Modeling impacts of atmospheric nitrogen deposition on terrestrial ecosystems:

1. Modeling impacts of atmospheric nitrogen deposition on terrestrial ecosystems: Implications for climate change J.J. Reyes1, J.C. Adam1, C. L. Tague2, J.S. Choate2 1. Civil and Environmental Engineering, Washington State University 2. Bren School, University of California – Santa Barbara GRS/GRC: July 2011 Bates College, Lewiston, ME Introduction Human activities are responsible for doubling the amount of reactive nitrogen (N) in the terrestrial biosphere, which has had numerous negative impacts on natural ecosystems, such as acidification, a decrease in biodiversity, and eutrophication (Gruber and Galloway, 2008). Perturbations in the nitrogen cycle also affect the carbon cycle and vice versa (Reich et al., 2006). Increasing carbon dioxide (CO2) levels from fossil fuel combustion may be causing a CO2 fertilization effect on forests, providing for a “carbon sink” that has helped to mediate some of the substantial additions of CO2 in the atmosphere (Gruber and Galloway, 2008). Similarly, additions of N in the terrestrial biosphere, such as through atmospheric deposition, may cause the same sequestration effect by inducing plant growth. CO2 fertilization can remain a carbon sink unabated, while forests fertilized by increased N may reach a saturation level and no longer remain carbon sinks (Gruber and Galloway, 2008; Aber et al., 1989). Furthermore, effects between different species of N, ammonium and nitrates, are important because deposition values are an order of magnitude different and their sources are also different (Holland et al., 1997). Changes in these N regimes must be investigated to assess how these perturbations affect C storage. Model Inputs Observed wet nitrogen deposition measurements from the National Atmospheric Deposition Program (NADP) were used as external inputs into RHESSys. The model can accept ADN as nitrates (NO3-) and/or ammonium (NH4+) species. As a default, RHESSys typically uses a constant value of ADN. However, a daily time-series dataset from NADP weekly measurements was created to better represent the temporal variation of and environmental impacts on ADN. Below are time-series data of wet nitrogen deposition for the HJ Andrews Long-Term Ecological Research (LTER) site in central Oregon (see Study Area) from the NADP. Ranges of chronic additions to the NADP time-series (left) from 1 kg/ha/yr to64 kg/ha/yr are created (not shown). By simulating future N regimes, possible levels of future C sequestration can be examined. Conclusions Using RHESSys and perturbed NADP data, nitrate additions result in potential long-term C storage over the 20-year period, while ammonium additions show no changes in the selected output data. While there are no changes from additions of ammonium, these results demonstrate that C continues to accumulate in forest vegetation and soils over time using a constant and present time-series data of N deposition. For nitrates, soil C increases at different rates for different perturbations. There are only minute increases in N uptake and plant productivity. Increases in these two factors would signify an increase in biomass and potential increased storage of C. For NPP, model estimates of the inter-annual variation in climate is a strong driver of variation in NPP. Model estimates of the impact of temporal variation in N deposition on NPP is negligible compared to this climate effect. Elevation (m) High: 1627.3 Objective The objective of this study is to examine the relative contribution of atmospheric deposition of nitrates and ammonium to potential carbon sequestration via chronic increases to past time-series data. Meters 0 2,500 5,000 7,500 Low: 411.5 Preliminary Results Chronic increases in nitrate deposition Constant N NADP Future Work Simulated data have not been field checked against observations. Future work involves evaluation of model behavior with soil C, soil N, and streamflow nitrate data. • Study Area • H. J. Andrews Long-Term Ecological Research (LTER) site • Extends to the Lookout Creek Watershed within the McKenzie Basin located in central Oregon • Located within the • Columbia River Basin • Abundant information on • streamflow and stream • chemistry concentrations • since 1970s BioEarth (http://www.cereo.wsu.edu/bioearth/) This research also represents the first step in developing BioEarth by looking at the one-way linkages between the atmosphere and terrestrial biosphere. Other linkages between components of the Earth system will also be considered. • Model Description • RHESSys is the Regional Hydro-ecologic Simulation System (Tague and Band, 2004). • It is a physical model that incorporates hydrology with relevant biogeochemical cycling within ecosystems. • Algorithms for carbon (C) and nitrogen (N) cycling within the soil and vegetation are adapted from the CENTURY and BIOME-BGC models. • The model was calibrated by comparing observed and simulated streamflow and adjusting two parameters, the saturated hydraulic conductivity (K) and decay of K with depth. • The simulations represent a hypothetical Douglas Fir forest stand. • The 20+ years simulated (1980-2003) represent the growing stage of the forest, as indicated by increasing trends (see Results). Therefore, results are interpreted within this context. Funding Source: WSU Laboratory of Atmospheric Research (LAR) Atmospheric Policy Trajectory (APT) Mean relative difference plots are shown above comparing plant C and soil C at the start and end of the simulation time period (1980 and 2004). (Note different percentage scales on y-axes). The percent relative difference was determined using the constant (default RHESSys) ADN value as the base For larger nitrate additions, soil plays a larger role in C storage. References Aber, J., et al. (1998). Bioscience, 48(11), 921-934. Gruber, N., & Galloway, J. N. (2008). Nature, 451(7176), 293-296. Reich, P. B., et al. (2006). Annual Review of Ecology Evolution and Systematics, 37, 611-636. Tague, C. L., & Band, L. E. (2004). Earth Interactions, 8.