Source identification and contribution of land uses to the observed values of heavy metals in soil samples of the border between the Northern Ireland and Republic of Ireland by receptor models and redundancy analysis
Introduction
Among environmental contaminants, heavy metals (HMs) have attracted great attention of researchers due to toxic properties of some species besides their difficult degradation (Zhong et al., 2014, Liang et al., 2017). In this context, heavy metals refer to any chemical element with atomic density greater than 4 g/cm3 (Duruibe et al., 2007) whereas, trace elements imply those elements that occur in small amounts in natural environment (Blum et al., 2009) and when present in excessive bioavailable concentrations are known as toxic elements. Nonetheless, the term HM has become less popular in environmental sciences so, it was replaced with potentially toxic elements (PTEs) in some recent researches (Pourret and Hursthouse 2019).
In this respect, the traditional method for source identification of HMs is multivariate statistical methods, but application of receptor models has become more popular among researchers. These models were originally developed for dealing with air contaminants however, since then they have been widely applied for source identification of pollutants in soil (Lv and Wang 2019), surface water (Shi et al., 2012) and groundwater (Meng et al., 2018) ecosystems.
Some receptor models such as positive matrix factorization (PMF) (Zhang et al., 2018a, Wu et al., 2019, Men et al., 2019) and Unmix (Jain et al., 2017, Gulgundi and Shetty, 2019) are able to estimate the uncertainty of extracted pollution sources which lack in conventional methods of source apportionment like multivariate statistical methods including factor analysis, principal component analysis and cluster analysis. In this respect, the most common receptor models include PMF (Zhang et al., 2018b, Wu et al., 2019), principal component analysis/absolute principal component scores (PCA/APCS) (Singh et al., 2008), Unmix (Jain et al., 2018, Guan et al., 2019) and chemical mass balance (CMB) (Feng et al., 2019, Cheng et al., 2020). To this extent, despite lots of past and recent studies conducted about source apportionment of HMs through receptor models (Yang et al., 2013, Salim et al., 2019, Guan et al., 2019, Lv, 2019, Wang et al., 2020b), there is not any common conclusion regarding the efficiency of these approaches among researchers. For instance, Salim et al. (2019) employed spatial storm water runoff data in order to identify the contributing pollution sources using PMF and PCA/APCS models and concluded that PMF, as the superior model, is able to reveal a larger number of sources. In particular, in a recent study, Guan et al. (2019) compared the efficiency of grouped principal component analysis/absolute principal component scores (GPCA/APCS), PMF and Unmix for identification the contribution of different sources to heavy metal contamination in soil samples of Wuwei, China and concluded that GPCA/APCS was the optimal method. Since there is a dearth of information regarding the performance of receptor models for source identification of soil pollution so, further research is required to unveil the efficiency of these models.
On the other hand, from the management point of view, it is important to understand whether or not there is any difference between the levels of HMs in various types of land uses. Such data will provide the basis for prioritization of management goals and their efficient control in this regard. Different methods were employed to investigate the impacts of land uses on degradation the quality of environmental media such as pollution indices (Trujillo-González et al., 2016) and multivariate statistical methods including ensemble of multiple linear regression and constrained least squares (Kang et al., 2010), cluster analysis (Kändler et al., 2017), geographically weighted regression (Wu et al., 2017) and redundancy analysis (RDA) (Gabarrón et al., 2019, Zhang et al., 2019b). Among these techniques, RDA proved to be one of the most efficient methods and has been applied widely in different parts of the world, recently. Despite the fact that receptor models can help to identify the primary sources of contamination in an area of study however, more in depth analysis of these sources is not possible by only relying on receptor models. The joint application of receptor models and redundancy analysis would enhance our understanding of the exact land uses and contributing factors for contamination of soil and prioritization of management goals for their control. This study is the first joint application of the three most applied receptor models, redundancy analysis together with multivariate statistical methods for identification the sources of HMs accumulation in soil and analysis the contribution of land uses to the observed concentrations of HMs, to the best of our knowledge. The only recent application of a receptor model and redundancy analysis was the study of Jia et al. (2020) focusing on source apportionment by the combined method of absolute principal component score/multiple linear regression (APCS/MLR) and redundancy analysis for this purpose. Accordingly, the main objectives of this study were (i) source apportionment of HMs (e.g. Cr, Ni, Co, Zn, Cu, Hg, Cd, As, Pb) in soil samples of the border of Republic of Ireland and Northern Ireland using receptor models including PMF, Unmix and PCA/APCS (ii) uncertainty analysis of identified sources using both PMF and Unmix models (iii) assessment the relationship between land use types, pH together with soil organic carbon and levels of HMs.
