Assessing uncertainty propagation through physically based models of soil water flow and solute transport

Soil hydrological models are inherently imperfect because they abstract and simplify ¿real¿ hydrological patterns and processes. Indeed, an important aim of modeling is to establish the simplest description possible for adequately addressing a particular problem. Also, models are frequently based on input data that are known to be inadequate for some practical purpose. Thus, uncertainties in model outputs originate from uncertainties in input data, which include measurement and interpolation errors, and in models that include conceptual, logical, mathematical, and computational errors. Understanding the causes and consequences of uncertainty in soil hydrological modeling is useful for: (i) establishing the utility of data and models as decision-support tools; (ii) directing resources towards improving data and models, and (iii) seeking alternative ways of managing soils when the opportunities for accurate modeling are limited. This chapter focuses on statistical methods for assessing uncertainty in soil data and models, propagating uncertainties through models, and assessing the contribution of different sources of uncertainty to the overall uncertainties in model predictions. In addition, it explores the impacts of scale, and changes between scales, on the outcomes of an uncertainty analysis. It concentrates on physically based models of soil water flow and solute transport, and provides numerous examples from the literature her