Seismic tomography techniques are frequently used to obtain velocity models of the weathering layer, which are then used to calculate static corrections times. The main idea in tomography is to use the misfit between synthetic and observed first-arrival (refraction) times to correct an initial velocity model. Due to the nature of the seismic method where, in general, sources and receivers are placed on only one edge of the medium, the Earth surface, it can be shown that inverting for velocities using the observed times does not give a unique solution. Instead, a family of solutions is obtained that explains the same data set. To give a meaning to a nonunique solution, uncertainty analysis techniques are necessary. They can give information about the family of solutions such as which zones have a larger error, which zones are better resolved, is there a dispersion measurement (such as variance) associated with the solution model family, and is there a more probable solution model. To give clues to the answers to these questions, we have used a real seismic data set and applied four uncertainty assessment methods: jackknifing, checkerboard test, Monte Carlo, and bootstrapping.