Confidential manuscript submitted to Geophysical Research Letters Sensitivity of blocks and cyclones in ERA5 to spatial resolution and definition

18 ERA5, the fifth-generation reanalysis of the European Center for Medium-Range 19 Weather Forecasts, provides long time series of atmospheric fields at high spatial and temporal 20 resolution. It allows detailed studies of atmospheric flow features such as blocks or cyclones. We 21 investigate characteristics of blocks and cyclones in ERA5 using different algorithms, compare 22 the results to ERA5’s predecessor, ERA-interim, and investigate how these characteristics 23 depend on spatial resolution. Generally, ERA5 and ERA-interim characterize blocks and 24 cyclones similarly. For Lagrangian detection and tracking methods, blocks are more robust than 25 cyclones to changes in resolution and reanalysis choice. Eulerian methods are robust to changes 26 in resolution. Thus, ERA5 provides a state-of-the-art reanalysis for the synoptic-scale extra27 tropical circulation. We find that cyclone characteristics are strongly dependent on spatial 28 resolution and therefore recommend that model and reanalysis data should be mapped to a 29 common resolution for verification. 30 Plain Language Summary 31 Reanalyses are among the most widely used data sets in the geosciences as they provide a 32 state of the atmosphere that is complete in both space and time by combining a state-of-the-art 33 weather prediction model with historical observations. Their applications range from 34 climatological studies to the closer examination of extreme events. Reanalyses are often the 35 primary tool to assess the performance of climate models. Recently, the ERA5 reanalysis was 36 published and thus many users may consider using this new product. However, due to its 37 novelty, ERA5 is not yet investigated extensively. We examine mid-latitudinal atmospheric flow 38 features such as blocks and cyclones and their dependence on input resolution and choice of 39 reanalysis. We find that blocks and cyclones characteristics are very similar in ERA5 and its 40 predecessor ERA-interim. Input resolution often plays a more important role on block and 41 cyclone characteristics than the choice between ERA5 and ERA-interim, particularly in case of 42 cyclone characteristics. For many applications the full resolution of ERA5 may not be necessary, 43 easing the computational requirement to use this high resolution dataset. In case of modeling 44 studies where reanalysis data is compared to modeled data, we recommend using the same 45 resolution. 46 Index terms and key words 47 AGU index set: 1610 Atmosphere, 3309 Climatology, 3364 synoptic-scale meteorology 48 6 own keywords: Block, cyclone, reanalysis, resolution, ERA5, ERA-interim 49