Prediction models for monitoring heavy-metal accumulation by wheat (Triticum aestivum L.) plants grown in sewage sludge amended soil. Issue 10 (23rd August 2020)

• Record Type: Journal Article
• Authors: Eid, Ebrahem M.; Galal, Tarek M.; El-Bebany, Ahmed F.
• Is Part Of: International journal of phytoremediation. Volume 22:Issue 10(2020)
• Page Start: 1000
• Page End: 1008
• Publication Date: 2020-08-23
• Subjects: Bio-concentration -- heavy metals -- prediction -- regression models -- wheat; Phytoremediation -- Periodicals; 628.5
• DOI: 10.1080/15226514.2020.1725866
• Languages: English
• ISSNs: 1522-6514

Abstract

Abstract: Prediction of heavy-metal concentration in the edible parts of economic crops, based on their concentration in soil and other environmental factors, is urgently required for human risk assessment. The present investigation aimed to develop regression models for predicting heavy-metal concentration in wheat plants via their contents in sewage sludge amended soil, organic matter (OM) content and soil pH. The concentration of heavy metals in the plant tissues reflected its concentration in the soil with high Fe followed by Al, Mn, Cr, Zn, Ni, Co, Cu, and Pb. Soil OM content had a significant positive correlation with all investigated heavy-metal concentrations in the different tissues of wheat plants, while soil pH was negatively significant with most heavy metals except spike Pb and grain Cr. The bio-concentration factor of Al, Cu, and Zn from soil to wheat root was >1, while that of shoot, spikes, and grains was <1 for all heavy metals. Significantly valid regression models were developed with fluctuated coefficient of determination ( R 2 ), high model efficiency (ME) values and low mean normalized average error (MNAE). The significant positive correlations between the concentration of some heavy metals in the soil and the same in wheat tissues indicate the potential of this plant as a biomonitor for these metals in contaminated soils. The significant correlations between heavy-metal concentrations in soil and its properties (pH and OM) with metal concentrations in wheat plants support the prediction model as an appropriate option. This study recommends the use of models with R 2 greater than 50% and recommend other researchers to use our models according to their own specific conditions.