In simulations, the computational method chosen to implement a model can critically affect quantitative results. Here I show that two reasonable implementations of a neighborhood model of plant growth produce predictions that differ quantitatively in several important respects. I conclude that for modelling forest canopies, pixel—based implementations should prove superior to implementations that are based on indices of neighborhood crowding.
No takes yet. Share an insight, caveat, or question.
Robert A. Armstrong (1993) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: