Disentangling the role of climate, atmospheric CO2, nitrogen and sulphur deposition on changes in tree WUE and nitrogen availability for UK forests
sources of FUNDING
Human activities in the last century have substantially increased the concentrations of reactive nitrogen (N) in the atmosphere and their deposition back to the biosphere, including terrestrial and aquatic ecosystems. How does the extra N input from the atmosphere affect the trade-off between between CO2 assimilation through photosynthesis and water loss during transpiration (i.e., water-use efficiency, WUE), and alter N availability?
Most of our understanding on the effects of N deposition on tree WUE relies on soil manipulation experiments, where N was added into the soil, commonly as ammonium nitrate (e.g., the +20 years experiment at Harvard forest in the US). However, this approach implies the atmospheric N reaches the soil, thus contributing mostly to biogeochemical processes, ruling out any possible interaction between N deposition and tree canopies. Is this the case? Only a few experiments (back in 2009) had examined the contribution of atmospheric N, by spraying or misting above tree canopies: the experiment at Deepsyke forest, in the United Kingdom and at Howland Forest, in the USA. Between the two sites, the Deepsyke experiment was unique in providing the opportunity to evaluate the effects of frequent (450 applications during the growing season) aerial N and sulfur (S) misting onto the canopy of a Sitka spruce forest for a relatively long period (between 5–and 8 years). During my postdoc the University of Edinburgh (2009-2012) working alongside Maurizio Mencuccini, I took advantage of this experimental site and I started a collaboration with scientists at CEH leading the Deepsyke experiment (Lucy Sheppard, Mark Sutton and Peter Levy). I applied the triple isotope approach in tree rings (δ13C, δ18O, and δ15N) to investigate the effects of treatments on changes in tree intrinsic water-use efficiency (i.e., iWUE, the ratio between photosynthesis, A, and stomatal conductance, gs) and N availability. We found that N applications, with or without S, increased the ratio between A and gs in favour of A, thus supporting a fertilizer effect of added N. However, while we observed a pulse effect in the case of N+S, N alone had a slow but long-lasting positive effect on iWUE. After the treatments ceased, the trees quickly adjusted to the reductions of N deposition, but not to the reduction in S deposition S deposition which had a negative effect on iWUE by reducing A. This result suggests that even though S deposition have been significantly reduced across Europe, forest might still be recovering from the high exposure to acid deposition. Finally, by comparing tree-ring δ15N values measured at our experimental plots with those reported in previous studies were soil manipulation experiments were carried out, we found opposite results: δ15N in tree rings becomes enriched when the cumulative added N to the soil increases, while in the case of canopy N applications δ15N tended to be more negative. This result provides first evidences of differences in the N dynamics captured by the two approaches: speeding up of biogeochemical processes and opening of N cycling with consequent losses from the system in the case of soil manipulation. Whereas N retained in the system, including tree canopies, in the case of aerial spraying.
One of the limitations of manipulation experiments, however, is that they may only reveal treatment-pulse responses, associated with the doses of N applied, normally higher than the N input from ambient N deposition. Moreover, they miss delayed feedbacks related to 1) gradual changes in the atmospheric N input and 2) the interaction between N deposition and climate variability and finally 3) they are site and species-specific making the generalization of results difficult. Comparisons across climate and N deposition gradients and the combination of multiple isotopes offer a useful approach for advancing understanding of spatial and temporal patterns of variations in WUE and its main drivers. In collaboration with colleagues at the Forest Research (Elena Valnguelova, Rona Pitman, Sue Benham, James Morison, Mike Perks) I explored changes in δ13C, δ18O and δ15N measured in tree rings for the last three decades (1980-2010), for four species: Pinus sylvestris L., Picea sitchensis Bong. Carr. Quercus robur L. and Fagus sylvatica L. at twelve UK forests along a climate and atmospheric N and S deposition gradients. Sites were included in UK and European long-term monitoring program (i.e., ICP forests). We assessed whether changes in nitrogen and sulfur deposition and atmospheric CO2 affects WUE and N availability after accounting for spatial and temporal climate variability (Guerrieri et al. under revision). Compared to other studies where modeled atmospheric deposition data were considered, we used measured N and S deposition at the site, as derived from concentrations of nitrogen and sulfur compounds in rainfall and throughfall water. Results are included in a paper under revision, so stay tuned!