Growth rules in clonal plants and environmental predictability--a simulation study

The adaptive nature of clonal growth rules was studied using a Monte Carlo simulation based on a stochastic, spatially explicit growth model. In this model, the development of a clone was controlled by environment-dependent growth rules, acting upon elementary developmental decisions, such as suppression or activation of meristems at certain locations, or modification of internode length. The neutral model, used as a control, simulated a plant of rigid, non-environment-dependent form. Growth proceeded in a spatially and temporally heterogeneous environment. A wide range of environmental types - modelled as mosaics of resource-rich and resource-poor sites - were used for the tests, and analyzed in the framework of an information theory model. The measures of information content (e.g. predictability) of the environment derived from this model allowed direct comparison of patterns in space and time. Seven sets of growth rules were examined. The growth forms resulting in different environmental settings were characterized by simple 'cost' variables which may act as factors under selection in the different species. These costs concern the morphology of the developed clones (cost of exploration and intraclonal interference) and the degree of matching the resource pattern in the environment (cost of placing ramets into resource-poor sites). Developmental decisions at individual meristems were based on site quality in their immediate neighborhood. As a consequence, it was found that the information content of local environmental signals, as characterized by predictability, played a crucial role in determining the costs of a particular growth type. The simulation shows that the effectiveness of resource acquisition is determined by both environmental resource pattern and architectural constraints. Each of the different growth forms proved adaptive only in a certain (often narrow) range of environmental types. The importance of considering not only the global pattern of resource availability, but its predictability in space and time is emphasized.