Field measurement and numerical modeling are the most popular and fundamental approaches for studying mixing pattern in rivers and coastal waters. Due to the limitations associated with both of these methods they should be used together to verify each other. Extensive field measurement was conducted on the effluent plume from the outfall of the Capital Region waste water treatment plant in the North Saskatchewan River. Tracer was injected at the outfall location and the mixing pattern was investigated by tracking the tracer concentration over a 83 km reach of the river. Flow velocity and depth were also measured simultaneously using an acoustic Doppler current profiler. An integrated in situ fluorometer-GPS measurement technique was introduced and used for field tracer studies in meandering rivers. The full transverse mixing length for the river was estimated to be 130 km. A stream-tube orthogonal curvilinear mesh generation algorithm was also developed for numerical modeling of meandering rivers. The method eliminates the effect of transverse velocity field using the stream-tube concept. The field measured velocity data were used for calculating the stream tube width in each cross-sectional strip. The stream-tube grid was used to develop a practical and efficient coupled field-numerical model for estimating the transverse mixing coefficient in meandering rivers. In this model the computational costs associated with solving the hydrodynamic sub-model is reduced by generating the velocity field from measured data. Using the calibrated model, the average transverse mixing coefficient was calculated for the surveyed reach. Extensive field study was also conducted on the near-field and far-field of a thermal plume discharged by the Ras Laffan Industrial City in Qatar. Three-dimensional perspective of the plume behavior was obtained using field measured temperature and velocity data. Different characteristics of the observed plume including the extent of different zones of the plume, plume thickness, detachment depth and variation of the minimum dilution were investigated and compared with available theories. The contribution of each effective mixing mechanism was also calculated using the field measured data. Vertical confinement was found to be the main effective parameter on the near-field mixing rate which reduced the minimum dilution rate up to 80%. An innovative remote sensing technique was introduced to investigate the near-field mixing of thermal surface plumes. The method generates a calibrated thermal image of the plume using LandSat thermal infrared (TIR) satellite images. Using a combination of remote sensing and numerical modeling, the near-field dynamics of the plume was found to be influenced by the wind action. It was also observed that the previous classification for determining the effect of wind on the plume dynamics did not successfully predict the plume behavior in shallow water. Two non-dimensional parameters, WI1=Uwl/U0 (ratio of the long-shore wind speed (Uwl) to the discharge velocity (Uo) and WI2= Uwc/U0 (ratio of the cross-shore wind (Uwc) to the discharge velocity), were introduced to quantify the effect of wind on the plume dilution and deflection. The plume trajectory was found to be sensitive to a longshore wind greater than 2 m/s, which is half of the reported value for deep water conditions. The surveyed coastal outfall was also modeled using a nested coupled hydrodynamic-wave approach. Validation of the model with field measured and remote sensing data showed that the employed approach can be used for engineering applications such as designing outfall systems and environmental impact assessment purposes. The calibrated model was used to investigate the effect of various effective factors on the mixing process such as lateral confinement, wave-flow interaction, wave dissipation factors and turbulence models. Lateral confinement was found to reduce the mixing potential of the outfall by 50% at the end of the near-field.