Causes and prediction of changes in extractable phosphorus during flooding

Abstract

In a laboratory experiment, samples of 18 soils, which are known to be flooded in the field, were flooded for up to 32 days. Both untreated and phosphate-treated (50 mg P kg-1) soils were studied. It was attempted to identify which chemical properties measured on the dry untreated soils, and the changes in pH, Eh and extractable Fe and Mn over the flooding periods, controlled the changes in sodium acetate buffer (pH 3.0) extractable phosphorus during flooding. It was shown that the reduction of iron(III) oxides was the dominant source of the P released during flooding. However, the amount of P released was strongly inhibited by re-sorption. Direct measurement of the amount of iron(III) reduced during flooding and measurement of phosphate sorption were required to predict the amount of P released during flooding. Organic matter contributed toward the P released during flooding. Its contribution appeared to be by mineralization, rather than by accelerating FeIII reduction. The reduction of MnIII and MnIII was a minor source of P in the untreated soils. Changes in soil pH during flooding were responsible for desorption of freshly applied P, but did not appear to affect P release in the untreated soils. The Vertisols and some of the Alfisols showed very little, or no release of P during flooding.