|
|
|
|
WIMOVAC DiscussionWimovac currently provides a firm foundation for future modelling work which meets the initial specification for a modular object orientated general model of vegetation and soil response to climate change. A key aim of creating an open, easy to use model including context sensitive online information has been achieved in a uniquely interactive fashion. However further development of model modules for a more detailed treatment of carbon partitioning and allocation, tree related processes and nutrient uptake is indicated. Attention to herbivory and refinement of links between ecophysical and ecosystem processes is also required. Probably the greatest weakness of all existing models of vegetation response to climate change is the lack of a truly mechanistic understanding of partitioning and allocation of carbon from photosynthesis to the growth and maintenance of plant structures. There is a considerable body of evidence that many plant responses to environmental change are not in the fundamental efficiency of core processes such as photosynthesis or nutrient uptake but rather in the relative allocation patterns of resources to sources and sinks within the plant. An understanding of allocation and its downstream effects on root:shoot ratio and concomitant effects on all aspects of plant physiology, from productivity to water use, is vital if we are to reliably predict the effects of global climate change on vegetative systems. The Farrar, et al. (1996) model described here represents a first step in the construction of such a model and represents a unique opportunity to develop a mechanistic allocation model in the context of a broader scope vegetation growth model. The phasic growth model adopted by wimovac is necessarily simple for use with the available datasets and should be improved by addition of mechanistic processes as these become apparent, or as a minimum expanded by the addition of plant specific empirical ‘rules’ for development, such as a need for a cold period before successful germination or flowering. Many existing crop and vegetation growth models include such empirical rules based upon experimental observation and the modular structure of wimovac insures that these may be incorporated without difficulty. Despite the large number of empirical, phenomenological, hierarchical and soil property orientated root models, few have addressed both the soil and crop factors effecting root growth simultaneously. One such attempt has been made by Jones et al., (1991){Allan Jones, 1991 #2036} and this model represents an opportunity to include a largely mechanistic approach to modelling root response to a number of factors including: aluminium toxicity, calcium deficiency, course root fragment content of soil layers, soil water content, temperature, bulk density and texture. The Jones et al., (1991) model is relatively complex and difficult to parameterise but its underlying structure and approach is compatible with the soil-plant-atmosphere components of wimovac and could be incorporated relatively easily in order to strengthen this component of the model. In many natural plant-soil systems phosphate, although not required in as large quantities as nitrogen, is an important limiting nutrient. The Century model proposed by Parton, et al. (1988) originally included a sub model of soil phosphate cycling which was not implemented here due to difficulties with parameterisation. However the similarity of structure between the phosphate sub-model and that of the soil C and N sub models (Figure 20 and Figure 21) allows for its easy incorporation at a later date if sufficient information for parameterisation becomes available. There is currently little direct experimental evidence to support the wide spread acceptability of the mechanistic ozone model proposed by Martin (1996) and described briefly here. However the model does offer an intriguing potential to create a general purpose model of vegetation growth that incorporates the effects of increased atmospheric ozone concentration, at the leaf level, in its responses to climate change. Vegetation model temporal scale issues are likely to be centred upon not only the expected changes in climate but also upon the acclimation/adaptation of plant metabolism and growth. Traditional empirical models by their nature do not usually incorporate processes by which acclimation may be expressed in a scientifically sound manner and although such responses may be incorporated in the leaf photosynthetic models of Farquhar and von Cammerer (1980, 1982) and Collatz (1992), there is currently little experimental evidence for a systematic approach to these changes. There is consequently a requirement for further long term experimental work exploring the effects of climate change on key physiological parameters. From a practical point of view the user interface of wimovac could be improved by incorporating a higher level of file organisation than currently exists. In the present version of wimovac the user is left free to organise parameter files, meteorological data, experimental data and simulation results on the hard disk of their computer in whatever manner they feel appropriate. This should be improved by taking a project based approach in which all resources required in a specific simulation set are located within a named project. The project can then be organised on the hard disk automatically and can be accompanied by important notes, and dates describing the material. The user can then be presented with a meaningful summary of the project rather than ambiguous file names. |
Send mail to humph@essex.ac.uk with questions or comments about this web site.
|
