COS 75-9
Variation in wood nutrient stoichiometry along a soil fertility gradient in a Panamanian montane forest

Wednesday, August 7, 2013: 4:20 PM
L100F, Minneapolis Convention Center
Katherine D. Heineman, Program for Ecology, Evolution, and Conservation Biology, University of Illinois-Urbana Champaign, Urbana, IL
Benjamin L. Turner, Smithsonian Tropical Research Institute, Ancon, Panama
James W. Dalling, Plant Biology, University of Illinois at Urbana-Champaign, Urbana, IL

Tropical montane forests often display marked variation in soil nutrient availability and productivity, reflecting landscape-level variation in topography and geology. While variation in soil fertility is known to influence foliar chemistry and litter decomposition, less is known about how nutrient availability influences wood chemistry. Since woody stems comprise ~70% of aboveground forest biomass, the sensitivities of wood elemental ratios to natural soil fertility could have implications for estimates ecosystem nutrient pools. To address this issue, we quantified wood and foliar P, Ca, K, and Mg in 60 tree species across six 1-ha plots which span a steep soil fertility gradient in a Panamanian lower montane forest. We predicted that the concentration of each nutrient in wood would increase with its availability in the soil, and that wood chemistry should be a stronger indicator of soil nutrient availability than foliar chemistry because wood has lower structural nutrient requirements and is not susceptible to herbivory pressure when nutrient concentrations are high. 


Our results revealed that although average nutrient concentrations were an order of magnitude larger in leaves relative to wood, there was substantially more variation in species mean wood nutrient concentrations, with each elemental concentration varying by at least 25-fold, compared to a maximum variation of 7-fold in leaves. Abundance-weighted site wood Ca and K concentrations were strongly positively correlated with the availability of these elements in the soil, while foliar concentrations of Ca and K showed no relationship with soil nutrient availability. In contrast, foliar P more closely paralleled soil P availability than did woody P. However, wood P was negatively correlated with soil and leaf N:P ratios, suggesting that P can accumulate in woody tissues at sites where N is more limiting. Consequently, wood nutrient concentrations of tropical forest trees may be informative indicators of plant nutrient limitation and the bioavailability of nonlimiting elements in the soil.