pb distribution in near-surface soils of Florida cities: Gainesville and Miami. Geoderma, 119, 113–120.CrossRefGoogle Scholar
  11. Cutler, D. R., Edwards, T. C., Jr, Beard, K. H., Cutler, A., Hess, K. T., Gibson, J., et al. (2007). Random forests for classification in ecology. Ecology, 88(11), 2783–2792.CrossRefGoogle Scholar
  12. De’ath, G., & Fabricius, K. E. (2000). Classification and regression trees: A powerful yet simple technique for ecological data analysis. Ecology, 81(11), 3178–3192.CrossRefGoogle Scholar
  13. Dietrich, K. N., Berger, O. G., Succop, P. A., Hammond, P. B., & Bornschein, R. L. (1993). The developmental consequences of low to moderate prenatal and postnatal lead exposure: Intellectual attainment in the Cincinnati Lead Study cohort following school entry. Neurotoxicology Teratology, 15, 37–44.CrossRefGoogle Scholar
  14. Duggan, M. J., & Inskip, M. J. (1985). Childhood exposure to Pb in surface dust and soil: A community health problem. Public Health Review, 13, 1–54.Google Scholar
  15. Facchinelli, A., Sacchi, E., & Mallen, L. (2001). Multivariate statistical and GIS-based approach to identify heavy metal sources in soil. Environmental Pollution, 114, 313–324.CrossRefGoogle Scholar
  16. Freeman, E., & Frescino, T. (2009). ModelMap. Methods in R, 18.Google Scholar
  17. Gahegan, M. (2000). The case for inductive and visual techniques in the analysis of spatial data. Journal of Geographical Systems, 2(1), 77–83.CrossRefGoogle Scholar
  18. Gahegan, M. (2003). Is inductive machine learning just another wild goose or might it lay the golden egg? International Journal of Geographical Information Science, 17(1), 69–92.CrossRefGoogle Scholar
  19. Golubiewski, N. (2006). Urbanization increases grassland carbon pools: Effects of landscaping in Colorado’s front range. Ecological Applications, 16(2), 555–571.CrossRefGoogle Scholar
  20. Griffith, D. A., Johnson, D. L., & Hunt, A. (2009). The geographic distribution of metals in urban soils: The case of Syracuse, NY. GeoJournal, 74(4), 275–291.Google Scholar
  21. Imperato, M., Adamo, P., Naimo, D., Arienzo, M., Stanzione, D., & Violante, P. (2003). Spatial distribution of heavy metals in urban soils of Naples city (Italy). Environmental Pollution, 124(2), 247–256.CrossRefGoogle Scholar
  22. JMP 7.0.2. (2007). SAS Institute Inc.Google Scholar
  23. Koller, K., Brown, T., Spurgeon, A., & Levy, L. (2004). Recent developments in low-level Pb exposure and intellectual impairment in children. Environmental Health Perspectives, 112(9), 987–994.CrossRefGoogle Scholar
  24. Lanphear, B. P., Dietrich, K., Auinger, P., & Cox, C. (2000). Cognitive deficits associated with blood Pb concentrations < 10 mg/dL in US children and adolescents. Public Health Report, 115, 521–529.CrossRefGoogle Scholar
  25. Li, X., Lee, S., Wong, S., Shi, W., & Thornton, I. (2004). The study of metal contamination in urban soils of Hong Kong using a GIS-based approach. Environmental Pollution, 129(1), 113–124.CrossRefGoogle Scholar
  26. Liaw, A., & Wiener, M. (2009). Classification and regression by random forest. Resampling Methods in R, The boot Package,18.Google Scholar
  27. Machemer, S., & Hosick, T. (2004). Determination of soil Pb variability in residential soil for remediation decision making. Water, Air, and Soil Pollution, 151(1), 305–322.CrossRefGoogle Scholar
  28. Markus, J., & McBratney, A. B. (2001). A review of the contamination of soil with Pb II. Spatial distribution and risk assessment of soil Pb. Environment International, 27, 399–411.CrossRefGoogle Scholar
