SYSTEMS MODELING AT SPRRS AND THE RAPPS MODEL

Systems modeling is the process of developing computer models that explain or predict the behavior of entire systems. In range research, we are interested in the grassland ecosystem, or in the case of improved forage, the agro-ecosystem. These systems are very complicated, involving multiple interactions of weather, soil, plants, animals, decomposers and human management. Range scientists have tried to understand these systems by improving the understanding of individual components and processes. This is usually done by experiments that manipulate one or a small number of factors while holding all others constant. Such studies are extremely useful for determining the importance of individual factors such as soil fertility, rainfall, plant genotype, or grazing duration for forage production and cattle gain. Most of the tremendous advances in range science over the last 75 years have been the result of such a reductionist program of research.

However, experimentation alone is not adequate for predicting the effects of simultaneous variations in multiple factors, or for understanding how grassland ecosystems function. It's simply impossible to experimentally manipulate all the relevant factors. Also, as more and more individual experiments are done, it becomes increasingly difficult to integrate their results into a well-defined conceptual framework. Our interest in systems modeling has evolved from a desire to resolve these twin problems of a complex system and information overload.

We are using systems modeling as a research tool to analyze grassland function and to integrate the results of some of our different experiments. At present, we are developing a model named RAPPS (RAnge Plant ProfileS). RAPPS is a computer model of range grass growth and production, currently focusing on Old World Bluestems, eastern gamagrass, and Big bluestem. It incorporates a great deal of data gathered here and at other range research sites, along with equations and logic derived from plant physiology, soil science and agricultural meteorology. As the model is put together, we find ourselves including plenty of working hypotheses based on the understanding and even the intuition of our scientists and other range workers. RAPPS thus shows considerable promise of becoming a tool for collecting and organizing results of Old World bluestem research.

One of the most robust components of RAPPS is the leaf photosynthesis and transpiration sub-model. This is a set of equations that calculates fluxes of gases in and out of plant leaves. Photosynthesis and transpiration, of course, are the fundamental processes in a plant's energy production and water use, so it is important that a plant growth model represent them adequately. The submodel is based on several equations that are well-accepted among plant physiologists. The behavior of these equations has been adjusted and "tuned" until they correspond closely to laboratory measurements made over the last several years.

Another nearly-complete part of the model is the determination of stem morphology and phenology. When RAPPS is run on a computer, it counts off the days in the year, reading weather data from the disk, and determining when enough degree-days have accumulated and when daylength is appropriate for a plant to go into its various morphological and phenological stages: new leaf initiation, tillering, flowering, seeding, and, ultimately, senescence.

RAPPS has recently been combined with the GRASS model developed by Gary Coughenour at theNatural Resources Ecology Lab. GRASS and RAPPS had complementary structures which allowed for a reasonably easy combination of these two independently developed plant growth models. The combined model is a multiple-year range plant simulator which has the option of simulating grazing. As the combined model becomes more complete, it will begin to have value beyond its current status as a research tool, and could well be used for assisting with management decisions.