Four different literature parameterizations for the formation and evolution of urban secondary organic aerosol (SOA) frequently used in 3-D models are evaluated using a 0-D box model representing the Los Angeles metropolitan region during the California Research at the Nexus of Air Quality and Climate Change (CalNex) 2010 campaign. We constrain the model predictions with measurements from several platforms and compare predictions with particle- and gas-phase observations from the CalNex Pasadena ground site. That site provides a unique opportunity to study aerosol formation close to anthropogenic emission sources with limited recirculation. The model SOA that formed only from the oxidation of VOCs (V-SOA) is insufficient to explain the observed SOA concentrations, even when using SOA parameterizations with multi-generation oxidation that produce much higher yields than have been observed in chamber experiments, or when increasing yields to their upper limit estimates accounting for recently reported losses of vapors to chamber walls. The Community Multiscale Air Quality (WRF-CMAQ) model (version 5.0.1) provides excellent predictions of secondary inorganic particle species but underestimates the observed SOA mass by a factor of 25 when an older VOC-only parameterization is used, which is consistent with many previous model–measurement comparisons for pre-2007 anthropogenic SOA modules in urban areas. Including SOA from primary semi-volatile and intermediate-volatility organic compounds (P-S/IVOCs) following the parameterizations of Robinson et al. (2007), Grieshop et al. (2009), or Pye and Seinfeld (2010) improves model–measurement agreement for mass concentration. The results from the three parameterizations show large differences (e.g., a factor of 3 in SOA mass) and are not well constrained, underscoring the current uncertainties in this area. Our results strongly suggest that other precursors besides VOCs, such as P-S/IVOCs, are needed to explain the observed SOA concentrations in Pasadena. All the recent parameterizations overpredict urban SOA formation at long photochemical ages (≈ 3 days) compared to observations from multiple sites, which can lead to problems in regional and especially global modeling. However, reducing IVOC emissions by one-half in the model to better match recent IVOC measurements improves SOA predictions at these long photochemical ages. Among the explicitly modeled VOCs, the precursor compounds that contribute the greatest SOA mass are methylbenzenes. Measured polycyclic aromatic hydrocarbons (naphthalenes) contribute 0.7% of the modeled SOA mass. The amounts of SOA mass from diesel vehicles, gasoline vehicles, and cooking emissions are estimated to be 16–27, 35–61, and 19–35%, respectively, depending on the parameterization used, which is consistent with the observed fossil fraction of urban SOA, 71(±3) %. The relative contribution of each source is uncertain by almost a factor of 2 depending on the parameterization used. In-basin biogenic VOCs are predicted to contribute only a few percent to SOA. A regional SOA background of approximately 2.1 μg m−3 is also present due to the long-distance transport of highly aged OA, likely with a substantial contribution from regional biogenic SOA. The percentage of SOA from diesel vehicle emissions is the same, within the estimated uncertainty, as reported in previous work that analyzed the weekly cycles in OA concentrations (Bahreini et al., 2012; Hayes et al., 2013). However, the modeling work presented here suggests a strong anthropogenic source of modern carbon in SOA, due to cooking emissions, which was not accounted for in those previous studies and which is higher on weekends. Lastly, this work adapts a simple two-parameter model to predict SOA concentration and O/C from urban emissions. This model successfully predicts SOA concentration, and the optimal parameter combination is very similar to that found for Mexico City. This approach provides a computationally inexpensive method for predicting urban SOA in global and climate models. We estimate pollution SOA to account for 26 Tg yr−1 of SOA globally, or 17% of global SOA, one-third of which is likely to be non-fossil.