Development and comparison of regression models for the uptake of metals into various field crops
Language English Country England, Great Britain Media print-electronic
Document type Journal Article, Research Support, Non-U.S. Gov't
PubMed
26448504
DOI
10.1016/j.envpol.2015.09.043
PII: S0269-7491(15)30090-7
Knihovny.cz E-resources
- Keywords
- Field crops, Heavy metals, Linear regression, Plant uptake, Prediction models,
- MeSH
- Brassica rapa chemistry MeSH
- Food Contamination MeSH
- Zea mays chemistry MeSH
- Soil Pollutants analysis MeSH
- Environmental Monitoring methods statistics & numerical data MeSH
- Multivariate Analysis MeSH
- Triticum chemistry MeSH
- Soil chemistry MeSH
- Regression Analysis MeSH
- Models, Theoretical * MeSH
- Metals, Heavy analysis MeSH
- Crops, Agricultural chemistry MeSH
- Publication type
- Journal Article MeSH
- Research Support, Non-U.S. Gov't MeSH
- Geographicals
- Czech Republic MeSH
- Names of Substances
- Soil Pollutants MeSH
- Soil MeSH
- Metals, Heavy MeSH
Field crops represent one of the highest contributions to dietary metal exposure. The aim of this study was to develop specific regression models for the uptake of metals into various field crops and to compare the usability of other available models. We analysed samples of potato, hop, maize, barley, wheat, rape seed, and grass from 66 agricultural sites. The influence of measured soil concentrations and soil factors (pH, organic carbon, content of silt and clay) on the plant concentrations of Cd, Cr, Cu, Mo, Ni, Pb and Zn was evaluated. Bioconcentration factors (BCF) and plant-specific metal models (PSMM) developed from multivariate regressions were calculated. The explained variability of the models was from 19 to 64% and correlations between measured and predicted concentrations were between 0.43 and 0.90. The developed hop and rapeseed models are new in this field. Available models from literature showed inaccurate results, except for Cd; the modelling efficiency was mostly around zero. The use of interaction terms between parameters can significantly improve plant-specific models.
References provided by Crossref.org