Exploring the database of a soil environmental survey using a geo-self-organizing map: A pilot study

doi:10.1007/s11442-019-1644-8

Abstract

Abstract:

A model integrating geo-information and self-organizing map (SOM) for exploring the database of soil environmental surveys was established. The dataset of 5 heavy metals (As, Cd, Cr, Hg, and Pb) was built by the regular grid sampling in Hechi, Guangxi Zhuang Autonomous Region in southern China. Auxiliary datasets were collected throughout the study area to help interpret the potential causes of pollution. The main findings are as follows: (1) Soil samples of 5 elements exhibited strong variation and high skewness. High pollution risk existed in the case study area, especially Hg and Cd. (2) As and Pb had a similar topological distribution pattern, meaning they behaved similarly in the soil environment. Cr had behaviours in soil different from those of the other 4 elements. (3) From the U-matrix of SOM networks, 3 levels of SEQ were identified, and 11 high risk areas of soil heavy metal-contaminated were found throughout the study area, which were basically near rivers, factories, and ore zones. (4) The variations of contamination index (CI) followed the trend of construction land (1.353) > forestland (1.267) > cropland (1.175) > grassland (1.056), which suggest that decision makers should focus more on the problem of soil pollution surrounding industrial and mining enterprises and farmland.

Key words: self-organizing map, geo-information, heavy metal, soil environmental quality, Hechi

LIAO Xiaoyong, TAO Huan, GONG Xuegang, LI You. Exploring the database of a soil environmental survey using a geo-self-organizing map: A pilot study[J].Journal of Geographical Sciences, 2019, 29(10): 1610-1624.

