Quantifying flow fields inside various microfluidic devices, especially in the interfacial region, is important since surface (vs. bulk) effects are significant at micro-to nano- scales. Over the last few years, the nano-particle image velocimetry (nPIV) technique, based upon evanescent-wave illumination of fluorescent colloidal tracer particles, has been developed and used to measure the two velocity components parallel to and within a few hundred nanometers of the wall. The first objective of this doctoral research is to develop and quantify the accuracy of an extension of nPIV — "multilayer nPIV" for obtaining velocities at different distances from the wall using a single set of nPIV images. The exponential decay in intensity of evanescent waves suggests that it may be possible to divide a single nPIV image into a few sub-images based upon tracer image intensity; velocities at different distances from the wall can then be extracted using standard PIV image processing algorithms. The second objective of this research is to evaluate the accuracy of nPIV data, including phenomena such as Brownian diffusion and non-uniform nature of the illumination. The proposed research involves both experimental and numerical work. The feasibility of multilayer nPIV will first be evaluated using artificial images of plane Couette flow incorporating hindered Browniandiffusion, evanescent-wave illumination and image noise. The effects of Brownian diffusion will be studied using hindered Brownian dynamics simulations. Finally, a "proof of concept" experiment will be performed to determine the feasibility of the technique for actual nPIV images. The results of this thesis have the potential to lead to a technique with the capability to measure velocity profiles within a few hundred nanometers of the wall, and a fundamental understanding of the accuracy and reliability of this technique.