Unexpected seasonal variability of Hedley-occluded phosphorus in phosphorus-supplied subtropical forest soils

The bioavailability of phosphorus (P) in soils is regulated by the dynamics of its various pools. Yet, these temporal dynamics (i.e., transformation rates) remain poorly quantified, introducing substantial uncertainty into models incorporating P cycling. To improve predictive understanding of soil P bioavailability, we conducted a field experiment in a subtropical forest in China, sampling soils from two layers (topsoil: 0 − 10 cm; subsoil: 10 − 20 cm) at − 6 months (pre-treatment) and at 0.5, 1, 2, 4, 6, 8, 10, and 12 months after a one-time application of inorganic P (Pi) at four rates (0, 25, 50, and 100 kg P ha⁻¹). Soil P pools were quantified using a modified Hedley fractionation procedure. While organic P pools remained stable following P additions, Pi pools in the topsoil increased sequentially according to their theoretical bioavailability: readily available Pi (0.5 months), labile Pi (0.5 months), moderately labile Pi (2 months), primary mineral Pi (2 months) and occluded Pi (4 months). In the subsoil, corresponding increases were observed at 1, 0.5, 4, 2, and 4 months, respectively. Unexpectedly, field-observed variability in occluded Pi greatly exceeded predictions from seven land surface models, suggesting that these models may underestimate its dynamics. Moreover, this unexpected variability in occluded Pi was significantly and positively associated with soil pH and iron and aluminum oxides. Our findings reveal that occluded Pi is more dynamic than traditionally assumed, highlighting the need to refine its definition and representation in models to better interpret empirical data and simulate P bioavailability in terrestrial ecosystems.