Characterising Brazilian biomass burning emissions using WRF-Chem with MOSAIC sectional aerosol
Characterising Brazilian biomass burning emissions using WRF-Chem with MOSAIC sectional aerosol
复制标题
DOI：
10.5194/gmd-8-549-2015
复制
发表时间：
2015-01-01
影响因子：
5.1
通讯作者：
McFiggans, G.
中科院分区：
文献类型：
--
作者：
Archer-Nicholls, S.;Lowe, D.;McFiggans, G.
南美洲生物质燃烧分析（SAMBBA）野外考察活动在2012年旱季对气溶胶进行了详细的原位飞行测量，以表征生物质燃烧产生的气溶胶，并增进对其对天气和气候影响的理解。通过将一个分段气溶胶方案与烟羽抬升参数化相结合，对包含化学过程的天气研究与预报模型（WRF - Chem）进行了改进，以更好地模拟该地区生物质燃烧产生的气溶胶。使用了巴西生物质燃烧排放模型（3BEM）的火灾排放数据，这些数据是利用PREP - CHEM - SRC准备的，并映射到CBM - Z和MOSAIC物种。模型结果已根据遥感产品、AERONET站点以及SAMBBA考察活动中的四个飞行测量案例研究进行了评估。WRF - Chem预测在热带森林地区的高空（6 - 8千米）存在颗粒有机物气溶胶浓度升高的层（5 - 20微克/立方米），而飞行测量显示在2 - 4千米高度以上急剧下降。这种差异归因于烟羽抬升参数化高估了注入高度。利用2012年火灾季节的活跃火灾规模和燃烧面积的估计值对3BEM排放产品进行了修改，这减小了火灾规模。火灾排放的增强因子从1.3提高到5，以保持合理的气溶胶光学厚度（AOD）。较小的火灾规模降低了排放物的注入高度，但WRF - Chem仍显示在4 - 5千米高度之间气溶胶浓度升高。在塞拉多东部（类似热带草原）地区，模拟和测量的气溶胶浓度在大约4千米高度以上均下降。与MODIS卫星数据和AERONET站点相比，WRF - Chem在考察活动前半段对西部热带森林火灾地区的AOD量级（在0.3 - 1.5之间）模拟得较好，但在下半年降水较多时往往高估了AOD。在塞拉多东部地区，WRF - Chem往往低估AOD。在改进的排放情景下，东部的模拟气溶胶浓度较高。WRF - Chem中一次有机物与黑碳的比率通常在8 - 10之间。这低于西部飞行测量值（B734的四分位间距为11.6 - 15.7，B739为14.7 - 24.0），但与东部飞行的B742（8.1 - 10.4）相似。然而，单次散射反照率在西部飞行中接近测量值（模型中为0.87 - 0.89；飞行B734测量值为0.86 - 0.91，飞行B739测量值为0.81 - 0.95），但在东部飞行B742中过高（模型中为0.86 - 0.87，测量值为0.79 - 0.82）。这表明WRF - Chem在模拟气溶胶成分和光学特性计算方面都需要改进。
The South American Biomass Burning Analysis (SAMBBA) field campaign took detailed in situ flight measurements of aerosol during the 2012 dry season to characterise biomass burning aerosol and improve understanding of its impacts on weather and climate. Developments have been made to the Weather Research and Forecast model with chemistry (WRF-Chem) model to improve the representation of biomass burning aerosol in the region, by coupling a sectional aerosol scheme to the plume-rise parameterisation. Brazilian Biomass Burning Emissions Model (3BEM) fire emissions are used, prepared using PREP-CHEM-SRC, and mapped to CBM-Z and MOSAIC species. Model results have been evaluated against remote sensing products, AERONET sites, and four case studies of flight measurements from the SAMBBA campaign.WRF-Chem predicted layers of elevated aerosol loadings (5-20 mu g sm(-3)) of particulate organic matter at high altitude (6-8 km) over tropical forest regions, while flight measurements showed a sharp decrease above 2-4 km altitude. This difference was attributed to the plume-rise parameterisation overestimating injection height. The 3BEM emissions product was modified using estimates of active fire size and burned area for the 2012 fire season, which reduced the fire size. The enhancement factor for fire emissions was increased from 1.3 to 5 to retain reasonable aerosol optical depths (AODs). The smaller fire size lowered the injection height of the emissions, but WRF-Chem still showed elevated aerosol loadings between 4-5 km altitude. Over eastern cerrado (savannah-like) regions, both modelled and measured aerosol loadings decreased above approximately 4 km altitude.Compared with MODIS satellite data and AERONET sites, WRF-Chem represented AOD magnitude well (between 0.3-1.5) over western tropical forest fire regions in the first half of the campaign, but tended to over-predict them in the second half, when precipitation was more significant. Over eastern cerrado regions, WRF-Chem tended to under-predict AODs. Modelled aerosol loadings in the east were higher in the modified emission scenario. The primary organic matter to black carbon ratio was typically between 8-10 in WRF-Chem. This was lower than the western flight measurements (interquartile range of 11.6-15.7 in B734, 14.7-24.0 in B739), but similar to the eastern flight B742 (8.1-10.4). However, single scattering albedo was close to measured over the western flights (0.87-0.89 in model; 0.86-0.91 in flight B734, and 0.81-0.95 in flight B739 measurements) but too high over the eastern flight B742 (0.86-0.87 in model, 0.79-0.82 in measurements). This suggests that improvements are needed to both modelled aerosol composition and optical properties calculations in WRF-Chem.