You are here:
Atmospheric inorganic nitrogen (N) deposition to Latin American cities: Comparison of field and Geos-Chem model estimates
Citation:
Ponette-Gonzalez, A., H. Lewis, B. Henderson, D. Carnelos, G. Pineiro, K. Weathers, AND D. Schwede. Atmospheric inorganic nitrogen (N) deposition to Latin American cities: Comparison of field and Geos-Chem model estimates. NADP 2020 Fall Meeting and Scientific Symposium, NA, October 28 - 30, 2020.
Impact/Purpose:
In Latin America, atmospheric deposition can be an important source of nitrogen for terrestrial ecosystems. Measurements of deposition in Latin America are scarce and modeling provides important spatial and temporal characterization of deposition which will be informative to control strategies in that region.
Description:
In Latin America, atmospheric deposition is a major, and sometimes the primary, vector of nitrogen (N) inputs to terrestrial ecosystems. In this study, our objectives were to: (1) estimate rates of atmospheric wet inorganic N (NH4-N + NO3-N) deposition to 16 Latin American cities using observed and modeled N deposition data; and (2) evaluate the performance of a global 3-D chemical transport model in predicting observed wet N deposition rates within and across sites. Published estimates of wet or bulk (hereafter wet) inorganic N deposition measured between 2006-2010 were compiled and annual wet, dry particulate, and dry inorganic N deposition to Latin America was modeled using the GEOS-Chem Chemical Transport Model. We evaluated the performance of GEOS-Chem for modeling inorganic wet N deposition to Latin American cities using spatial assessments, linear regression analysis, and the normalized mean bias. Mean annual inorganic N deposition for sites with in-situ measurements of both NH4-N and NO3-N ranged from 2.78-16.8 kg N ha-1 yr-1. Ammonium-N was generally the dominant form of N deposited, comprising 48-90% of observed inorganic N. Spatial patterns of observed N were similar to modeled estimates of wet N deposition. However, the model underestimated observed NH4-N deposition at most sites, while the model both over- and under-estimated NO3-N deposition. Use of modeled wet plus dry particulate N as a predictor of observed N did not improve R2 values but resulted in a slight improvement in normalized mean bias for NH4-N and inorganic N. GEOS-Chem captured variation in N deposition across sites better than variation among years at any given site. Our preliminary findings suggest that even at coarse spatial resolution, GEOS-Chem performs relatively well when compared to in-situ measured deposition in under-sampled urban areas of Latin America.
