The reduction of chemical networks

A way to model chemical processes in space is to use special databases of chemical reactions, those expected to proceed efficiently under cosmic conditions. The relevant astrochemical databases (UMIST $90$, $95$, $99$, Ohio New Standard Database, etc.) contain hundreds of species and thousands of reactions (see Millar et al. [3,4], Terzieva & Herbst [5], Le Teuff et al. [6]).

While one can only wonder if these databases contain enough correct information on all possible reactions and species to accurately reproduce observed abundances of any particular species, the other question to ask is if all this information is always needed.

Recently, Ruffle et al. [7] have shown that it is possible to quantify this task by introducing a certain criterion that would allow, in each particular case, to indicate abundances of which species have to be followed in order to have a reasonable estimate for an abundance of species under investigation.

We perform a similar analysis of the UMIST $95$ chemical reaction database in order to find if it is possible to reduce the number of species and/or reactions in the chemical network used in the self-consistent modelling of the chemical and dynamical evolution of protostellar objects.