A pasture simulation model for dry matter production, and fluxes of carbon, nitrogen, water and energy

A mechanistically based ecosystem model for managed productive pastures was developed to simulate the annual production of plant biomass and the nitrogen balance, and seasonal patterns of growth, and of fluxes of latent, sensible and soil heat. The model is composed of submodels for plant growth, microclimate, soil biology, and soil physics. At a time step of one hour, above-and below-ground plant growth is calculated in relationship to fluxes of carbon, nitrogen, water and energy. Plant growth is separated into reproductive and vegetative developmental stages. The model is driven by hourly weather input data for radiation, temperature, vapour pressure, wind speed, and precipitation. For two sites located on the Swiss Plateau at 548 and 915 m above sea level the difference between measured and simulated annual dry matter was in the range of -6-21% depending on location, year, and cutting regime. Analysis of the annual dynamics of simulated and measured dry matter production revealed the importance of the differentiation between reproductive and vegetative growth. Largest deviations in yield were obtained for growth periods with extremely dry or cold and rainy weather. Simulated energy fluxes agreed well with measurements, whereas annual nitrogen yield tended to be underestimated by the model. Analysis of sensitivity to model parameters and initial conditions revealed the high importance of a parameter for the nitrogen concentration of the structural plant dry matter. This could indicate that the nitrogen sink limitation of the model could be too low. It is concluded that on an annual basis productivity, energy balance and nitrogen dynamics of productive pastures under mid-European conditions can be simulated accurately with a high temporal resolution using today's climatic conditions between 500 and 1000 m above sea level, and that the model could be used to assess the sensitivity of pastures to future climatic changes. (C) 1998 Elsevier Science B.V.