Hydrograph separation is one of many approaches used to analyse shifts in source water contributions to stream flow resulting from climate change in remote watersheds. Understanding these shifts is vital, as shifts in source water contributions to a stream can shape water management decisions. Because remote watersheds are often inaccessible and have poorly characterized contributing water sources, or end-members, it is critical to understand the implications of using different hydrograph separation techniques in these data-limited environments. To explore the uncertainty associated with different techniques, results from two hydrograph separation techniques, mass balance and principle component analysis, were compared using 3 years of aqueous geochemical data from the East River watershed located in the Elk Mountains of Central Colorado. Solute concentrations of the end-members were characterized by both a limited set of direct chemical measurements of different sources and detailed seasonal instream chemistry to examine the influences of uncertain end-member compositions in a data-limited environment. Annual volumetric end-member contributions to stream flow had relatively good agreement across separation techniques. Large variations in time were observed in the hydrograph separations, depending on the end-member type, and estimated flow contributions varied between the selected solutes. End-member concentrations characterized by stream chemistry showed several limitations including a reduced number of distinguishable end-members and differences in timing of flow contributions. Results highlight the benefits of using multiple hydrograph separation techniques by providing a ‘weight-of-evidence’ approach to environments with limited end-member concentration data.