Alternative Dynamic Regimes for Canopy-Forming Kelp: A Variant on Density-Vague Population Regulation

Different populations within at least some species show a range of qualitatively different population dynamics. Motivated by such an observation for giant kelp, we ask if a single set of underlying biological processes could be responsible for the different types of demography. We develop a family of models for canopy-forming marine plants, based on the idea that recruitment can occur under appropriate conditions of temperature, bottom irradiance, and other unknown or unpredictable factors represented through a two-state birth-and-death process. The properties of the simplest model are investigated in some detail. In a constant environment, we can obtain a stable equilibrium set by the level of adult mortality, or we can obtain a limit cycle about that population density for which self-shading is sufficiently strong to inhibit recruitment. Seasonal variation in temperature or average daily irradiance may have no effect, may cause simple cycles in population density with a period of one year, or may cause complex patterns of multiyear cycles with no recruitment in certain years. Stochastic effects, which in the model act to further interrupt recruitment, may enhance the likelihood of multiyear fluctuations. These results suggest that the way in which the less well understood factors, represented by the stochasticity, interact with light and nutrient levels may be of crucial importance in determining the population dynamics of giant kelp. Our models predict cycle periods of the same order (3-5 yr) as have been observed in some natural populations of giant kelp. Alternatively, with higher mortality rates and strong seasonal forcing, our model predicts annual bursts of recruitment. We suggest that the same factors producing these different dynamics in the models may explain the qualitative differences in dynamics between the giant kelp populations of central and southern California. The regulatory mechanisms discussed in the paper have some features in common with the concept of "density-vague" regulation. We contrast our work with work on this concept and propose directions for future theoretical studies on density-vague behavior.