Online coverage path planning is a canonical multi-robot task, where the objective is to minimize the time it takes for robots to visit every point in an unknown area. Two general major approaches have been explored in the literature: a stigmergic approach, inspired by ant behavior, relies on active marking of the environment. In contrast, the collaborative approach relies instead on localization, memory of positions, and global communications. In this paper, we report on a new approach, inspired by territorial bird chirping, which borrows from both previous approaches: it relies on localization and memory, but not on global communications. We provide a detailed analytic and empirical evaluation of this model.