The phosphorus cycle, labile nutrients, and soil water dynamics of Acacia koa forests: Effects of soil development and silviculture

Koa (Acacia koa, Gray), is an important endemic tree species in Hawaii. Little is known about the species, although previous research showed that its growth was limited by low soil phosphorus (P). Three sites were established on 3,000 yr-old, 20,000 yr-old, and 4.1 million yr-old soils to measure unmanaged koa stand dynamics after canopy closure and site nutrient characteristics. Koa stand age ranged from 23 to 35 years. A split-plot study was established at the youngest site to investigate the responses koa to silviculture. The main treatment was intraspecific thinning; sub-plot treatments were control, exotic grass control (H), and H + phosphorus (P) fertilisation (HF). The PRS-probe resin membrane was found as an effective means to measure the flux of soil labile nutrients. However, it did not sorbed nutrients as an infinite sink. Rather, membrane nutrient sorption was at a dynamic equilibrium with labile soil nutrients. Labile P and magnesium correlated with koa stand dynamics. Most Hedley fractionation P pools correlated with koa stand dynamics, with the hydrochloric-acid extractable inorganic P pool the strongest. Phosphorus fertilisation increased labile P and it remained high three years after the last P application. Fertilisation also increased foliar P, P retranslocation, litterfall P return, and most soil P pools. Litterfall koa biomass increased with H and HF treatments; however, HF was not greater until three years after the last P application. Other species litterfall returned proportionally more fertilised P to the forest floor than koa. The increase in other species litterfall P provided some compensation to the decrease in koa litterfall P after thinning. The sodium hydroxide- and hydrochloric-acid extractable inorganic P pools had the largest increases to applied P. These two pools were most likely the primary sources of elevated labile P; however, the mechanism is yet to be determined. Volumetric soil water did not change with the H treatment, but decreased with the HF treatment. This appeared to indicate that koa dominated stand was more P limited than water limited for the unthinned stands. Koa water stress only declined with a combination of thinning and HF treatments.