Moisture convergence from different sources is an importantprerequisite for a heavy-precipitation event. The contributions fromdifferent source regions can, however, hardly be quantified fromobservations, and their assessment based on model results iscomplex. Two conceptually different numerical methods are widely usedfor the quantification of moisture sources: Lagrangian approachesbased on the analysis of humidity variations along backwardtrajectories and Eulerian methods based on the implementation ofmoisture tracers into a numerical model. In this study the moisturesources for a high-impact, heavy-precipitation event that affectedeastern Europe in May 2010 are studied with both Eulerian andLagrangian moisture source diagnostics. The precipitation event wasconnected to a cyclone that... developed over northern Africa, moved overthe Mediterranean towards eastern Europe and induced transport ofmoist air towards the Carpathian Mountains. Heavy precipitation andmajor flooding occurred in Poland, the Czech Republic and Slovakiabetween 16 and 18 May 2010. The Lagrangian and Eulerian diagnosticsconsistently indicate a wide spatial and temporal range of moisturesources contributing to the event. The source with the largest share is localevapotranspiration from the European land surface, followed bymoisture from the North Atlantic. Further contributions comefrom tropical western Africa (10–20° N) and the Mediterranean Sea.Contrary to what could be expected, the Mediterranean contribution ofabout 10% is relatively small. A detailed analysis ofexemplary trajectories corroborates the general consistency of the twoapproaches, and underlines their complementarity. The Lagrangianmethod allows for mapping out moisture source regions withcomputational efficiency, whereas the more elaborate Eulerian modelrequires predefined moisture sources, but includes also processes suchas precipitation, evaporation and turbulent mixing. However, in theEulerian model, uncertainty concerning the relative importance ofremote versus local moisture sources arises from different options toparameterise moisture tagging at the surface. Ultimately a moresophisticated parameterisation scheme will be required to reduce thisuncertainty.