Variation in Canopy Litterfall Along a Precipitation and Soil Fertility Gradient in a Panamanian Lower Montane Forest

Fertilization experiments in tropical forests have shown that litterfall increases in response to the addition of one or more soil nutrients. However, the relationship between soil nutrient availability and litterfall is poorly defined along natural soil fertility gradients, especially in tropical montane forests. Here, we measured litterfall for two years in five lower montane 1-ha plots spanning a soil fertility and precipitation gradient in lower montane forest at Fortuna, Panama. Litterfall was also measured in a concurrent nitrogen fertilization experiment at one site. Repeated-measures ANOVA was used to test for site (or treatment), year, and season effects on vegetative, reproductive and total litterfall. We predicted that total litterfall, and the ratio of reproductive to leaf litterfall, would increase with nutrient availability along the fertility gradient, and in response to nitrogen addition. We found that total annual litterfall varied substantially among 1-ha plots (4.78Mg/ha/yr to 7.96Mg/ha/yr), and all but the most aseasonal plot showed significant seasonality in litterfall. However, litterfall accumulation did not track soil nutrient availability; instead forest growing on relatively infertile soil, but dominated by an ectomycorrhizal tree species, had the highest total litterfall accumulation. In the fertilization plots, significantly more total litter fell in nitrogen addition relative to control plots, but this increase in response to nitrogen (13%) was small compared to variation observed among 1-ha plots. These results suggest that while litterfall at Fortuna is nutrient-limited, compositional and functional turnover along the fertility gradient obscure any direct relationship between soil resource availability and canopy productivity. Resumen Se demostro con experimentos de fertilizacion en bosques tropicales incrementos en la caida de hojarasca asociados con la adicion de nutrientes; empero, se desconoce la relacion entre la disponibilidad de nutrientes del suelo con la produccion de hojarasca en gradientes naturales de fertilidad del suelo, especialmente en bosques montanos tropicales. Medimos la produccion de hojarasca durante dos anos en cinco parcelas (1-ha) en bosques montanos bajos, sobre gradientes de fertilidad y precipitacion y en un experimento paralelo de fertilizacion de nitrogeno en la Reserva Forestal Fortuna, Panama. Predijimos incrementos de hojarasca total y reproductiva con la disponibilidad de nutrientes en el gradiente de fertilidad del suelo y en respuesta a la adicion de nitrogeno (13%). Encontramos que la hojarasca total varia sustancialmente entre parcelas (4.78 Mg/ha/yr hasta 7.96 Mg/ha/yr) y que en todas las parcelas, excepto la menos estacional, hay estacionalidad significativa en la cantidad de hojarasca; pero, la acumulacion de hojarasca no coincide con la disponibilidad de nutrientes en el suelo, porque el bosque sobre suelo poco fertil con dominio de una especie ectomicorricica, presento mayor acumulacion de hojarasca. Las parcelas con fertilizacion tenian significativamente mas hojarasca que las control; pero, el incremento en hojarasca en respuesta a la adicion de nitrogeno fue menor respecto a la variacion entre parcelas. Los resultados sugieren que mientras la produccion de hojarasca en Fortuna es limitada por nutrientes, la variacion en la composicion de especies a traves de gradientes de fertilidad ocultan cualquier relacion directa entre la disponibilidad de nutrientes del suelo y la productividad del dosel.