The interconnections between groundwater and river systems remain poorly understood in many catchments throughout the world, and yet they are fundamental to effectively managing water resources. Groundwater extraction from aquifers that are connected to river systems will reduce river flows, and this has implications for riverine ecosystem health, water security, aesthetic and cultural values, as well as water allocation and water management policies more generally. The decline in river flows as a consequence of groundwater extractions has the potential to threaten river basin industries and communities reliant on water resources. ¶ In this thesis the connectivity between groundwater and river systems and the impact that groundwater extractions have on river flows were studied in one of Australia’s most developed irrigation areas, the Namoi River catchment in New South Wales. ¶ Gauged river reaches in the Namoi River catchment were characterised according to three levels of information: 1) presence of hydraulic connection between aquifer-river systems; 2) dominant direction of aquifer-river flux; and 3) the potential for groundwater extraction to impact on river flows. The methods used to characterise the river reaches included the following analyses: 1) a comparison of groundwater and river channel base elevations using a GIS/Database; 2) stream hydrographs and the application of a baseflow separation filter; 3) flow duration curves and the percentage of time a river flows; 4) vertical aquifer connectivity from nested piezometer sites; and 5) paired stream and groundwater hydrographs. ¶ The theoretical responses for gaining, losing and variably gaining-losing river reaches were conceptualised along with the processes that operate in these systems. Subsequently, a map was prepared for the Namoi River catchment river reaches indicating aquifer-river connectivity and dominant direction of flux. Large areas of the Upper Namoi River catchment were found to have connected aquifer-river systems, with groundwater extraction bores located in close proximity to the rivers. Accordingly, the potential for groundwater extraction to impact on river flows in these areas was considered significant. The Lower Namoi was assessed as having mostly disconnected aquifer-river systems. ¶ In order to investigate the impacts of groundwater extraction on river flows in connected aquifer-river systems, a simple integrated aquifer-river model entitled IHACRES_GW was developed for use at the catchment scale. The IHACRES_GW model includes a dynamic, spatially-lumped rainfall-runoff model, IHACRES, combined with a simple groundwater bucket model that maintains a continuous water balance account of groundwater storage volumes for the upstream catchment area relative to the base of the stream, assumed to be the stream gauging station. The IHACRES_GW model was developed primarily: 1) to improve upon existing water allocation models by incorporating aquifer-river interactions; 2) to quantify the impacts of groundwater extraction on river flows within unregulated, connected aquifer-river systems; 3) to inform water policy on groundwater extraction; and 4) to be able to utilise the model in future integrated assessment of water allocations options at the catchment scale. ¶ The IHACRES_GW model was applied within the Cox’s Creek subcatchment in order to test its validity. The model was used to simulate a range of extraction scenarios which enabled the impacts of groundwater extractions on river flows to be assessed. In particular, the historical impacts of groundwater extraction on the timing, magnitude and frequency of baseflow events were quantified over a 15-year (1988-2003) simulation period. The IHACRES_GW model was also used to evaluate the implications of water sharing plans for the Cox’s Creek subcatchment. ¶ A spatially-lumped modelling approach in the management of water resources has a number of limitations, including those arising from the lack of spatial considerations. However, it offers a number of advantages including facilitating a better understanding of large-scale water management issues, assessing the impacts of water allocation and groundwater extraction on river flows at the catchment scale, and informing water sharing plans. In particular, this type of modelling approach lends itself to integrated assessments of water allocation options in which hydrological, ecological and socioeconomic data sets are combined, and where data is commonly aggregated to a larger scale of interest in response to the requirements of policy makers. The research findings from this thesis provide some insights into how to better manage the impacts of groundwater extraction in connected aquifer-river systems.