This chapter examines the question of uncertainty in environmental modeling in relation to the apparent equifinality of model structures and parameter sets in reproducing the observed behavior of environmental systems. The chapter explores Monte Carlo based methodology for the assessment of uncertainty in the predictions of environmental models, which explicitly recognizes that because of limitations of both knowledge and observational data, it may be impossible to construct a unique model of such systems. One important feature of the approach is that it is possible to study whether parametric change in the system is detectable within the constraints of model uncertainty. One important implication of exploring the possibility of equifinality in environmental modeling is the focus on the importance of data for both, conditioning, rejecting, and critically testing models. Equifinality in modelled responses complicates matters, not least because it must throw doubt on parameter values for a wide variety of models that have been determined by calibration and reported in the literature as characteristic of a given vegetation type or soil or catchment. However, recognition of equifinality can also be seen as a very positive step in the attention on data that results.