Grantee Research Project Results
Final Report: Exploration of uncertainty in the simulation of power plant plume chemistry
EPA Grant Number: R826239Title: Exploration of uncertainty in the simulation of power plant plume chemistry
Investigators: Gillani, Noor V. , Wu, Yuling
Institution: The University of Alabama in Huntsville
EPA Project Officer: Hahn, Intaek
Project Period: January 3, 1998 through January 2, 2001 (Extended to June 30, 2002)
Project Amount: $450,000
RFA: Ambient Air Quality (1997) RFA Text | Recipients Lists
Research Category: Air , Air Quality and Air Toxics
Objective:
The objectives of this research project were to: (1) conduct performance evaluation and process analysis of the leading gas-phase chemical mechanisms in the simulation of rural power plant plume (PPP) chemistry (for which emissions inventories were available from the U.S. Environmental Protection Agency [EPA], viz., CB4, and Regional Acid Deposition Model 2 [RADM2]); (2) evaluate SO2 and NOx emissions from the principal power plants under study, and the sensitivity of plume chemistry simulations to these emissions and to surface emissions of volatile organic compounds (VOCs) and NOx; (3) investigate the role of mixing in the simulation of rural PPP chemistry; (4) investigate the role of dry deposition (HNO3 and H2O2) in the simulation of PPP chemistry; and (5) conduct integrated assessment of the research results, aimed at recommending improvements in the simulation of PPP chemistry in operational air quality models (OAQMs).
Summary/Accomplishments (Outputs/Outcomes):
This research project focused mainly on exploration of the relative sensitivities of simulated gas-phase chemistry of large, rural PPPs to the main influencing variables. The focus in the chemistry simulations was on the relative production of ozone and NOz in rural PPPs, and the main dependent variables explored for this were the ozone production efficiency (OPE) and the yields of ozone and NOz ( Y(O3), Y(NOz)). The concept of yield, Y, ozone, and NOz species per unit of NOx emission has been introduced in this work as a new concept, has been applied extensively, and has proven to be very beneficial. The particular role of each of the following factors influencing OPE and Yi (i.e., sources of uncertainty in the simulation of PPP chemistry) has been investigated. These include: (1) chemical mechanism used to represent the chemistry; (2) relevant emissions, including those from the power plant and those from the surface in the region of plume transport; (3) meteorological and dynamical variables, including sunlight, plume lateral and vertical spread, the mean wind field, and subgrid-scale (turbulent) fluctuations in species concentrations; (4) initial conditions (species concentrations) in the regional background and in the PPP at the time of initialization of the plume; (5) dry deposition; and (6) spatial resolution of the Lagrangian Reactive Plume Model (LRPM). The important roles of aerosol and cloud processes were not investigated in this project.
Approach
The approach was to conduct diagnostic analyses based on: (1) three-dimensional field data of plume and background sampling from the 1995 Nashville-Middle Tennessee Field Study of the Southern Oxidants Study (SOS95), and (2) appropriate numerical model simulations. Three core field-study datasets were used; the TVA helicopter data of three-dimensional plume sampling, the met/chem data of the surface monitoring network, and the upper-air met soundings of the profiler network. The modeling studies have been carried out at two different spatial-temporal scales; at the plume scale, extending over mesoscale distances of 100 km or more, a combination of the University of Alabama-Huntsville (UAH) Plume Dynamics Model (PDM) and the UAH-LRPM were applied; at the finer scale of large turbulent eddies, and extending over a horizontal domain of up to 40 km, a Regional Atmospheric Modeling System-Large Eddy Simulation model (RAMS-LES) with detailed chemistry (LESchem) was applied. Guidance for specification of the regional background (meteorology, emissions, and chemistry) were derived from regional models (Models-3/Community Multiscale Air Quality [CMAQ] and Ozone Transport Assessment Group [OTAG]) and local observations.
Substantial models development activity was carried out in this project. Two independent versions of PDM have been developed from scratch for this project: the PDM-s(igma) and the PDM-p(article). The PDM serves as a preprocessor of LRPM, supplying plume dynamics information. The UAH-LRPM is a substantially upgraded version of the old Gillani-LRPM, now with generalized chemistry input (CB4 and RADM2-cis1 and -cis4 implemented), access to two chemical solvers quasi-steady state approximation method (QSSA), and sparse-matrix vectorized gear (SMV-GEAR) enhanced (arbitrary) vertical resolution, with three optional models of vertical diffusion-eddy diffusion (Kz), asymmetric convective mixing (ACM), and a hybrid of Kz and ACM (ACM2)-generalized capability to receive empirical or simulated initial conditions and boundary conditions (IC/BC), and a post-processor for intergrated reaction rate/mass balance (IRR/MB) process analysis. The upgraded UAH-LRPM currently is the most sophisticated and best evaluated LRPM around, and already has served as a building block of the plume-in-grid module of CMAQ and for multiplume modeling of industrial point-source plumes in Houston. For LES, a module for detailed chemistry simulation (arbitrary, externally supplied mechanism) was added to the RAMS-LES dynamical model, and the resulting model (LESchem) was implemented with the CB4 mechanism.
