Why model root system architecture?

Abstract

In this short discussion, I shall attempt to classify some of the various models of root system architecture according to the different purposes they can serve. A major objective, which is recurrent in many modelling approaches, was to provide better representations of the root system as an input for uptake models or crop models. I show how the required representations followed the concepts of the uptake modelling approaches. A second objective was to study the morphology and morphogenesis of the root system per se. Some examples show how the models of root system architecture can provide new insights in studying root development, comparing genotypes, and quantifying the effects of the environment on the root system. Another promising domain in which such models were shown to be helpful is the integration at the root system level of physiological processes, especially on carbon allocation. This integration allows new studies of the functional relationships between the root system and the shoot system. Methodological problems remain a crucial point in the observation of root system. Therefore, several authors have tried to show how theoretical root systems, as they are simulated by architectural models, may help in the design and optimisation of new experiments. The possible integration of local and global processes, or processes at different space and time scales into a consistent frame (the model), is also very valuable in root research since observations are generally scattered in both time and space. Thus, this review shows the very large potential of such models in the general scope of root system and plant research. These models could probably serve additional purposes that have not been considered up to now, like the study of anchorage, in which the representation of the architecture of the proximal root system is probably a key step. For the next future, models will appear for integrating more physiological processes, relating to both the structure (morphogenesis, architecture) of the root system, and to its function (uptake, and resource allocation). The difficulty will be to find good compromises between the need for integrating more processes, and the complexity of such models.