This paper develops the conceptual and empirical basis for a class of empirical economic production models that can be linked to site-specific bio-physical models for use in integrated assessment research. Site-specific data are used to estimate econometric production models, and these data and models are then incorporated into a simulation model that represents the decision making process of the farmer as a sequence of discrete or continuous land use and input use decisions. This discrete/continuous structure of the econometric process model is able to simulate decision making both within and outside the range of observed data in a way that is consistent with economic theory and with site-specific bio-physical constraints and processes. An econometric-process model of the dryland grain production system of the Northern Plains demonstrates the capabilities of this type of model.