The work presented in this dissertation discusses large scale flow and transport in river networks and investigates advantages and disadvantages of grid-based and vector-based river networks. This research uses the Mississippi River basin as a continental-case study and the Guadalupe and San Antonio rivers and Seine basin in France as regional-case studies. The first component of this research presents an extension of regional river flow modeling to the continental scale by using high resolution river data from NHDPlus dataset. This research discovers obstacles of flow computations for river a network with hundreds of thousands river segments in continental scales. An upscaling process is developed based on the vector-based river network to decrease the computational effort, and to reduce input file size. This research identifies drainage area as a key factor in the flow simulation, especially in a wet climate. The second component of this research presents an enhanced GIS framework for a steady-state riverine nitrogen transport modeling in the San Antonio and Guadalupe river network. Results show that the GIS framework can be applied to represent a spatial distribution of flow and total nitrogen in a large river network with thousands of connected river segment. However, time features of the GIS environment limit its applicability to large scale time-varied modeling. The third component shows a modeling regional flow and transport with consideration of stream-aquifer interactions at a regional scale at high resolution. The STICS- Eau-Dyssée combined system is implemented for entire seine basin to compute daily nitrate flux in the Seine grid river network. Results show that river-aquifer exchange has a significant impact on river flow and transport modeling in larger river networks. / text