Soil and tree response to P fertilization in a secondary tropical forest supported by an Oxisol

Mature tropical forests are considered to be P limited and to cycle P efficiently. Whether P limitations are significant in younger secondary tropical forests, however, remains largely unexplored. This study evaluated P limitation by observing the P fertilizer response of a naturally regenerated 24-year-old forest and its soil. In February 1999, six 20 × 20-m plots were established in secondary forest in the Brazilian Amazon. After 1 year of pre-treatment tree measurements, 50 kg P ha −1 was applied in January 2000 and again in January 2001. Soil sorption of P was relatively low (∼100 μg g −1 ) in the surface 0–20 cm while sorption increased to ∼180 μg g −1 at 20–50 cm and approached ∼500 μg g −1 for the 50- to 200-cm layers. Soil P in 0–10 cm, measured as sequentially extractable fractions (resin, HCO 3 -Pi, NaOH-Pi, NaOH-Po, and 1 M HCl), increased shortly after fertilization and could account for nearly all the 50 kg P ha −1 added at each date. During the following 6 years, soil P in fertilized plots declined in all pools other than resin P, and by June 2006, concentrations returned to pre-fertilization levels. Despite the increase in extractable P with fertilization, increased tree growth was not detected from stand age 25 to 31 years. It appears that during secondary forest succession at this site, the forest P cycle was conservative so as to maintain available P at a sufficient concentration to meet forest P demands.