Groundwater models used to predict flow and transport through aquifers often have many parameters and take a significant amount of computational time. Because of this, it has been challenging to assess the uncertainty in these models. Keating et al. propose a method for analyzing the uncertainty of such models by selecting and analyzing a simpler surrogate model that has many of the characteristics of the process model to be evaluated. They compared two different methods for estimating the uncertainty of predictions made by the model and found them to be consistent.