Figures/Tables 10

Figure 1

Figure 2

Table 1

Figure 3

Table 2

Figure 4

Figure 5

Table 3

Figure 6

Table 4

References 36

[1]	Alvarez-Guerra M, González-Piñuela C, Andrés A , et al., 2008. Assessment of self-organizing map artificial neural networks for the classification of sediment quality. Environment International, 34(6):782-790.
[2]	Anagu I, Ingwersen J, Utermann J et al., 2009. Estimation of heavy metal sorption in German soils using artificial neural networks. Geoderma, 152(1/2):104-112.
[3]	Astel A, Tsakovski S, Barbieri P , et al., 2007. Comparison of self-organizing maps classification approach with cluster and principal components analysis for large environmental data sets. Water Research, 41(19):4566-4578.
[4]	Bação F, Lobo V, Painho M , 2004. Geo-self-organizing map (Geo-SOM) for building and exploring homogeneous regions. In: Egenhofer M J, Freksa C, Miller H J. Third International Conference on GIScience. Berlin: Springer, 22-37.
[5]	Buszewski B, Kowalkowski T , 2006. A new model of heavy metal transport in the soil using nonlinear artificial neural networks. Environmental Engineering Science, 23(4):589-595.
[6]	Cai L M, Huang L C, Zhou Y Z , et al., 2010. Heavy metal concentrations of agricultural soils and vegetables from Dongguan, Guangdong. Journal of Geographical Sciences, 20(1):121-134.
[7]	Chang D H, Islam S , 2000. Estimation of soil physical properties using remote sensing and artificial neural network. Remote Sensing of Environment, 74(3):534-544.
[8]	Cockx L, Van Meirvenne M, Verbeke L P C et al., 2009. Extracting topsoil information from EM38DD sensor data using a neural network approach. Soil Science Society of America Journal, 73(6):2051-2058.
[9]	Dotaniya M L, Meena V D, Rajendiran S et al., 2017. Geo-accumulation indices of heavy metals in soil and groundwater of Kanpur, India under long term irrigation of tannery effluent. Bulletin of Environmental Contamination and Toxicology, 98(5):706-711.
[10]	Gao B, Lu A, Pan Y , et al., 2017. Additional sampling layout optimization method for environmental quality grade classifications of farmland soil. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 10(12):5350-5358.
[11]	Guan Y, Shao C, Gu Q , et al., 2016. Study of a comprehensive assessment method of the environmental quality of soil in industrial and mining gathering areas. Stochastic Environmental Research and Risk Assessment, 30(1):91-102.
[12]	Huang K X, Qin L M, Wu S Z et al., 2012. Situation and remedial measures for heavy metals pollution in Hechi city of Guangxi. Journal of Guangxi Academy of Sciences, 28(4):320-324. (in Chinese)
[13]	Huang Y, Ye H, Zhang L , et al., 2017. Prediction of soil organic matter using ordinary kriging combined with the clustering of self-organizing map: A case study in Pinggu District, Beijing, China. Soil Science, 182(2):52-62.
[14]	Jaffar S T A, Luo F, Ye R et al., 2017. The extent of heavy metal pollution and their potential health risk in topsoils of the massively urbanized district of Shanghai. Archives of Environmental Contamination and Toxicology, 73(3):362-376.
[15]	Kohonen T , 1982. Self-organized formation of topologically correct feature maps. Biological Cybernetics, 43(1):59-69.
[16]	Kohonen T, Somervuo P , 2002. How to make large self-organizing maps for nonvectorial data. Neural Networks, 15(8/9):945-952.
[17]	Kong X T , 2014. China must protect high-quality arable land. Nature, 506(7486 ): 7.
[18]	Li X, Gao B Pan Y , et al., 2016. The soil heavy metal content mapping based on Sandwich model. In: 2016 Fifth International Conference on Agro-Geoinformatics (Agro-Geoinformatics), Piscataway:IEEE, 1-6.
[19]	Li Y, Li C K, Tao J J , et al., 2011. Study on spatial distribution of soil heavy metals in Huizhou City based on BP-ANN modeling and GIS. Procedia Environmental Sciences, 10:1953-1960.
[20]	Muleta M K, Nicklow J W , 2005. Decision support for watershed management using evolutionary algorithms. Journal of Water Resources Planning and Management, 131(1):35-44.
[21]	Nadal M, Schuhmacher M, Domingo J L , 2004. Metal pollution of soils and vegetation in an area with petrochemical industry. Science of the Total Environment, 321(1-3):59-69.
[22]	Nemerow N L , 1974. Scientific Stream Pollution Analysis. New York: McGraw-Hill.
[23]	Olawoyin R, Nieto A, Grayson R L et al., 2013. Application of artificial neural network (ANN)-self-organizing map (SOM) for the categorization of water, soil and sediment quality in petrochemical regions. Expert Systems with Applications, 40(9):3634-3648.
[24]	Pan Y, Li H , 2016. Investigating heavy metal pollution in mining brownfield and its policy implications: A case study of the Bayan Obo rare earth mine, Inner Mongolia, China. Environmental Management, 57(4):879-893.
[25]	Patel R M, Prasher S O, God P K , et al., 2002. Soil salinity prediction using artificial neural networks. Journal of the American Water Resources Association, 38(1):91-100.
[26]	Rivera D, Sandoval M, Godoy A , 2015. Exploring soil databases: A self-organizing map approach. Soil Use and Management, 31(1):121-131.
[27]	Sakizadeh M, Mirzaei R, Ghorbani H , 2017. Support vector machine and artificial neural network to model soil pollution: A case study in Semnan Province, Iran. Neural Computing and Applications, 28(11):3229-3238.
[28]	Somaratne S, Seneviratne G, Coomaraswamy U , 2005. Prediction of soil organic carbon across different land-use patterns. Soil Science Society of America Journal, 69(5):1580-1589.
[29]	Tóth G, Hermann T, Da Silva M R et al., 2016. Heavy metals in agricultural soils of the European Union with implications for food safety. Environment International, 88:299-309.
[30]	Vesanto J , 2002. Data exploration process based on the self-organizing map [D]. Helsinki: Helsinki University of Technology.
[31]	Wang J F, Haining R, Liu T J et al., 2013. Sandwich estimation for multi-unit reporting on a stratified heterogeneous surface. Environment and Planning A, 45(10):2515-2534.
[32]	Wang J F, Zhang T L, Fu B J , 2016. A measure of spatial stratified heterogeneity. Ecological Indicators, 67:250-256.
[33]	Wang Y B, Liu C W, Wang S W , 2015. Characterization of heavy-metal-contaminated sediment by using unsupervised multivariate techniques and health risk assessment. Ecotoxicology and Environmental Safety, 113:469-476.
[34]	Wu Y , 2015. Characteristics of soil heavy metal contamination around industrial and mining enterprises in Diaojiang river basin, Guangxi Zhuang Autonomous Region, China [D]. Beijing: University of Chinese Academy of Sciences. (in Chinese)
[35]	Yang C, Guo R, Wu Z et al., 2014. Spatial extraction model for soil environmental quality of anomalous areas in a geographic scale. Environmental Science and Pollution Research, 21(4):2697-2705.
[36]	Zhou P, Zhao Y, Zhao Z et al., 2015. Source mapping and determining of soil contamination by heavy metals using statistical analysis, artificial neural network, and adaptive genetic algorithm. Journal of Environmental Chemical Engineering, 3(4):2569-2579.

