An attempt is made to compare results of large-eddy simulation (LES) in a convective boundary layer using the model PALM with experimental data obtained from acoustic travel time tomography. This method provides two-dimensional data arrays, which are considered as more suitable for LES-validation than classical local or line-integrated measurements, because the tomographic data are area- or volume-averaged. For a quantitative comparison with experimental data in general, some prerequisites have to be considered: First of all, the initial and boundary conditions of the LES model have to be provided correctly by the experiment. Considering measurement errors, a sensitivity study was performed to investigate the influence of inaccurate initial and boundary conditions on the simulation results. This showed that for determining some boundary conditions, such as the surface temperature and the roughness length, high measurement accuracies are necessary, which are difficult to reach or which at least require considerable extra measurement efforts. The initial and boundary conditions provided by the Lindenberg experiment in 1999 turned out to be of insufficient accuracy to allow quantitative comparisons. However, a qualitative comparison was performed instead to investigate if the acoustic tomography method is a proper method for comparisons with LES models in general. It showed a good qualitative agreement with some quantitative differences. These differences can partly be explained by the sensitivity of the LES to initial and boundary conditions and by the limitations of the acoustic tomography.