Integrating a Hybrid Back Propagation Neural Network and Particle Swarm Optimization for Estimating Soil Heavy Metal Contents Using Hyperspectral Data

Piao Liu,Guangxing Wang,Zhou Shi,Lu Wang,Zhenhua Liu,Yuchun Pan,Yueming Hu

doi:10.3390/su11020419

Abstract

Soil heavy metals affect human life and the environment, and thus, it is very necessary to monitor their contents. Substantial research has been conducted to estimate and map soil heavy metals in large areas using hyperspectral data and machine learning methods (such as neural network), however, lower estimation accuracy is often obtained. In order to improve the estimation accuracy, in this study, a back propagation neural network (BPNN) was combined with the particle swarm optimization (PSO), which led to an integrated PSO-BPNN method used to estimate the contents of soil heavy metals: Cd, Hg, and As. This study was conducted in Guangdong, China, based on the soil heavy metal contents and hyperspectral data collected from 90 soil samples. The prediction accuracies from BPNN and PSO-BPNN were compared using field observations. The results showed that, 1) the sample averages of Cd, Hg, and As were 0.174 mg/kg, 0.132 mg/kg, and 9.761 mg/kg, respectively, with the corresponding maximum values of 0.570 mg/kg, 0.310 mg/kg, and 68.600 mg/kg being higher than the environment baseline values; 2) the transformed and combined spectral variables had higher correlations with the contents of the soil heavy metals than the original spectral data; 3) PSO-BPNN significantly improved the estimation accuracy of the soil heavy metal contents, with the decrease in the mean relative error (MRE) and relative root mean square error (RRMSE) by 68% to 71%, and 64% to 67%, respectively. This indicated that the PSO-BPNN provided great potential to estimate the soil heavy metal contents; and 4) with the PSO-BPNN, the Cd content could also be mapped using HuanJing-1A Hyperspectral Imager (HSI) data with a RRMSE value of 36%, implying that the PSO-BPNN method could be utilized to map the heavy metal content in soil, using both field spectral data and hyperspectral imagery for the large area.

Highlights

With the fast increase of industrial and chemical pesticide pollutants, many heavy metals enter the soil in many ways, which induces direct or indirect harms to the environment and humanity.Estimating soil heavy metal contents is necessary for monitoring the health of soil and for taking preventative measures to avoid contamination.The conventional method of estimating soil heavy metal contents is based on regular field samples and subsequent chemical analysis of the sampled soils in a laboratory, followed by spatial interpolation to acquire regional-scale maps of soil heavy metal contents
We integrated the correlation analysis and variance inflation factor (VIF) analysis for the selection of the hyperspectral data collected in the field, their transformations (FD, Second Derivative (SD), Logarithm of Reciprocal (LG), Reciprocal Transformation (RT), etc.), and the combinations of the transformations by addition, subtraction, multiplication, and division
The research results showed that the particle swarm optimization (PSO)-back propagation neural network (BPNN) method significantly increased the estimation accuracies of the soil heavy metal contents by greatly decreasing the Mean Relative Error (MRE) and RRMSE values

Summary

Introduction

The conventional method of estimating soil heavy metal contents is based on regular field samples and subsequent chemical analysis of the sampled soils in a laboratory, followed by spatial interpolation to acquire regional-scale maps of soil heavy metal contents. This is time-consuming and costly with low estimation accuracy at local areas [1,2]. Wu et al (2009) studied mid-infrared diffuse reflectance spectroscopy to accurately estimate heavy metal contents in soils for the mining areas located in Jiangning District and Baguazhou District [4]. Kooistra et al (2003) found that the soil spectral reflectance could be utilized to acquire the pollution levels of Zn and Cd in soils [5]

Objectives

Methods

Findings

Discussion

Conclusion