Does Nitrogen Constrain Carbon Cycling, or Does Carbon Input Stimulate Nitrogen Cycling? 1

The concept that nitrogen (N) availability can limit plant productivity is well established based on (1) N fertilization that stimulates productivity and (2) increases in productivity along gradients of soil fertility. Nitrogen limitations to plant productivity are regulated by processes such as mineralization, immobilization, and plant physiological adjustments. However, this productioncentric perspective might not fully explain patterns in carbon (C) sequestration in terrestrial ecosystems. Carbon sequestration involves both plant and soil pools. The plant pool, which is the main concern of production research, can be much smaller than the soil pool. To quantify terrestrial C sequestration, therefore, we have to develop an ecosystem perspective to examine how C and N interact in both plant and soil pools. Due to fossil fuel burning and deforestation, atmospheric CO2 concentration has increased by approximately 35% since the Industrial Revolution. In general, elevated CO2 enhances photosynthesis and stimulates initial C sequestration in terrestrial ecosystems. How sustainable the CO2-induced C sequestration can be depends, in part, on ecosystem N availability and supply. Thus, the interdependence of C and N cycles is an issue that is not only interesting to ecologists, but also has important implications for global change policy. Increased C influx into an ecosystem under elevated CO2 generally requires more N to support plant growth than is required at ambient CO2 and, in turn, sequesters N into long-lived plant biomass and soil organic matter pools. This N sequestration can decrease soil N availability for plant uptake and lead to progressive N limitation (PNL) over time. The PNL hypothesis states that N sequestration in long-term organic matter pools will, without new N input and/or decreases in N losses, lead to a decline in mineral N availability over time at elevated CO2 compared to ambient CO2. On the other hand, increased plant N demand and/or sequestration could induce changes in N supply. When elevated CO2 increases N use efficiency (NUE) and stimulates N transfer from the soil organic pools with narrow C:N ratios to plants with broad C:N ratios, PNL may be delayed. If additional C input at elevated CO2 stimulates capital gain of N through fixation, decreased losses, increased forage for soil N, or any combinations of them, PNL may not occur. If it does, CO2-induced C sequestration in ecosystems declines over time. In short, N will constrain C sequestration over time unless additional C input at elevated CO2 stimulates N gain in ecosystems. This Special Feature consists of six papers that examine various aspects of PNL against field data collected from ecosystems that have been exposed to elevated CO2 treatments. The first two papers show sustained CO2 stimulation of net primary production (NPP) in forest ecosystems. Norby and Iversen present data from a sweetgum forest stand in Oak Ridge, Tennessee, that has been exposed to free-air CO2 enrichment (FACE) for six years. The sustained CO2 stimulation of NPP was associated primarily with increased N uptake, since NUE did not change significantly under elevated CO2. Sufficient N supply from soil at Oak Ridge may help delay or even avoid PNL as elevated CO2 substantially stimulated root growth to explore N sources in deeper soil layers. At the Duke Forest FACE site, Finzi and colleagues demonstrate that the CO2 stimulation of NPP was sustained at 18–24% during the first six years of the experiment. Sustained NPP stimulation occurred together with significantly more N uptake by trees and higher NUE at elevated than at ambient CO2. Their mass balance analysis shows that significantly more N accumulated at elevated CO2 in plants and in forest floor litter. The forest ecosystem accrued N capital at an average rate of 12 g N·m22·yr21, perhaps due to N uptake from deeper in the soil profile.