  29. Maryland Department of the Environment. (2007). Maryland plan to eliminate childhood Pb poisoning by 2010. Baltimore, 110.Google Scholar
  30. Mielke, H. W. (1999). Pb in the inner cities. American Scientist, 87(1), 62–73.Google Scholar
  31. Mielke, H. W., Anderson, J. C., Berry, K. J., Mielke, P. W., Chaney, R. L., & 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.CrossRefGoogle Scholar
  32. Mielke, H. W., Dugas, D., Mielke, P. W., Jr, Smith, K. S., & Gonzales, C. R. (1997). Associations between soil Pb and childhood blood Pb in urban New Orleans and rural Lafourche Parish, Louisiana. Environmental Health Perspectives, 105(9), 950–954.CrossRefGoogle Scholar
  33. Mielke, H. W., Laidlaw, M. A. S., & Gonzales, C. R. (2011). Characterization of lead (Pb) from traffic in 90 U.S.A. urbanized areas: Review of urban lead dust and health. Environment International, 37, 248–257.CrossRefGoogle Scholar
  34. Mielke, H. W., & Reagan, P. L. (1998). Soil is an important pathway of human lead exposure. Environmental Health Perspectives, 106(1), 217–229.Google Scholar
  35. Mielke, H. W., Wang, G., Gonzales, C. R., Powell, E. T., Le, B., & Quach, V. N. (2004). PAHs and metals in the soils of inner-city and suburban New Orleans, Louisiana, USA. Environmental Toxicology and Pharmacology, 18(3), 243–247.CrossRefGoogle Scholar
  36. Miranda, M., & Dolinoy, D. (2005). Using GIS-based approaches to support research on neurotoxicants and other children’s environmental health threats. Neurotoxicology, 26(2), 223–228.CrossRefGoogle Scholar
  37. Miranda, L. M., Dolinoy, D. C., & Overstreet, M. A. (2002). Mapping for prevention: GIS models for directing childhood Pb poisoning prevention programs. Environmental Health Perspectives, 110(9), 947–953.CrossRefGoogle Scholar
  38. Motto, H. L., Daines, R. H., Chilko, D. M., & Motto, C. K. (1970). Pb in soils and plants: Its relationship to traffic volume and proximity to highways. Environmental Science and Technology, 4(3), 231–237.CrossRefGoogle Scholar
  39. Needleman, H. L., & Gatsonis, C. A. (1990). Low-level lead exposure and the IQ of children. Journal of the American Medical Association, 263, 673–678.CrossRefGoogle Scholar
  40. Nigg, J. T., Knottnerus, G. M., Martel, M. M., Nikolas, M., Cavanagh, K., Karmaus, W., et al. (2008). Low blood lead levels associated with clinically diagnosed attention-deficit/hyperactivity disorder and mediated by weak cognitive control. Biological Psychiatry, 63(3), 325–331.CrossRefGoogle Scholar
  41. Ordóñez, A., Loredo, J., DeMiguel, E., & Charlesworth, S. (2003). Distribution of heavy metals in the street dusts and soils of an industrial city in Northern Spain. Archives of Environmental Contamination and Toxicology, 44(2), 160–170.CrossRefGoogle Scholar
  42. Perry, J. N., Liebhold, A. M., Rosenberg, M. S., Dungan, J., Miriti, M., Jakomulska, A., et al. (2002). Illustrations and guidelines for selecting statistical methods for quantifying spatial pattern in ecological data. Ecography, 25(5), 578–600.CrossRefGoogle Scholar
  43. Peters, J., De Baets, B., Verhoest, N. E. C., Samson, R., Degroeve, S., De Becker, P., et al. (2007). Random forests as a tool for ecohydrological distribution modelling. Ecological Modelling, 207(2–4), 304–318.CrossRefGoogle Scholar
  44. Pouyat, R. V., Belt, K., Pataki, D., Groffman, P. M., Hom, J., & Band, L. (2007a). Urban land-use change effects on biogeochemical cycles. In J. G. Canadell, D. E. Pataki & L. F. Pitelka (Eds.), Terrestrial ecosystems in a changing world (pp. 45–58). Global Change, The IGBP Series, Springer, Berlin-Heidelberg-New York.Google Scholar
