The Arctic freshwater cycle is an essential component of the Earth's water cycle system and it has experienced dramatic changes during recent years. This study aims to comprehensively assess the Arctic freshwater cycle derived from the ERA5 (atmosphere), ERA5-land (land) and ORAS5 (ocean) reanalysis datasets distributed by the European Copernicus Climate Change Service. It is found that, in these datasets and during the 1979-2018 period, (1) river runoff, inflow through the Bering Strait (BS), and net precipitation are the main sources of freshwater into the Arctic Ocean. By contrast, the freshwater export occurs mainly through the Fram Strait (FS) and Canadian Arctic Archipelago (CAA). However, the ocean freshwater exports from the FS and CAA based on ORAS5 reanalysis data is found to be much lower than values reported in other studies. This underestimation likely causes an imbalance in the Arctic Ocean freshwater cycle. (2) In terms of seasonality, values of the moisture convergence and precipitation over the Arctic Ocean, the evaporation and precipitation over the drainage basin are the highest in summer. The total runoff peaks in June due to snow melting. The ocean liquid freshwater import through BS and the export through the CAA, are the highest in summer. Sea-ice freshwater exports is the strongest at the CAA in summer and at the FS in winter. (3) In terms of trends, both precipitation and evaporation over the Arctic Ocean have increased significantly from 1979 to 2018, whereas over land, evaporation has increased but precipitation and runoff have decreased. There is a notable increase in moisture convergence over land and an overall increase in the total water volume over the ocean and land. The decrease in sea ice freshwater storage associate with sea ice melt has led to increased storage of liquid freshwater in the Arctic Ocean on annual average. Overall, we suggest that ERA5 reanalysis data is suitable for the analysis of the atmospheric freshwater cycle at the large scale and on climatic time scales; that ERA5-land data for runoff, while freshwater transport from the FS and CAA are not well represented in ORAS5 data. The trends in liquid and sea-ice freshwater transport in the Arctic Ocean are highly sensitive to the period of reference. Therefore, caution should be exercised when using ORAS5 data to study and interpret the trend of freshwater transport  into or from the Arctic Ocean.