In many streams, discharge is often anticipated to be the residual between the inputs to watersheds (precipitation, groundwater inputs), outputs (evapotranspiration, groundwater outflow), and changes to water storage. This basic water balance approach fails to capture many of the aspects that drive streamflow, namely the sequence between fluxes and storages that must occur prior to streamflow generation. This is further complicated by the limitations of ever estimating total water storage, namely due to subsurface heterogeneity, poorly defined boundary conditions, and difficulties of associated with measuring certain fluxes (GW in and outflows) and storage reservoirs (soil water). From a hillslope perspective, the relationship between watershed storage and discharge is not so linear. There are sequences of storage thresholds which must be breached prior to other storage units being activated. As storage units fill (i.e., fill and spill storage + surficial soil storage) they activate other hydrologic processes (percolation) that spurs the filling of other storage components (groundwater storage). These thresholds are complex, time varying, with non-linear and hysteretic activations, and result in much more complex transit times than a simple water balance would suggest. We use records from ~13 years of intensive hydrologic monitoring at 3 adjacent low-relief, groundwater driven headwater streams that drain the Savannah River Site in the Upper Atlantic Coastal of Plain to explore the sequential storage thresholds that govern stream discharge at the site. Coupled with a spatially discrete water table model and remotely sensed estimates of surface water storage, we provide a more robust estimate of total water storage and help elucidate how much water is going unmeasured via deep groundwater flow pathways