We present a new tomographic model for P- and S-velocity anomalies beneath Europe (30°N-55°N, 5°W-40°E), extending in depth up to 700 km and constrained by inversion of data from the International Seismological Center (ISC) catalogue. The algorithm uses the traveltimes from events located in the study area recorded by all available worldwide stations, as well as times from teleseismic events recorded by European stations. The events from the ISC catalogue have been relocated and combined into composite events. All the traveltimes were corrected for crustal structure using the reference model EuCRUST-07. The resulting velocity anomalies show similar large-scale patterns as observed in previous studies, but have a higher resolution, which allows detection of some features in more detail. For example, it is now possible to assess the depth extension of the small slow velocity body beneath the Eifel region and Eger graben. The P and the S model show a good consistency in the uppermost 200 km below most of the European area and in some parts even in the deeper layers (e.g. beneath the Apennines and the Hellenic arc). The new model provides clear images of some principal features, which were previously detected in a limited number of studies, while the comparison between P- and S-velocity anomalies provides novel constraints to address on their nature (e.g. the gap in the Adriatic plate subducted below the central-southern Apennines). In this paper, we pay special attention to testing the reliability of the results. The random noise effect is evaluated using a test with independent inversion of two data subsets (with odd/even events). The spatial resolution is estimated using different checkerboard tests. Furthermore, we present a synthetic model with realistic patterns, which reproduces after performing forward and inverse modelling the same shape and amplitudes of the anomalies as in the case of the real data inversion. In this case, the parameters of the model can be used to assess the amplitudes of P and S anomalies that is critical for evaluation of other petrophysical parameters (temperature, density, composition, etc.) in the upper mantle. © 2009 The Authors Journal compilation © 2009 RAS.