Section snippets
Tellus project source data
The source data applied in the current research consisted of an integrated analysis under Tellus Project including soil, sediment and water compartments performed by the Geological Survey of Ireland (GSI). The Tellus project is a national programme to collect geochemical and geophysical data across the Republic of Ireland and Northern Ireland. Sampling was based on an adhoc gridded survey design with one sample being collected per 4 km2 of the survey grid cell. Surveying was completed in
PmF
Regarding the optimum number of factors to retain for source identification by the PMF model; there was a great difference between the R2 values when the number of factors increased from 2 to 4. More specifically, the initial model run by 2 factors could not capture the true variability of Cu, As, Cd and Hg resulting in R2 values of lower than 0.6 whereas, in the revised model by 4 factors the amount of R2 for all of the elements other than Hg and As was higher than 0.80 indicating the
Conclusion
In this study, the primary sources of heavy metals accumulation and contribution of land uses to concentrations of HMs in soil samples from the border of Republic of Ireland and Northern Ireland were investigated using three receptor models (PMF, Unmix, PCA/APCS) and redundancy analysis along with multivariate analysis. According to the findings, the following conclusions can be made:
- 1)
None of the receptor models could attribute arsenic to a separate source which might be either due to the fact
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors of this paper would like to express their sincere gratitude to the Geological Survey of Ireland, Geological survey of Northern Ireland and Environmental Protection Agency of Ireland for providing the data applied in the current research. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Parts of the Tellus source data are freely available under the open data license CC BY 4.0. The license conditions can be
References (87)
- et al.
Cadmium and other heavy metal concentrations in bovine kidneys in the Republic of Ireland
Sci. Total Environ.
(2014) - et al.
The use of microchemical analysis of alluvial gold grains in mineral exploration: experiences in Britain and Ireland
J. Geochem. Explor.
(2000) - et al.
An integrated chemical mass balance and source emission inventory model for the source apportionment of PM2. 5 in typical coastal areas
J. Environ. Sci.
(2020) - et al.
Determination of copper and mercury in phosphate fertilizers employing direct solid sampling analysis and high resolution continuum source graphite furnace atomic absorption spectrometry
Spectrochim. Acta, Part B
(2015) - et al.
Geostatistical mapping and quantitative source apportionment of potentially toxic elements in top-and sub-soils: A case of suburban area in Beijing, China
Ecol. Indicat.
(2020) - et al.
Multivariate statistical and GIS-based approach to identify heavy metal sources in soils
Environ. Pollut.
(2001) - et al.
Effect of land use and soil properties in the feasibility of two sequential extraction procedures for metals fractionation
Chemosphere
(2019) - et al.
Source apportionment of heavy metals in farmland soil of Wuwei, China: Comparison of three receptor models
J. Cleaner Prod.
(2019) Multivariate receptor modeling by N-dimensional edge detection
Chemomet. Intellig. Laborat. Syst.
(2003)- et al.
Spatial variation of PM2. 5 chemical species and source-apportioned mass concentrations in New York City
Atmos. Environ.
(2004)
Source apportionment of PM10 in Delhi, India using PCA/APCS, UNMIX and PMF
Particuology
Identification of the sources and influencing factors of potentially toxic elements accumulation in the soil from a typical karst region in Guangxi, Southwest China
Environ. Pollut.
Impact of land use on water quality in the upper Nisa catchment in the Czech Republic and in Germany
Sci. Total Environ.
Geochemistry of serpentine agricultural soil and associated groundwater chemistry and vegetation in the area of Atalanti, Greece
J. Geochem. Explor.
Linking land-use type and stream water quality using spatial data of fecal indicator bacteria and heavy metals in the Yeongsan river basin
Water Res.
