Abstract
Phosphate sorption coefficient (PSC) has long been used to charcterize Andosols and to calculate the P rate for the amendment of poor P status in Andosols. For the determination of PSC, 25 g kg^<-1> of ammonium phosphate adjusted to pH 7 was added to a soil sample at a solution : soil ratio of 2 : 1. However, three problems were pointed out for the PSC as follows : (i) the reactivity of soil and phosphate depends on final phosphate concentrations, (ii) the PSC of most matured Andosols show similar values near the upper limit and (iii) the reason for a solution pH of 7 is not clear. Recently, it was further revealed that PSC was overestimated in Gley soils and Gray lowland soils due to the precipitation of CaHPO_4・2H_2O and MgNH_4PO_4・6H_2O. Hence a new method to determine phosphate retention capacity was developd to solve these problems. In the new method, 0.07 mol L^<-1> sodium phosphate solution adjusted to pH 4.3 using 0.1 mol L^<-1> of acetate buffer was added to soils at solution : soil ratios of 2 : 1, 10 : 1 and 40 : 1. Among them, using two results interposing a final phosphate concentration of 0.05 mol L^<-1>, phosphate retention capacity was calculated at the final phosphate concentration of 0.05 mol L^<-1>. The phosphate retention capacity was free from precipitation of CaHPO_4・2H_2O and MgNH_4PO_4・6H_2O in most Gley soils and Gray lowland soils. A linear relationship was observed between phosphate retention capacity and the content of oxalate-extractable Al (Al_0) plus half iron (Fe_0) for Andosols and non-allophanic Andosols. The atomic ratio of the PSC / (AL_0+Fe_0) was 0.16-0.21 for allophanic Ardosols and 0.32 0.38 for non-allophanic Andosols.
