Early evidence from free-air carbon dioxide (CO2) enrichment (FACE) experiments suggested that stimulation of net primary productivity (NPP) was a general property of temperate forests. More recently however, a discrepancy in NPP responses has emerged: while some forests show a sustained enhancement of NPP for up to a decade, others show only a transient response. These findings challenge results from global land models, which commonly predict persistent increases in NPP with rising CO2. This discrepancy raises the critical question of whether and how forest response depends upon secondary site factors such as nutrient availability and historical land use. Here we explore how land-use history, successional stage, atmospheric nitrogen (N) deposition, and climatic conditions interact to generate the observed forest NPP responses to elevated CO2. We compare a series of CO2 enrichment and N addition experiments using the Princeton-Geophysical Fluid Dynamics Laboratory (GFDL) global carbon-nitrogen (C-N) model LM3V-N to understand the general rules that shape the response of NPP to elevated CO2.
Results/Conclusions
Results from one of the forest FACE sites show a stronger stimulation of modeled NPP in a very young stand on former agricultural land compared to an old growth forest. Our findings indicate that land-use history, successional stage, climate and N availability can significantly affect the NPP response to a step increase in atmospheric CO2 (550 ppm). In line with experimental evidence, we found that response in NPP declined over time due to progressively greater N limitation as stand age increased. At five FACE locations (Duke, ORNL, web, POP-EURO, and Aspen-FACE), our model predicts a variable response to CO2, due to interactions between N and water availability within the context of land-use history. In C-only models, the stimulation in NPP was driven strongly by the physiological response of trees to rising CO2. We demonstrate here that inclusion of feedbacks between the C and the N cycle allows to account for potential biogeochemical constraints. We suggest that these interactions are critical for capturing spatial and temporal variations in the response of NPP to elevated CO2.