Quantification of uncertainty in ensemble-based predictions of climate change and the corresponding hydrologic impact is necessary for the development of robust climate change adaptation plans. Although equifinality of hydrological modeling has been discussed for a long time, its impact on hydrologic analysis of climate change has not been studied enough to provide clear ideas that represent the relative contributions of uncertainty contained in both multi-GCM (general circulation model, hereafter multi-model) and multi-parameter ensembles toward the projections of hydrologic components. This study demonstrated that the uncertainty in multi-GCM ensembles could be an order of magnitude larger than that of multi-parameter ensembles for predictions of direct runoff, suggesting that selection of appropriate GCMs should be much more emphasized than calibration of hydrological model parameters when projecting direct runoff. When simulating soil moisture and groundwater, on the other hand, equifinality in hydrologic modeling was more influential than uncertainty in multi-GCM ensemble. In addition, the uncertainty in a hydrologic simulation of climate change impact was much more closely associated with the uncertainty in ensemble projections of precipitation than that in projected temperature, indicating a need to pay closer attention to precipitation data for the improvement of the reliability of hydrologic predictions. The uncertainty in the ensemble projections of the individual hydrologic components showed unique responses to the uncertainty in the precipitation and temperature ensembles. From among 35 GCMs incorporated, this study identified GCMs that contribute the most and least to uncertainty in an assessment of climate change impacts regarding hydrology of 61 Ohio River watersheds, thereby exhibiting a framework to quantify contributions of individual GCMs to the overall uncertainty in climate change modeling uncertainty.