Recent research has clarified the bias in estimated whole-stream metabolic rates due to the unidirectional influence of groundwater inflow on stream channel dissolved oxygen.  However, the bidirectional nature of stream - groundwater exchange suggests that consideration of a complete water balance might more accurately describe hydrologic controls of solute concentrations.  Our recent experiments in sequential stream reaches have demonstrated concurrent gain and loss of stream water, as indicated by conservative solute transport.  We found net discharge increases of 0-20% (relative to upstream discharge) in reaches that concurrently lost ~10% of added tracer mass.  We are evaluating the implications of a fully specified stream channel water balance model to estimates of whole-stream metabolism.  Relative to current conceptual models of whole-stream metabolism, we expect the additional effects of a full water balance result primarily from the degree to which channel volume is reduced by hydrologic loss.  Therefore, we are focusing on characterizing the sensitivity of simulated whole-stream metabolism estimates to the ranges of hydrologic loss evident in our study reaches.  This work will demonstrate the importance of bidirectional water balance models to understanding basic stream ecosystem function in the context of hyporheic interaction.