Resolving nitrogen gaseous pathways in the atmosphere-plant-microbial-soil continuum in the NOAA/GFDL Earth System Modeling Framework

Representing plant-microbe-soil organic matter interactions and their coupling with land surface processes are critical to understanding of ecosystem responses to climate change. More specifically, microbes play an important role in the nitrogen (N) cycle by providing acquisition pathways for plants and overcoming N limitation through mycorrhizal symbiosis and bacterial fixation. Even though biological nitrogen fixation acts as a primary N source for the organisms, ecosystem N availability is still strongly affected by N losses, including atmospheric volatilization.

One of the major challenges to accurately representing N availability in Earth System Models (ESM) is the representation of the atmospheric losses that are not necessarily controlled by the organisms. For instance, the conversion of soil ammonium into gaseous ammonia (i.e., volatilization) is driven by ambient environmental conditions and not directly controlled by the biological demand of plants and soil microbes. Thus, rapid losses of N via volatilization (e.g., after precipitation events) could induce feedback on soil microbial activity and plant growth by impeding biological assimilation.

Even though the representation of ammonia emissions is progressively integrated into ESMs, the focus has been mainly on parameterizing losses from agricultural or managed ecosystems. However, ammonia volatilization from natural soils occurs worldwide and can reach 9 TgN/yr, a non-negligible source, especially in alkaline drylands. Up to now, no proper representation of emissions of ammonia, applicable to unmanaged lands, has been included in ESMs and challenged by observations. In the future, these emissions are likely to follow the rising trends of nitrogen deposition and increasing precipitation due to climate change.

Here we describe a mechanistic parameterization of ammonia emissions in natural ecosystems with explicit treatment of microbes and vegetation dynamics in the fully integrated terrestrial component of the GFDL ESM, LM4.2-GIMICS-N. We apply observational constraints, including measurements of soil 15N isotope and estimates of nitrogen fluxes (BNF, nitrification, mineralization, and ammonia exchange) at different sites to reduce uncertainty in the model simulations. Finally, we examine the main drivers of ammonia volatilization across various ecosystems by considering aridity, soil pH, and nitrogen deposition as well as the key environmental conditions such as precipitation, temperature, and soil moisture.