[R-sig-dyn-mod] ode integration troubles

[Stephen Paul Ellner]

I'll add a suggestion that I've found very useful in situations like this: by using very small time steps, 
find out which state variable is the first to go awry (e.g., first to become negative, first to become NAN).
Usually the behavior you're seeing is because of trouble in one state variable, which then cascades to
others whose rate equations depend on it.

Steve 

Stephen P. Ellner (spe2 at cornell.edu)
Department of Ecology and Evolutionary Biology
E339 Corson Hall, Cornell University, Ithaca NY 14853-2701
Phone (607) 254-4221    FAX (607) 255-8088