A. Sandu et al., BENCHMARKING STIFF ODE SOLVERS FOR ATMOSPHERIC CHEMISTRY PROBLEMS .1.IMPLICIT VS EXPLICIT, Atmospheric environment, 31(19), 1997, pp. 3149-3164
In many applications of atmospheric transport-chemistry problems, a ma
jor task is the numerical integration of the stiff systems of ordinary
differential equations describing the chemical transformations. This
paper presents a comprehensive numerical comparison between five dedic
ated explicit and four implicit solvers for a set of seven benchmark p
roblems from actual applications.-The implicit solvers use sparse matr
ix techniques to economize on the numerical linear algebra overhead. A
s a result they are often more efficient than the dedicated explicit o
nes, particularly when approximately two or more figures of accuracy a
re required. In most test cases, sparse RODAS, a Rosenbrock solver, ca
me out as most competitive in the 1% error region. Of the dedicated ex
plicit solvers, TWOSTEP came out as best. When less than 1% accuracy i
s aimed at, this solver performs very efficiently for tropospheric gas
-phase problems. However, like all other dedicated explicit solvers, i
t cannot efficiently deal with gas-liquid phase chemistry. The results
presented may constitute a guide for atmospheric modelers to select a
suitable integrator based on the type and dimension of their chemical
mechanism and on the desired level of accuracy. Furthermore, ale woul
d like to consider this paper an open invitation for other groups to a
dd new representative test problems to those described here and to ben
chmark their numerical algorithms in our standard computational enviro
nment. (C) 1997 Elsevier Science Ltd.