  45. Pouyat, R. V., Yesilonis, I., Russell-Anelli, J., & Neerchal, N. K. (2007b). Soil chemical and physical properties that differentiate urban land-use and cover. Soil Science Society America Journal, 71, 1010–1019.CrossRefGoogle Scholar
  46. Schwarz, K., Pickett, S. T. A., Lathrop, R. G., Weathers, K. C., Pouyat, R. V., & Cadenasso, M. L. (2012). The effects of the urban built environment on the spatial distribution of lead in residential soils. Environmental Pollution, 163, 32–39.CrossRefGoogle Scholar
  47. Shinn, N. J., Bing-Canar, J., Cailas, M., Peneff, N., & Binns, H. J. (2000). Determination of spatial continuity of soil Pb levels in an urban residential neighborhood. Environmental Research, 82(1), 46–52.CrossRefGoogle Scholar
  48. S-Plus 6.1 for Windows. (2002). Insightful Corporation.Google Scholar
  49. Sutton, C. D. (2005). Classification and regression trees, bagging, and boosting. Handbook of Statistics, 24, 303–329.CrossRefGoogle Scholar
  50. United States Census Bureau. (2012). Topologically Integrated Geographic Encoding and Referencing (TIGER) database. May 20, 2012. <>.
  51. Wang, J., Ren, H., Liu, J., Yu, J., & Zhang, X. (2006). Distribution of Pb in urban soil and its potential risk in Shenyang City, China. Chinese Geographical Science, 16(2), 127–132.CrossRefGoogle Scholar
  52. Weathers, K. C., Lovett, G. M., Likens, G. E., & Lathrop, R. (2000). The effect of landscape features on deposition to Hunter Mountain, Catskill Mountains, New York. Ecological Applications, 10(2), 528–540.CrossRefGoogle Scholar
  53. Weathers, K. C., Simkin, S. M., Lovett, G. M., & Lindberg, S. E. (2006). Empirical modeling of atmospheric deposition in mountainous landscapes. Ecological Applications, 16(4), 1590–1607.Google Scholar
  54. Wong, C. S. C., Li, X., & Thornton, I. (2006). Urban environmental geochemistry of trace metals. Environmental Pollution, 142(1), 1–16.CrossRefGoogle Scholar
  55. Wright, J. P., Dietrich, K. N., Ris, M. D., Hornung, R. W., Wessel, S. D., Lanphear, B. P., et al. (2008). Association of prenatal and childhood blood lead concentrations with criminal arrests in early adulthood. PLoS Medicine, 5(5), 732–739.CrossRefGoogle Scholar
  56. Yesilonis, I., Pouyat, R., & Neerchal, N. K. (2008). Spatial distribution of metals in soils in Baltimore, Maryland: Role of native parent material, proximity to major roads, housing age and screening guidelines. Environmental Pollution, 156(3), 723–731.CrossRefGoogle Scholar
  57. Zhai, M., Kampunzu, H. A. B., Modisi, M. P., & Totolo, O. (2003). Distribution of heavy metals in Gaborone urban soils (Botswana) and its relationship to soil pollution and bedrock composition. Environmental Geology, 45(2), 171–180.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2013

Authors and Affiliations

  • Kirsten Schwarz
    • 1
    Email author
  • Kathleen C. Weathers
    • 3
  • Steward T. A. Pickett
    • 3
  • Richard G. LathropJr.
    • 4
  • Richard V. Pouyat
    • 5
  • Mary L. Cadenasso
    • 2
  1. 1.Department of Biological SciencesNorthern Kentucky UniversityHighland HeightsUSA
  2. 2.Department of Plant SciencesUniversity of California, DavisDavisUSA
  3. 3.Cary Institute of Ecosystem StudiesMillbrookUSA
  4. 4.Department of Ecology, Evolution, and Natural Resources, School of Environmental and Biological Sciences, Walton Center for Remote Sensing and Spatial AnalysisRutgers UniversityNew BrunswickUSA
  5. 5.US Forest ServiceArlingtonUSA

Personalised recommendations