A joint standard-exceeding risk assessment of multiple pollutants based on robust geostatistics with categorical land-use type data: A case study of soil nitrogen and phosphorus. (15th April 2022)

• Record Type: Journal Article
• Authors: Chen, Jian; Qu, Mingkai; Wang, Yan; Huang, Biao; Zhao, Yongcun
• Is Part Of: Environmental pollution. Volume 299(2022)
• Publication Date: 2022-04-15
• Subjects: Soil available nitrogen -- Soil available phosphorus -- Robust sequential Gaussian simulation with land-use types -- Spatial uncertainty -- Joint standard-exceeding probability -- Joint regulation of multiple soil pollutants; Pollution -- Periodicals; Pollution -- Environmental aspects -- Periodicals; Environmental Pollution -- Periodicals; Pollution -- Périodiques; Pollution -- Aspect de l'environnement -- Périodiques; Pollution -- Effets physiologiques -- Périodiques; Pollution; Pollution -- Environmental aspects; Periodicals; Electronic journals; 363.73
• DOI: 10.1016/j.envpol.2022.118901
• Languages: English
• ISSNs: 0269-7491

Abstract

Abstract: Joint standard-exceeding risk and its spatial uncertainty of soil available nitrogen (AN) and available phosphorus (AP) under the specific constraints are essential for guiding the joint regulation of pollutants but were rarely considered by previous studies. Moreover, traditionally-used spatial simulation models are not only non-robust but also ignoring valuable categorical information (e.g., land-use types), which may hinder the acquisition of high-precision spatial simulation results. This study first established optimally robust semi-variogram estimators to identify the spatial outliers of soil AN and AP in Jintan County, China. Then, robust sequential Gaussian simulation (RSGS) with land-use types (RSGS-LU) was proposed and further compared with RSGS, SGS-LU, and SGS in the spatial simulation accuracy. Last, a joint standard-exceeding probability model under the specific constraints was proposed, and the corresponding high-risk areas were delineated for the joint regulation of soil AN and AP. Results showed that: (i) 23 and 17 spatial outliers were identified for soil AN and AP, respectively; (ii) removing outliers or combining land-use types could improve the spatial simulation accuracy of soil AN and AP; (iii) RSGS-LU generated the highest spatial simulation accuracy for both soil AN and AP; (iv) the area with the joint standard-exceeding (AP > 30 mg kg −1 ∪ AN > 130 mg kg −1 ) probability >75% accounted for 9.98% of the county's area; (iv) the area with the joint standard-exceeding (AP > 30 mg kg −1 ∩ AN > 130 mg kg −1 ) probability >75% accounted for 2.29% of the county's area. It is concluded that RSGS-LU and joint standard-exceeding probability model under the specific constraints could provide more accurate and flexible spatial decision support for the joint regulation of soil AN and AP at a regional scale. Moreover, the methods recommended in this study also provide valuable tools for the joint standard-exceeding risk assessment of other multiple soil pollutants. Graphical abstract: Image 1 Highlights: Spatial outliers of soil AN and AP were identified based on robust semi-variograms. Robust sequential Gaussian simulation with land-use types (RSGS-LU) was proposed. RSGS-LU obtained a higher spatial simulation accuracy than SGS-LU, RSGS, and SGS. A joint standard-exceeding probability model under specific constraints was proposed. Flexible spatial decision support for the joint regulation of AN and AP was obtained.