期刊
ATMOSPHERIC ENVIRONMENT
卷 131, 期 -, 页码 371-381出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.atmosenv.2016.02.020
关键词
NOx emissions; Data assimilation; Bayesian; CMAQ DDM; Remote sensing
资金
- TCEQ Air Quality Research Program project [AQRP 14-014]
Reliable emission inventories are key to precisely model air pollutant concentrations. The relatively large reduction in NOx emissions that is well corroborated by satellite and in-situ observations over southeast Texas has resulted in discrepancies between observations and regional model simulations based on the National Emission Inventory (NEI) provided every three years in U.S. In this study, a Bayesian inversion of OMI tropospheric NO2 is conducted to update anthropogenic sources of NEI-2011 and soil-biogenic sources from BEIS3 (Biogenic Emission Inventory System version 3) over southeast Texas and west Louisiana during the 2013 DISCOVER-AQ Texas campaign. Results reveal that influences of the a priori profile used in OMI NO2 retrieval play a significant role in inconsistencies between model and satellite observations, which should be mitigated. A posteriori emissions are produced using the regional Community Multiscale Air Quality (CMAQ) model associated with Decoupled Direct Method (DDM) sensitivity analysis. The inverse estimate suggests a reduction in area (44%), mobile (30%), and point sources (60%) in high NOx areas (ENOx > 0.2 mol/s), and an increase in soil (similar to 52%) and area emissions (37%) in low NOx regions (ENOx < 0.02 mol/s). The reductions in anthropogenic sources in high NOx regions are attributed to both uncertainty of the priori and emissions policies, while increases in area and soil-biogenic emissions more likely resulted from under-estimation of ships emissions, and the Yienger-Levy scheme used in BEIS respectively. In order to validate the accuracy of updated NOx emissions, CMAQ simulation was performed and results were evaluated with independent surface NO2 measurements. Comparing to surface monitoring sites, we find improvements (before and after inverse modeling) for MB (1.95, -0.30 ppbv), MAB (3.65, 2.60 ppbv), RMSE (6.13, 4.37 ppbv), correlation (0.68, 0.69), and IOA (0.76, 0.82). The largest improvement is seen for morning time surface NO2 for which RMSE and MB are reduced considerably by similar to 60% (5.05 to 2.07 ppbv) and 109% (3.49 to -0.3 ppbv). However, under prediction of model NO2 in Houston on 09125-09/26 likely resulting from unprecedented local NOx sources from the Houston Ship Channel (HSC) becomes more evident. Overall, results show that use of OMI tropospheric NO2 columns can substantially reduce the uncertainty of bottom-up emissions for a regional study which should be extended to larger domains (e.g. entire U.S. or North America). (C) 2016 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据