ABSTRACT
This chapter examines how the aggregation of input data influences the predictive capabilities of the forest process models across a changing spatial or temporal resolution. It examines the influence of data aggregation on multiple-scale forest process model development, use, and validation, using PnET-IIS, a physiologically-based model for predicting the forest hydrology and productivity at the stand, ecosystem, and regional scales. The chapter outlines overall model structure and data requirements and discusses model use and limitations at three spatial scales. PROSPER, a phenomenological, one-dimensional model linked the atmosphere, vegetation, and soils. PROSPER used electrical network equations to balance water allocation between vegetation and the three soil layers. The flows of water within and between the soil and the plant were a function of the soil hydraulic conductivity, soil water potential, root characteristics in each soil layer, and surface water potential.
