Silica sources for arsenic mitigation in rice: machine learning-based predictive modeling and risk assessment.

Khanam, Rubina; Nayak, Amaresh Kumar; Kulsum, Pedda Ghouse Peera Sheikh; Mandal, Jajati; Shahid, Mohammad; Tripathy, Rahul; Bhattacharyya, Pratap; Selvam, Panneer; Munda, Sushmita; Manickam, Sivashankari; Debnath, Manish; Bandaru, Raghavendra Goud

Abstract

Arsenic (As) is a well-known human carcinogen, and the consumption of rice is the main pathway for the South Asian people. The study evaluated the impact of the amendments involving CaSiO , SiO nanoparticles, silica solubilizing bacteria (SSB), and rice straw compost (RSC) on mitigation of As toxicity in rice. The translocation of As from soil to cooked rice was tracked, and the results showed that RSC and its combination with SSB were the most effective in reducing As loading in rice grain by 53.2%. To determine the risk of dietary exposure to As, the average daily intake (ADI), hazard quotient (HQ), and incremental lifetime cancer risk (ILCR) were computed. The study observed that the ADI was reduced to one-third (0.24 μg kg bw) under RSC+SSB treatments compared to the control. An effective prediction model was established using random forest model and described the accumulation of As by rice grains depend on bioavailable As, P, and Fe which explained 48.5, 5.07%, and 2.6% of the variation in the grain As, respectively. The model anticipates that to produce As benign rice grain, soil should have P and Fe concentration more than 30 mg kg and 12 mg kg , respectively if soil As surpasses 2.5 mg kg . [Abstract copyright: © 2023. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.]

Citation

Khanam, R., Nayak, A. K., Kulsum, P. G. P. S., Mandal, J., Shahid, M., Tripathy, R., …Bandaru, R. G. (in press). Silica sources for arsenic mitigation in rice: machine learning-based predictive modeling and risk assessment. Environmental Science and Pollution Research, https://doi.org/10.1007/s11356-023-30339-5

• Journal Article Type: Article
• Acceptance Date: Oct 4, 2023
• Online Publication Date: Oct 18, 2023
• Deposit Date: Nov 8, 2023
• Publicly Available Date: Oct 19, 2024
• Journal: Environmental Science and Pollution Research
• Print ISSN: 0944-1344
• Publisher: Springer Verlag
• Peer Reviewed: Peer Reviewed
• DOI: https://doi.org/10.1007/s11356-023-30339-5
• Keywords: Silicon, Human exposure, Random forest model, Machine learning, Arsenic, Rice