Streamflow depletion describes the reduction of baseflow to a stream resulting from groundwater extraction. Traditionally, this was conceptualised as the reduction of baseflow at a specified time following a finite period of continuous extraction. Here we termed this concept “instantaneous streamflow depletion” (ISD). Since the 1950s, a range of analytical solutions have been developed to predict ISD. The subsequent advent of numerical groundwater flow simulation enabled more detailed assessments, including the use of both forward and adjoint modelling approaches. However, traditional analytical and numerical approaches both require the evaluation of a large number of model runs, in order to assess a range of potential bore-stream separation distances (i.e. potential bore locations). In the present study, an alternative concept was considered: the total volumetric reduction in baseflow to a stream resulting from continuous groundwater extraction over a finite period. We termed this concept “cumulative streamflow depletion” (CSD), which is relevant to combined assessments of surface water and groundwater balances. We derived an adjoint state approach to provide efficient estimates of CSD. A single numerical adjoint model run provided estimates of CSD impacts at all potential locations of interest. The composition of the loading term that forms the basis of the adjoint state approach was derived analytically from first principles. The method was validated using multiple synthetic case studies representing variations in (a) streambed resistance and (b) stream and bore base elevations. The results of adjoint solutions compared almost exactly to those calculated using equivalent closed-form analytical solutions. The adjoint method was also used to estimate the potential for CSD in an alluvial groundwater flow system in the Gloucester Basin, based on an existing numerical forward model. In summary, the adjoint state method derived in this study was found to provide an efficient means of estimating cumulative streamflow depletion based on existing forward models.