Cadmium and zinc in soil solution extracts following the application of phosphate fertilizers
Sci. Total Environ.
Combination of Unmix and PMF receptor model to apportion the potential sources and contributions of PAHs in wetland soils from Jiaozhou Bay, China
Mar. Pollut. Bull.
Spatial distribution and source identification of heavy metals in surface soils in a typical coal mine city, Lianyuan, China
Environ. Pollut.
Source identification of potentially hazardous elements and their relationships with soil properties in agricultural soil of the Pinggu district of Beijing, China: multivariate statistical analysis and redundancy analysis
J. Geochem. Explor.
Multivariate and geostatistical analyses of the spatial distribution and origin of heavy metals in the agricultural soils in Shunyi, Beijing, China
Sci. Total Environ.
Multivariate receptor models and robust geostatistics to estimate source apportionment of heavy metals in soils
Environ. Pollut.
PMF receptor models and sequential Gaussian simulation to determine the quantitative sources and hazardous areas of potentially toxic elements in soils
Geoderma
Uncertainty analysis in source apportionment of heavy metals in road dust based on positive matrix factorization model and geographic information system
Sci. Total Environ.
Apportionment and evolution of pollution sources in a typical riverside groundwater resource area using PCA-APCS-MLR model
J. Contam. Hydrol.
Lead within ecosystems on metalliferous mine tailings in Wales and Ireland
Sci. Total Environ.
Multiscale analysis of heavy metal contents in soils: spatial variability in the Duero river basin (Spain)
Geoderma
Inputs of trace elements in agricultural soils via phosphate fertilizers in European countries
Sci. Total Environ.
Characterization of metal pollution in soils under two landuse patterns in the Angouran region, NW Iran; a study based on multivariate data analysis
J. Hazard. Mater.
Multiscale analysis of heavy metal contents in Spanish agricultural topsoils
Chemosphere
Comparison of two receptor models PCA-MLR and PMF for source identification and apportionment of pollution carried by runoff from catchment and sub-watershed areas with mixed land cover in South Korea
Sci. Total Environ.
Impact of landscape pattern at multiple spatial scales on water quality: A case study in a typical urbanised watershed in China
Ecol. Ind.
Multivariate and geostatistical analyses of the spatial distribution and sources of heavy metals in agricultural soil in Dehui
Northeast China. Chemosphere
Heavy metal accumulation related to population density in road dust samples taken from urban sites under different land uses
Sci. Total Environ.
Screening biochars for heavy metal retention in soil: role of oxygen functional groups
J. Hazard. Mater.
Cadmium governance in Europe's phosphate fertilizers: Not so fast?
Sci. Total Environ.
Quadratic discriminant analysis model for assessing the risk of cadmium pollution for paddy fields in a county in China
Environ. Pollut.
Geo-statistical and multivariate analyses of potentially toxic elements' distribution in the soil of Hainan Island (China): A comparison between the topsoil and subsoil at a regional scale
J. Geochem. Explor.
Identifying quantitative sources and spatial distributions of potentially toxic elements in soils by using three receptor models and sequential indicator simulation
Chemosphere
Elucidating the differentiation of soil heavy metals under different land uses with geographically weighted regression and self-organizing map
Environ. Pollut.
Spatial patterns and origins of heavy metals in Sheyang River catchment in Jiangsu, China based on geographically weighted regression
Sci. Total Environ.
Source apportionment of polycyclic aromatic hydrocarbons in soils of Huanghuai Plain, China: comparison of three receptor models
Sci. Total Environ.
Beyond mere pollution source identification: Determination of land covers emitting soil heavy metals by combining PCA/APCS, GeoDetector and GIS analysis
Catena
Environmental geochemical mapping and multivariate geostatistical analysis of heavy metals in topsoils of a closed steel smelter: Capital Iron & Steel Factory, Beijing, China
J. Geochem. Explor.
Cited by (31)
Incorporating source apportionment and bioaccessibility into human health risk assessment of heavy metals in a soil-rice system in the Jiulong River basin, southeast China
2023, Journal of Food Composition and Analysis