Effects of Nitrogen and Phosphorus Availability on Fine-Root Dynamics in Hawaiian Montane Forests

Abstract

Although it is often assumed that root dynamics are similar to leaf dynamics in relation to nutrient availability, this hypothesis is rarely tested. Using sequential soil coring and a decomposition experiment, patterns of fine-root (<2 mm diameter) biomass, belowground net primary productivity, and root turnover rates were examined over a 1-yr period in three sites along a forest chronosequence in the Hawaiian Islands. These sites form a natural fertility gradient but have similar species composition, climate, and geology. The youngest site (300 yr old) was low in N availability, the oldest site (4.1 × 106 yr old) was low in P availability, and the intermediate-aged site (20 000 yr old) was higher in both N and P. The youngest and oldest sites also contained long-term fertilization plots that were fertilized with N or P, or were unfertilized (control). Along the natural fertility gradient, the oldest site generally differed from the other two sites; this site had the greatest standing stock of live root mass, the greatest root length density, the lowest belowground net primary productivity (BNPP), and the slowest root turnover rate. The 300-yr-old and 20 000-yr-old sites were similar in all of these variables, except that root turnover was slightly faster at the 300-yr-old site. The response of fine roots to fertilization was consistent with the response along the natural fertility gradient. Despite N limitation to aboveground growth at the 300-yr-old site, fertilization had small effects on root variables, with the only significant effects being a small increase in the standing stock of live root biomass in N-fertilized plots and an increase in root tissue P concentrations in P-fertilized plots. In contrast, fertilization at the 4.1 × 106 yr-old site altered root dynamics more than fertilization at the 300-yr-old site. In the P-fertilized plots, P concentrations increased, BNPP tended to be greater, and root turnover rates increased. These results suggest that root dynamics differ dramatically between ecosystems low in N and ecosystems low in P, even though each system is regarded as “infertile.” N availability had a smaller effect on root dynamics than did P availability, suggesting that the simple dichotomy between fertile and infertile sites that is often evoked to explain plant characteristics may be unjustified.