With the advent of a big data revolution in air pollution research comes a necessity for refining existing statistical techniques and developing new ones that are robust, scalable, and informative. This work discusses the development of novel statistical methods applied to air pollution and can be broken into two parts: improving an existing method to analyze air pollution measurements taken at stationary monitors and developing new methods to analyze novel measurements taken with a mobile platform. In the first main chapter of this thesis, Dynamic Principal Component Analysis (DPCA) is discussed as an alternative to Principal Component Analysis (PCA) for analyzing time-dependent relationships between air pollutant variables measured at stationary monitors. DPCA is shown to offer several advantages over PCA, in that it generates a set of statistics which are time dependent and uncovers a wealth of information not readily apparent in a conventional static application. The second part of this work discusses development of new statistical methods to analyze a burgeoning and increasingly important method of measuring air pollution, which is through mobile monitoring. Mobile monitoring offers excellent spatial coverage at the expense of temporal coverage, making many conventional statistical techniques frequently employed by the atmospheric sciences community ill-suited for analysis of these measurements. Subsequent chapters discuss mobile monitoring as a measurement strategy, a novel algorithm to derive and quantify the background in mobile monitoring measurements, and another novel algorithm to identify and analyze detected plumes in time series. Results from these algorithms showcase the power mobile measurements have to inform stakeholders about the spatial variability of air pollution not obvious through the use of stationary monitors alone. It is hoped that through the refinement and development of these techniques and others that an important and exciting era of air pollution research can be ushered and its long-term mysteries solved.