Growth enhancement of Picea abies trees under long-term, low-dose N addition is due to morphological more than to physiological changes

Human activities have drastically increased nitrogen (N) inputs into natural and near-natural terrestrial ecosystems such that critical loads are now being exceeded in many regions of the world. This implies that these ecosystems are shifting from natural N limitation to eutrophication or even N saturation. This process is expected to modify the growth of forests and thus, along with management, to affect their carbon (C) sequestration. However, knowledge of the physiological mechanisms underlying tree response to N inputs, especially in the long term, is still lacking. In this study, we used tree-ring patterns and a dual stable isotope approach (D¹³C and D¹⁸O) to investigate tree growth responses and the underlying physiological reactions in a long-term, low-dose N addition experiment (+23kgNha^-1a ^-1). This experiment has been conducted for 14 years in a mountain Picea abies (L.) Karst. forest in Alptal, Switzerland, using a paired-catchment design. Tree stem C sequestration increased by ∼22, with an N use efficiency (NUE) of ca. 8kg additional C in tree stems per kg of N added. Neither earlywood nor latewood D¹³C values changed significantly compared with the control, indicating that the intrinsic water use efficiency (WUE _i) (A/g_s) did not change due to N addition. Further, the isotopic signal of D¹⁸O in early- and latewood showed no significant response to the treatment, indicating that neither stomatal conductance nor leaf-level photosynthesis changed significantly. Foliar analyses showed that needle N concentration significantly increased in the fourth to seventh treatment year, accompanied by increased dry mass and area per needle, and by increased tree height growth. Later, N concentration and height growth returned to nearly background values, while dry mass and area per needle remained high. Our results support the hypothesis that enhanced stem growth caused by N addition is mainly due to an increased leaf area index (LAI). Higher LAI implies that more photosynthetically active radiation is absorbed and therefore canopy-level photosynthesis is increased. We conclude that models assuming that N deposition increases tree growth through higher leaf-level photosynthesis may be mechanistically inaccurate, at least in forest canopies that are not (yet) completely closed.