The dynamics of phosphorus extractability, adsorption, and desorption rates as influenced by phosphorus applications and incubation times

Abstract

In a study to investigate the fate of the applied P in soils, a red-sandy clayey soil (Ferric Luvisols) from Rustenburg (high P fixing) and a red-sandy loam soil (Ferric Acrisols) from Loskop (low P fixing) were used. Sequential P fractionations were used to determine the content of the different P pools to show which pool the applied P was transformed to. The soils treatments consisted of different Prates (0, 25, 50, 100, 150, and 200 mg kg-1), and incubation periods (1, 60, 120, 180, and 240 days) under a laboratory conditions. The sequential P fractionation procedure consisted of extraction with hydrous ferric oxide in a dialysis membrane tube (DMT-HFO), 0.5M NaHC03, O.1M NaOH-P, 1.0M HCI, concentrated HCI, and concentrated H2S04 + H2O2. Approximately 30 to 60 % of the added P were transformed into less labile P pools within one day and 80-90 % after 60 days. This transformation was faster in the Rustenburg than in the Loskop soil showing a higher P fixation capacity. A major part of the P transformation was to the -OH-P1 pool with a recovery of about 30%. In the second experiment an attempt was made to determine P desorption rates by successive DMT-HFO extractions (1, 7, 14, 28, and 56 days) after the transformations of the applied P. This was followed by the sequential extractions to determine the changes and distribution of the added P into different P pools as well as which pools the P was des orbed from. The Rustenburg and Loskop soils were treated to different Prates (0, 25, 50, 100, and 200 mg P kg-1) and incubation periods (1, 120, and 240 days). The cumulative DMT -HFO extraction curves for 56 days showed that desorption could continue for a much longer period. This property is important in the economical management of fertilizer applications rates. Results showed the transformations and distribution of the applied P during incubation periods and proved that all the stable soil P pools contributed to the labile P pool by different proportions after prolonged successive DMT-HFO extractions. Although Rustenburg soil is considered a high P fixing soil, the P release rates under laboratory conditions were high enough to meet the requirements of cotton and tobacco crops. Root systems of these crops do not exploit 100 % soil volume as this laboratory method, which could explain why these crops experience P deficiencies. It is envisage that by using this method the P releasing properties of a soil could be used to develop a P desorption model to determine how much extractable P, with a specific extractant, in a particular soil, should be available at the beginning of a growing season to sustain a high enough P releasing rate to meet the requirements of a certain crop up to the end of the growing season. To do this, a model to describe root development that represents the percentage of the soil exploited P desorption rates that simulate P uptake by plant roots will be necessary.