channels. Elevations in Ca²⁺ concentration after intracellular release through IP₃ receptors (IP₃R) can either propagate in the form of waves spreading through the entire cell or produce spatially localized puffs. The appearance of waves and puffs is thought to implicate random initial openings of one or a few channels and subsequent activation of neighboring channels because of an “autocatalytic” feedback. It is much less clear, however, what determines the further time course of release, particularly since the lifetime is very different for waves (several seconds) and puffs (around 100 ms). Here we study the lifetime of Ca²⁺ signals and their dependence on residual Ca²⁺ microdomains. Our general idea is that Ca²⁺ microdomains are dynamical and mediate the effect of other physiological processes. Specifically, we focus on the mechanism by which Ca²⁺ binding proteins (buffers) alter the lifetime of Ca²⁺ signals. We use stochastic simulations of channel gating coupled to a coarse-grained description for the Ca²⁺ concentration. To describe the Ca²⁺ concentration in a phenomenological way, we here introduce a differential equation, which reflects the buffer characteristics by a few effective parameters. This non-stationary model for microdomains gives deep insight into the dynamical differences between puffs and waves. It provides a novel explanation for the different lifetimes of puffs and waves and suggests that puffs are terminated by Ca²⁺ inhibition while IP₃ unbinding is responsible for termination of waves. Thus our analysis hints at an additional role of IP3 and shows how cells can make use of the full complexity in IP₃R gating behavior to achieve different signals.