Monitoring of coastal hydrodynamics over a relatively large spatial extent and a long temporal period is essential to assess the impact of coastline interventions on several functions. An example of a recent coastline intervention is the Sand Engine, a pilot project for optimization of (future) coastline maintenance. This innovative pilot project is therefore subject to an intensive research and monitoring program that allows for development and validation of new measurement techniques. Continuous series of spatial hydrodynamics measured by remote sensing instruments have the potential to complement the temporally and spatially limited measurements of in-situ instruments. In this thesis, the accuracy of hydrodynamics measured with a land-based X-band radar at the Sand Engine is examined. The mapping of the sea surface by an X-band radar results in wave dispersion patterns in the radar domain over time. This allows for the extraction of spatial maps of the hydrodynamic parameters d and U by spectral (Fourier) analysis of wave dispersion patterns on a limited spatial scale (computational cube size). Previous research has shown that the spatial and temporal scale of wave dispersion analysis are critical factors for the accuracy of remotely-sensed currents. The recently developed open-source XMFit algorithm (Friedman(2014)) is used to extract X-band radar currents. It is not exactly known what X-band radar derived currents exactly represent. In addition, there is no agreement whether a radar measures Eulerian or Lagrangian currents. Therefore, the objective is to find the relation between spatial scale of wave dispersion analysis and X-band radar current measurement accuracy at the Sand Engine by performing point and spatial validation of currents obtained by analysis of wave dispersion on different spatial scales. The Sand Engine study site was selected due to availability of both Eulerian ADCP and Lagrangian drifter measurements. The relation between measurement accuracy and computational cube size was found to be hyperbolic for both the point and spatial validation, which is connected to the hyperbolic relation between cube size and spectral resolution dk. In-situ current estimates are consistently overestimated by XMFit for grid points within the first 500 m from the coastline. Non-linear wave effects cause anomalies in the image spectrum compared to the definition of a useful spectrum. These anomalies do not follow linear theory on which the method is based and therefore have a negative effect on the current estimate of XMFit. In the point validation at two ADCP locations, XMFit current estimates were compared to pseudo-Lagrangian ADCP currents. The pseudo-Lagrangian ADCP currents are obtained by correcting Eulerian ADCP measurements for the surface Stokes drift. A direct comparison of XMFit to the Stokes-drift corrected ADCP measurements proves that the correlation in cross-shore direction is high. This supports that X-band radar currents are Lagrangian. The spatial validation by drifter measurements is performed for another temporal period. The accuracy is significantly lower then during the point validation period: a significant offset in current direction between drifter and radar measurements is observed. A sensitivity analysis showed that, for this period in time, malfunctioning of the spectral filter to remove aliased spectral energy causes this offset. It was proven that no offset in direction is observed when optimal spectral filtering parameters are selected. Finally it is emphasized that optimization of spectral quality is the key to even more accurate current estimates. Further study to accuracies should focus on areas with low hydrodynamic complexity to better understand the fundamentals of radar currents before examining the performance of XMFit for complex flow patterns.