A software framework for process flow execution of stochastic multi-scale integrated models

摘要: Abstract Dynamic environmental models use a state transition function, external inputs and parameters to simulate the change of real-world processes over time. Modellers specify function required in process calculation each time step component model, self-contained numerical module representing an individual spatio-temporal process. Depending on application case model – such as standalone execution or integrated source input needs be specified. The can thereby obtained by file operation execution. Alternatively, from other models, is part model. Using different notations these requirements, however, requires modification per therefore would reduce generic applicability To address this problem, we propose object notation means requirements This provides modellers with uniform syntax express within function. During initialisation, objects parameterised sources. In addition syntax, allows modeller request-reply flow coupled (i.e. requests data needed for its own progress another component). We extend approach Monte Carlo simulations implement software framework prototype. prototype, build exemplary coupling components land change, hydrology Eucalyptus tree growth at temporal discretisations obtain probability bioenergy plant growing hypothetical catchment. presented independently increases reusability models.