A chemical transport model (CTM) is a type of computer numerical model which typically simulates atmospheric chemistry and may be used for air pollution forecasting.
While related general circulation models (GCMs) focus on simulating overall atmospheric dynamics (e.g. fluid and heat flows), a CTM instead focuses on the abundance and fluxes of one or more chemical species. Accordingly, a CTM must solve only the continuity equation for its species of interest while a GCM must solve all the primitive equations for the atmosphere. Hence, CTMs are potentially computationally much cheaper. However, a CTM will be expected to accurately represent the entire cycle for the species of interest, including emissions, fluxes (e.g. advection, convection), chemical production/loss, and deposition. The transport of species by the winds is computed by the model's advection scheme, which ideally conserves mass and avoids overshoots, numerical diffusion, and numerical dispersion. That being said, the tendency, especially as the cost of computing declines over time, is for GCMs to incorporate CTMs for species of special interest to climate dynamics, especially shorter-lived species such as nitrogen oxides and volatile organic compounds; this allows feedbacks from the CTM to the GCM's radiation calculations, and also allows the meteorological fields forcing the CTM to be updated at higher time resolution than may be practical in studies with offline CTMs.
CTMs may be classified according to their methodology and their species of interest, as well as more generic characteristics (e.g. dimensionality, global or regional, degree of resolution).
Jacob (1999) classifies CTMs as Eulerian/"box" or Lagrangian/"puff" models, depending on whether the CTM in question focuses on
An Eulerian CTM solves its continuity equations using a global/fixed frame of reference, while a Lagrangian CTM uses a local/moving frame of reference.