Sensitivity analysis of simulated SOA loadings using a variance‐based statistical approach. (8th April 2016)

• Record Type: Journal Article
• Authors: Shrivastava, Manish; Zhao, Chun; Easter, Richard C.; Qian, Yun; Zelenyuk, Alla; Fast, Jerome D.; Liu, Ying; Zhang, Qi; Guenther, Alex
• Is Part Of: Journal of advances in modeling earth systems. Volume 8:Number 2(2016)
• Page Start: 499
• Page End: 519
• Publication Date: 2016-04-08
• Subjects: uncertainty quantification -- sensitivity analysis -- secondary organic aerosols; Geological modeling -- Periodicals; Climatology -- Periodicals; Geochemical modeling -- Periodicals; 551.5011
• DOI: 10.1002/2015MS000554
• Languages: English
• ISSNs: 1942-2466

Abstract

Abstract: We investigate the sensitivity of secondary organic aerosol (SOA) loadings simulated by a regional chemical transport model to seven selected model parameters using a modified volatility basis‐set (VBS) approach: four involving emissions of anthropogenic and biogenic volatile organic compounds, anthropogenic semivolatile and intermediate volatility organics (SIVOCs), and NOx ; two involving dry deposition of SOA precursor gases, and one involving particle‐phase transformation of SOA to low volatility. We adopt a quasi‐Monte Carlo sampling approach to effectively sample the high‐dimensional parameter space, and perform a 250 member ensemble of simulations using a regional model, accounting for some of the latest advances in SOA treatments based on our recent work. We then conduct a variance‐based sensitivity analysis using the generalized linear model method to study the responses of simulated SOA loadings to the model parameters. Analysis of SOA variance from all 250 simulations shows that the volatility transformation parameter, which controls whether or not SOA that starts as semivolatile is rapidly transformed to nonvolatile SOA by particle‐phase processes such as oligomerization and/or accretion, is the dominant contributor to variance of simulated surface‐level daytime SOA (65% domain average contribution). We also split the simulations into two subsets of 125 each, depending on whether the volatility transformation is turned on/off. For each subset, the SOA variances are dominated by the parameters involving biogenic VOC and anthropogenic SIVOC emissions. Furthermore, biogenic VOC emissions have a larger contribution to SOA variance when the SOA transformation to nonvolatile is on, while anthropogenic SIVOC emissions have a larger contribution when the transformation is off. NOx contributes less than 4.3% to SOA variance, and this low contribution is mainly attributed to dominance of intermediate to high NOx conditions throughout the simulated domain. However, we note that SOA yields have a more complex nonlinear dependence on NOx levels, which needs to be addressed by more integrated model‐measurement approaches focused on gaining a better process‐level understanding of anthropogenic‐biogenic interactions. The two parameters related to dry deposition of SOA precursor gases also have very low contributions to SOA variance. This study highlights the large sensitivity of SOA loadings to the particle‐phase processes such as oligomerization that rapidly cause large decrease in the volatility of SOA, which is neglected in most previous models. Key Points: SOA is most sensitive to volatility transformation due to particle‐phase processes such as oligomerization Particle‐phase volatility transformation changes dominant SOA source contributions Variance‐based sensitivity analysis is a promising approach to rank SOA parameters