A methodology to qualify and quantify uncertainty and sensitivity measures, mathematically related to a representative metamodel and correlation coefficients, using the Latin hypercube approach, is evaluated in the context of the spatially distributed hydrological model MIKE SHE. The characteristics of various outputs, such as cumulative catchment discharge, average soil water content, and groundwater elevation, are examined at different time and space scales. The soil hydraulic parameters make up the varying input. The input uncertainties and the corresponding Latin hypercube parameter perturbations are based on U.S. Department of Agriculture texture figures. Results indicate important differences between the measures, even in ranking, making combined interpretation of the measures necessary. The real-world problem of correlation among parameters adds significant complexity to the assessment of uncertainty and sensitivity and cannot be disregarded without caution. The presented methodology, though still CPU intensive, allows the hydrological modeler to cope with this complexity.