Biosphere-atmosphere exchange of ammonia from a fertilized corn canopy in Central Illinois

Ammonia flux;ammonia;relaxed eddy accumulation;flux gradient;agriculture;fertilizer;air emissions

Nelson, Andrew Joseph

关键词: Ammonia flux;    ammonia;    relaxed eddy accumulation;    flux gradient;    agriculture;    fertilizer;    air emissions;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/101277/NELSON-DISSERTATION-2018.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Chemical fertilizer use is the primary source of anthropogenic atmospheric NH3 in Illinois. Atmospheric NH3 is a precursor to secondary particulate matter (PM) and it also contributes to eutrophication and soil acidification following wet or dry deposition. Uncertainty in NH3 emissions due to sparse measurements limit the ability of air quality models to predict secondary PM in the atmosphere and depositional loads of nitrogen to ecosystems. To address this challenge, it is necessary to improve understanding of NH3 emission through experimental measurements and development and evaluation of NH3 emission models. This research seeks to provide new measured NH3 fluxes through an experimental field campaign in central Illinois, to investigate dependence of NH3 flux on environmental conditions and field management practices, and to evaluate closure of field measurements and modeled NH3 fluxes. Two experimental field systems to quantify NH3 flux were deployed above a corn canopy in Central Illinois during the 2014 corn-growing season: a relaxed eddy accumulation (REA) system to quantify NH3 flux in 4 h intervals and a flux-gradient (FG) system to quantify 0.5 h averaged NH3 flux. A new method to better the quantify the effect of flux footprint on REA flux measurements was developed, resulting in a subset of 82 concurrent measurements. Mean NH3 flux was 205 +/- 300 ng m-2 s-1 with REA and 110 +/- 256 ng m-2 s-1 using FG for all concurrent measurements. REA and FG measurements were in agreement at a 0.95 confidence level. The FG method resolved NH3 emission peaks at 0.5 h averaging time that were otherwise un-observed with 4 h REA averaging. Two early-season emission periods were identified (DOY 130-132 and 140-143), where the timing and intensity of such emissions are attributed to a combination of the use of urease inhibitor and localized soil temperature and precipitation. Though general NH3 flux trends were similar between this study and those at non-Midwest sites, differences in short-term NH3 emission (i.e., hours to days) were found. REA measurements were also used to evaluate closure with DeNitrification DeComposition (DNDC) model predictions of NH3 flux. A new method to address practical issues of evaluating closure was developed to account for flux footprints extending outside the measurement site and differences in measurement and model temporal resolution. Modeled fluxes replicated experimental trends satisfactorily using association statistics during the first 33 days after fertilizer application when measured fluxes were to the atmosphere (association statistic, ra2 > 0.74). DNDC did not replicate measured trends as well during later time periods when depositional fluxes were measured (ra2 < 0.52). Evaluation of closure between model predictions and measurements identified a potential under-prediction of NH3 deposition within the DNDC model. These new measurements, and evaluation of closure with the DNDC model, are important to improving understanding of NH3 flux across varied ecosystems and for future upscaling of local measurements to regional scales using modeled parameterizations.