Foliar phosphorus concentration exerts stronger control of photosynthesis than does nitrogen across diverse woody species in Panama

Background/Question/Methods

Phosphorus (P) availability exerts a strong control on the productivity of lowland tropical forests. However, the observed variation in productivity across P gradients is poorly represented in terrestrial biosphere and Earth system models because the models lack routines to estimate P availability and its influence on productivity. While detailed biogeochemical models of P availability are currently being developed, the physiological effect of phosphorus limitation is often represented as a simple downscaling of GPP, if it is included at all. To inform the development of a physiological model of P limitation, we are gathering data to support empirical relationships between foliar P content and photosynthetic parameters. We measured the A-Ci relationship in leaves of many woody species, including lianas and understory, sub-canopy, and canopy trees at three locations in Panama. Measurements with LiCor 6400 systems were made from the ground in Gamboa and from canopy cranes in a seasonally dry forest, Parque Natural Metropolitano near Panama City, and a moist forest, Parque Nacional San Lorenzo near Colon. After determination of leaf area and dry mass, leaves were analyzed for nitrogen (N) and P content at the Smithsonian Tropical Research Institute.

Results/Conclusions

Foliar C:N:P ratios varied widely across species, with N:P ratios ranging from 7 to 42. Taking an N:P ratio > 16 as roughly indicative of P limitation, the majority of observations suggested P limitation, and few of the observations with lower N:P ratios, which might suggest N limitation, were from canopy trees. N and P concentrations declined with increasing leaf mass per unit area, but the relationships were weak. Although global analyses have supported a strong relationship between foliar N and photosynthesis, there was no clear indication of relationships between photosynthetic parameters and leaf N in our Panama data set. However, Vcmax, Jmax, and TPU were weakly correlated with leaf P concentration across all species. Vcmax tended to decline with increasing N:P ratio, suggesting greater control of photosynthesis by P than be N at these sites in Panama. Current carbon cycle models that rely on constant C:N:P ratios across all plant functional types, and those that ignore P altogether, may not work well in the tropics. Future studies that focus on understanding differences in nitrogen and phosphorous use efficiencies and their relationships to photosynthesis among co-occurring species may lead to better models of gross primary production of tropical forests.