Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Comparative Study
. 2017 Dec 26;15(1):34.
doi: 10.3390/ijerph15010034.

Comparison Study on the Estimation of the Spatial Distribution of Regional Soil Metal(loid)s Pollution Based on Kriging Interpolation and BP Neural Network

Zhenyi Jia  1 , Shenglu Zhou  2 , Quanlong Su  3 , Haomin Yi  4 , Junxiao Wang  5
Affiliations
Comparative Study

Comparison Study on the Estimation of the Spatial Distribution of Regional Soil Metal(loid)s Pollution Based on Kriging Interpolation and BP Neural Network

Zhenyi Jia et al. Int J Environ Res Public Health. .

Abstract

Soil pollution by metal(loid)s resulting from rapid economic development is a major concern. Accurately estimating the spatial distribution of soil metal(loid) pollution has great significance in preventing and controlling soil pollution. In this study, 126 topsoil samples were collected in Kunshan City and the geo-accumulation index was selected as a pollution index. We used Kriging interpolation and BP neural network methods to estimate the spatial distribution of arsenic (As) and cadmium (Cd) pollution in the study area. Additionally, we introduced a cross-validation method to measure the errors of the estimation results by the two interpolation methods and discussed the accuracy of the information contained in the estimation results. The conclusions are as follows: data distribution characteristics, spatial variability, and mean square errors (MSE) of the different methods showed large differences. Estimation results from BP neural network models have a higher accuracy, the MSE of As and Cd are 0.0661 and 0.1743, respectively. However, the interpolation results show significant skewed distribution, and spatial autocorrelation is strong. Using Kriging interpolation, the MSE of As and Cd are 0.0804 and 0.2983, respectively. The estimation results have poorer accuracy. Combining the two methods can improve the accuracy of the Kriging interpolation and more comprehensively represent the spatial distribution characteristics of metal(loid)s in regional soil. The study may provide a scientific basis and technical support for the regulation of soil metal(loid) pollution.

Keywords: BP neural network; cross validation; soil metal(loid)s; spatial interpolation.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Study site and spatial distribution of each kind of sample. The 126 total samples were divided into two kinds of sets, including 88 training sets (70%) and 38 validation sets (30%), respectively. In addition, another 88 densification sets were added to increase the sampling density.
Figure 2
Figure 2
Spatial distribution maps of the geo-accumulation indices of As and Cd before and after densification. The classification of the geo-accumulation index was obtained by using natural break interval approach. (a) The geo-accumulation index of As before densification; (b) the geo-accumulation index of As after densification; (c) the geo-accumulation index of Cd before densification; (d) the geo-accumulation index of Cd after densification.
Figure 3
Figure 3
Statistics of each pollution level before and after densification. For As, the pollution grades were clean and slight pollution. For Cd, the pollution grades were clean, slight, intermediate, and heavy pollution. (a) The area percentage of each pollution level before and after densification of As, respectively; (b) the area percentage of each pollution level before and after densification of Cd, respectively.
Figure 4
Figure 4
Spatial distribution maps of the mean square errors of the estimation results before and after densification. (a) The MSE of the geo-accumulation indices of As before densification; (b) the MSE of the geo-accumulation indices of As after densification; (c) the MSE of the geo-accumulation indices of Cd before densification; (d) the MSE of the geo-accumulation indices of Cd after densification.
See this image and copyright information in PMC

References

    1. Kaldor J., Harris J.A., Glazer E., Glaser S., Neutra R., Mayberry R., Nelson V., Robinson L., Reed D. Statistical association between cancer incidence and major-cause mortality, and estimated residential exposure to air emissions from petroleum and chemical plants. Environ. Health Perspect. 1984;54:319–332. doi: 10.1289/ehp.8454319. - DOI - PMC - PubMed
    1. Kelly J., Thornton I., Simpson P.R. Urban Geochemistry: A study of the influence of anthropogenic activity on the heavy metal content of soils in traditionally industrial and non-industrial areas of Britain. Appl. Geochem. 1996;11:363–370. doi: 10.1016/0883-2927(95)00084-4. - DOI
    1. Laidlaw M.A.S., Filippelli G.M., Brown S., Paz-Ferreiro J., Reichman S.M., Netherway P., Truskewycz A., Ball A.S., Mielke H.W. Case studies and evidence-based approaches to addressing urban soil lead contamination. Appl. Geochem. 2017;83:14–30. doi: 10.1016/j.apgeochem.2017.02.015. - DOI
    1. Luo X.S., Yu S., Zhu Y.G., Li X.D. Trace metal contamination in urban soils of China. Sci. Total Environ. 2012;421–422:17–30. doi: 10.1016/j.scitotenv.2011.04.020. - DOI - PubMed
    1. Wang M., Chen W., Peng C. Risk assessment of Cd polluted paddy soils in the industrial and township areas in Hunan, Southern China. Chemosphere. 2016;144:346–351. doi: 10.1016/j.chemosphere.2015.09.001. - DOI - PubMed

Publication types