Diagnostic Applications. Diagnostic applications in this project included those using PDM/LRPM and those using LESchem. The LRPM applications particularly were extensive and included tailoring (calibration) of the model simulation to specific observation scenarios (the "standard" run), and performance of sensitivity analyses for each of the influencing variables identified above. In this report, detailed results of LRPM application are illustrated from a particular case study, the 0945 release of the Cumberland plume of July 16, 1995 (a Sunday). The sensitivity of plume chemistry to chemical mechanism was explored for CB4 and RADM2, both with a single-product isoprene mechanism, and also for RADM2, with two alternate isoprene mechanisms (one-product, -cis1, and four-product, -cis4). Our study, based on LESchem with detailed chemistry (CB4), is of a pioneering nature for PPPs, and presently is somewhat idealized. Nevertheless, the findings are quite illuminating, significant, and counter-intuitive. For example, for most important reactions of photochemistry, there is a slowdown of chemistry in the PPP, as expected, from the effects of turbulent concentration fluctuations (related to the segregation of plume NOx and background VOCs), but surprisingly, there is significant enhancement of HNO3 production from such fast chemistry, especially in plume edges. Interestingly, for the other reactions with rate deceleration, the maximum deceleration is not in the plume core, as might be expected, but rather it is in the plume edges. Overall, the decelerational effect on plume chemistry is most pronounced in the fresh plume, but the effect, though decreasing with plume age, can persist for hours, particularly for large NOx emissions.
We have summarized our main finding regarding the relative importance of the factors influencing plume ozone and NOz chemistry by classifying the factors into three levels by importance. Those in the highest level of importance are sunlight, power plant NOx source strength (QNOx), background reactivity, lateral mixing intensity, and U0 (the horizontal wind speed involved in the conversion of QNOx to C0). Of second level importance are: initialization of the background, choice of chemical mechanism, the turbulent chemistry of concentration fluctuations, plume vertical spread, reasonable specification of the horizontal wind field, and the spatial resolution of the plume. Of third level importance are dry deposition and the particular model of vertical diffusion.
We believe that this systematic and comprehensive study of the sensitivity of PPP gas-phase chemistry to influencing variables significantly has enhanced our understanding of quantitative aspects of plume chemistry and also has revealed important areas needing further work. The two most exciting areas for future work relate to LES studies and to the issue of NOy loss. The project also has developed first class modeling tools and analytical procedures for plume studies. A journal publication based on this work is in the review process, and a Ph.D. Thesis focused on the role of mixing processes in plume chemistry is in preparation.
Journal Articles on this Report : 1 Displayed | Download in RIS Format
Other project views: | All 2 publications | 1 publications in selected types | All 1 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Gillani NV, Meagher JF, Valente RJ, Imhoff RE, Tanner RL, Luria M. Relative production of ozone and nitrates in urban and rural power plant plumes. 1. Composite results based on data from 10 field measurement days. Journal of Geophysical Research-Atmospheres 1998;103(D17):22,593-22,616. |
R826239 (Final) |
Exit |
Supplemental Keywords:
air, ambient air, atmosphere, plume, power plant plume, PPP, rural plumes, Middle Tennessee, Nashville, Southern Oxidants Study, power plant emissions, ozone, nitrates, nitric acid, OH, H2O2, NOx, NOz, NOy, volatile organic compounds, VOCs, isoprene, ozone production efficiency, National Ambient Air Quality Standards, NAAQS, ozone yield, NOz yield, sources of uncertainty, simulation, power plant plume chemistry, gas-phase chemistry, turbulent chemistry, chemical mechanisms, CB4, RADM2, power plant emissions, surface emissions, plume transport, meteorology, regional background, boundary layer, plume dispersion, atmospheric mixing, plume-background mixing, turbulent diffusion, vertical diffusivity, asymmetric convective mixing, mixing parameterizations, plume dynamics model, sigma model, particle model, Lagrangian reactive plume model, large eddy simulation, LES, LES-with-chemistry, plume-in-grid, multiplume modeling, initial conditions/boundary conditions, IC/BC, dry deposition, NOy loss, spatial resolution., RFA, Scientific Discipline, Air, Toxics, particulate matter, air toxics, Environmental Chemistry, HAPS, Atmospheric Sciences, tropospheric ozone, ambient air quality, anthropogenic stresses, Lagrangian approach, air pollutants, diagnostic modeling, air quality models, ambient air, Sulfur dioxide, air quality criteria, ambient monitoring, chemical composition, aerosol/ cloud interactions, air pollution models, chemical kinetics, isoprene, hazardous air pollutants (HAPs), atmospheric monitoring, aerosol cloud, biogenic emissions, power plant plume chemistry , operational air quality simulation models, field measurements, atmospheric chemistry, convective boundary layerProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.