Modelling of seed dormancy

PHIL ALLEN; ROBERTO LUIS BENECH-ARNOLD; DIEGO BATLLA; KENT BRADFORD

Seed Development, Dormancy and Germination

Blackwell Publishing

Lugar: Reino Unido; Año: 2006;

Understanding seed dormancy sufficiently to predict germination and seedling emergence from natural seed banks has long been a goal of both seed ecologists and agriculturalists. These predictions require more complex models than for germination of agronomic species that lack appreciable seed dormancy, but offer the potential for significant economic and environmental benefits where such models succeed in predicting seed outcomes. For example, models to predict the timing of weed seed germination would be useful for scheduling seedling control efforts (Benech and Sánchez, 1995; Grundy, 2003). Seed dormancy models could also aid in the design of practices for managing native or introduced plant populations. Although several empirical models have been developed that predict germination and emergence with some success, one of the most significant limitations to the advancement of such models is the existence of dormancy in seed populations (Forcella et al., 2000; Vleeshouwers and Kropff, 2000). A detailed understanding of how environmental factors regulate dormancy status is needed in order to develop an adequate theoretical framework for the construction of predictive models that address dormancy changes in seed banks (see also Chapter 4). Several environmental factors associated with dormancy in the field are identified and their relationships diagramed in Figure 1. While our focus in this chapter is primarily on those stages most closely responsible for changes in dormancy status, it is important to recognize that predictions related to seed germination can span all the processes that begin with inputs related to weather and soil and end with outcomes related to seedling establishment and survival. Since field measurements relating to specific variables of interest can be difficult to obtain, we often accept less accurate or less precise measurements for practical reasons (e.g., atmospheric temperature and precipitation instead of seed zone temperature and water potential). It is also generally easier to measure seedling emergence from the soil rather than germination per se. The decision whether to accept a particular measure as a proxy for a variable that is theoretically more desirable but in practice more difficult to obtain often depends on the complexity of interactions among environmental factors and the desired level of accuracy for model predictions. In many cases a model must be constructed and tested in order to determine whether particular variables lead to sufficiently accurate predictions related to dormancy and germination. A further challenge in modeling of seed dormancy is that the level of dormancy (i.e., the range of conditions under which germination can proceed) is not the same for all individuals within a seed population, and models need to explicitly account for this variation. In the sections below we discuss environmental factors affecting dormancy in natural seed banks according to the general framework presented in Figure 1, with an emphasis on classifying and understanding the effects of these factors on seed dormancy changes under field situations. The aim of this classification is to facilitate conceptualization of dormancy systems for modeling purposes. Later sections are devoted to presenting some examples of population-based models that relate seed dormancy behavior to the effects of those environmental factors.