Elements	Mean (mg·kg^-1)	Standard deviation (mg·kg^-1)	Variation coefficient (%)	Skewness	Kurtosis	Chinese standard (mg·kg^-1)	Exceeding rate (%)
As	27.05	94.86	351	16.03	297.07	15	42.69
Cd	1.47	4.11	280	9.15	123.09	0.2	51.85
Cr	68.27	56.86	83	2.25	9.83	90	21.25
Hg	0.48	0.72	150	5.20	39.84	0.15	73.88
Pb	76.06	192.39	250	8.50	84.73	35	62.00

Rows	Columns	Map size	Variable		Log		Range
Rows	Columns	Map size	QE	TE	QE	TE	QE	TE
11	8	88	0.47	0.02	0.62	0.03	0.04	0.11
12	8	96	0.46	0.01	0.60	0.02	0.04	0.11
11	9	99	0.45	0.02	0.60	0.04	0.04	0.15
*13	*8	*104	0.45	0.03	0.59	0.03	*0.04	*0.06
12	9	108	0.44	0.02	0.58	0.04	0.04	0.12
11	10	110	0.44	0.03	0.58	0.04	0.04	0.18
13	9	117	0.43	0.02	0.57	0.04	0.04	0.08
12	10	120	0.43	0.02	0.56	0.04	0.04	0.14
11	11	121	0.42	0.03	0.56	0.03	0.04	0.17
13	10	130	0.42	0.02	0.55	0.03	0.03	0.14
12	11	132	0.42	0.02	0.55	0.03	0.03	0.18
13	11	143	0.41	0.02	0.54	0.03	0.03	0.15

SEQ	Mean content of elements (mg·kg^-1)					Counts
SEQ	As	Cd	Cr	Hg	Pb	Counts
Level 1	16.74	0.46	47.1	0.33	47.24	413	2.85
Level 2	27.52	3.94	169.1	0.66	69.17	84	15.18
Level 3	290.04	8.9	86.88	3.29	782.23	16	49.76

Land-use types	SEQ			Counts	CI
Land-use types	Level 1	Level 2	Level 3	Counts	CI
Construction land	36 (70.6%)	12 (23.5%)	3 (5.9%)	51	1.353
Cropland	167 (86.1%)	20 (10.3%)	7 (3.6%)	194	1.175
Forestland	176 (75.9%)	50 (21.6%)	6 (2.6%)	232	1.267
Grassland	34 (94.4%)	2 (5.6%)	0 (0%)	36	1.056