River-aquifer exchange is highly spatially variable. For this reason, methods that estimate river gain or loss over small scales cannot be easily extrapolated to provide flux estimates over the tens to hundreds of kilometres of river length required for regional water management. Flux estimates at large scales can be provided by analysis of head gradients, differential flow gauging, river chemistry and groundwater chemistry. Hydraulic gradients allow estimation of exchange fluxes over scales of hundreds of metres or more, if piezometers located at similar distances from the river are available for measurement of aquifer heads. However, the difficulty of measuring hydraulic conductivity at this scale, and the need for piezometers at regular intervals along the river, mean that this method is likely to be inaccurate and difficult to apply in many catchments. Flow gauging has the potential to estimate fluxes over large scales, although it is difficult to apply in rivers where there is significant surface water pumping or where there are large numbers of tributaries. River chemistry can potentially provide estimates of exchange flux over distances of tens to hundreds of kilometres with a spatial resolution of hundreds of metres. Groundwater chemistry can provide estimates of loss rates at small to regional scales, depending upon the availability of piezometers for groundwater sampling, and the tracers that are used. Each method has its own advantages and limitations, and a combination of methods will usually provide most reliable flux estimates.