Soil Aggregate Size Affects Phosphorus Desorption from Highly Weathered Soils and Plant Growth

Wiersum, 1962; Cornforth, 1968; Misra et al., 1988), but freshly applied P may penetrate only a thin layer Because plant absorption of P depends on the desorption of P around soil aggregates (Gunary et al., 1964; Linquist et from soil, understanding P desorption from soils may improve the precision of P diagnosis and fertilization recommendations. Many al., 1997), suggesting that larger soil aggregates with soils with high P retention due to high levels of Fe and Al are also relatively less surface area than small aggregates may highly aggregated. Extractable P is sometimes higher on larger size reduce P fixation and result in increased availability of aggregates, which will probably result in increased P release from recently applied P. With equivalent levels of extractable aggregates. The effects of aggregate size on P availability of three P, one might expect that desorption from larger aggrehighly weathered soils were quantified with a column-leaching study gates may be lower if P diffuses deeply into the agand a pot experiment. Phosphorus desorption by leaching from small gregates. aggregates was greater than that from large aggregates when P had Soil P release may affect both P supply to plants and been added to the bulk soil (Kapaa and Leilehua soils) and a mixture P discharge from watersheds. Desorption of P would, of different-sized aggregates (Leilehua soil). When aggregates were therefore, probably be more pertinent than P sorption separated and then P added, however, P desorption was greater from large aggregates (4–6 mm) than from small aggregates (,0.5 mm). when evaluating plant-available P and P buffering propConformity of the P desorption data to the parabolic diffusion and erties of soil, an observation noted many years ago by expanded Elovich equations suggests that P desorption is probably Fox and Kamprath (1970). Several experimental techcontrolled by diffusion processes. A pot experiment showed that total niques have been used to investigate P desorption. P in lettuce (Lactuca sativa L.) and soybean [Glycine max (L.) Merr.] These include extraction of soil P with P-free solution shoots, and the root dry weights of plants grown in the large aggregates (Kafkafi et al., 1967; Barrow, 1983), addition of materi(2–6 mm) were higher than for plants grown in the small aggregates als with high capacities to bind P in order to deplete P (,0.5 mm) after equal amounts of P were added to the separated in soils (Amer et al., 1955; Yang and Skogley, 1992), and aggregate fractions. Increased P uptake with increased aggregate size leaching of soil columns with P-free solutions (Sawhney, was attributed to increased P release from aggregates because of 1977; van der Zee and Gjaltema, 1992). The soil column reduced P fixation. The results suggest that soil management that favors soil aggregation may, in some cases, increase availability of leaching method was used in our experiments because applied P. Perhaps the distribution of soil aggregates should be considit prevented the breakup of soil aggregates resulting ered in making P management decisions. from the vigorous shaking required by the other methods. Leaching soil columns also permitted the removal of desorbed P with time, which simulates nutrient reA assessment of P availability in soils is bemoval by plant uptake more closely than batch equilicoming increasingly important from perspectives bration. of sustainability of agriculture and protecting the enviThe objectives of this study were (i) to investigate ronment. Although influences of soil chemical properthe effect of aggregate size on P availability as measured ties on P release from soils and uptake by plants are by extractable P and P desorption from different soils well documented, prediction of P requirements remains and (ii) to test the hypothesis that increased aggregation imprecise (Chen et al., 1997). For example, Cassman et increases P uptake by plants through decreased sorption al. (1993) found that the application of 100 kg P ha21 of applied P. to an Ultisol produced the maximum yield of soybean, although the recommended P application rate, based MATERIALS AND METHODS on the Fox and Kamprath method (Fox and Kamprath, Soils 1970), was more than 500 kg P ha21 (Cassman et al., 1981). One possible solution to such imprecision is a Two Oxisols (Kapaa and Wahiawa series) and an Ultisol (Leilehua series) from Hawaii were selected for this study. better understanding of the processes controlling P sorpSelected properties of the soils and P rate for each soil are tion and desorption from soils, particularly high P sorbpresented in Table 1. ing soils. Evidence shows that soil aggregate size influEight aggregate-size fractions were obtained from the Wahences P sorption and bioavailability (Wiersum, 1962; iawa and Kapaa soils by dry-sieving to obtain sizes of , 0.053, Gunary et al., 1964; Cornforth, 1968; Misra et al., 1988; 0.053 to 0.125, 0.125 to 0.25, 0.25 to 0.5, 0.5 to 1, 1 to 2, 2 to Linquist et al., 1997). Nutrient uptake from small aggre4, and 4 to 6 mm. The target P concentrations (1.6 mg P L2) gates often is greater than that from large aggregates in solution for the Wahiawa and Kapaa soils were maintained from 1971 to 1989 (Rajbhandari, 1991), and 0.5 M NaHCO3extractable P in these soils was very high (Table 1). For the X. Wang and R.S. Yost, Dep. of Tropical Plant and Soil Sciences, Univ. of Hawaii, Honolulu, HI, 96822; and B.A. Linquist, International Rice Research Institute, P.O. Box 933, Manila 1099, Philippines. Received Abbreviations: Delta P, difference in extractable P between before 4 June 1999. *Corresponding author ([email protected]). planting and after planting; GMD, geometric mean diameter; SE, standard error. Published in Soil Sci. Soc. Am. J. 65:139–146 (2001).