|
(M/EH128) A Robust Model for Predicting Phytoavailability of Copper and Zinc. Saxe, Jennifer1, Impellitteri, Christopher1, Allen, Herbert1, 1 ABSTRACT- Trace quantities of copper and zinc are essential for plant health. Higher doses of these metals are toxic to plants, however. Soils may contain copper or zinc at concentrations exceeding the global geochemical average by several orders of magnitude due to local geology or anthropogenic sources. Analysis of soil digests for total copper or zinc is commonly used to assess the potential for copper or zinc toxicity from soil exposure. The total metal in a soil however, is not a good predictor of the metal concentrations found in plant tissue. Because of this, toxic effects may or may not be experienced by plants grown in soils that are indicated as risky due to high total metal concentrations. It is already well understood that trace metal species' bioavailability in environmental systems is altered depending on local pH and the presence of competing cations. These, as well as additional soil characteristics were used to develop a neural network phytoavailability model for copper and zinc. The model predicts the metal content in shoots of three plant species grown in widely varied field soils. Lettuce (Lactuca sativa), Indian mustard (Brassica juncea) and barley (Hordeum vulgare) were grown in the laboratory on domestic and international soils. The metal measured using soil extractions that are less stringent than complete digestions is more appropriate than total metal for assessing bioavailable fractions in soil. Soil extracts using 0.01 M HCl (Cu) or 0.01 M CaCl2 (Zn) are best correlated with plant metal levels. Soil characteristics including a measure of acid neutralization capacity, soil carbon content, soil texture, and macronutrient levels are necessary for predicting plant metal concentrations over a comprehensive range of soil types. Key words: bioavailability, plants, copper, zinc |
