UDASH-SalaciaML-2-Arctic represents an updated and simplified version of the UDASH dataset, that has been created to develop an artificial intelligence algorithm, that we name SalaciaML-2-Arctic to support the visual/human quality control (QC) of the data. UDASH-SalaciaML-2-Arctic can be directly used with our algorithms, provided under the GitHub repository (https://github.com/GastonKreps/SalaciaML-2-Arctic), to reproduce our results, extend the methods and more. Additionally, we have implemented SalaciaML-2-Arctic as an user-friendly app at https://mvre.autoqc.cloud.awi.de. The following steps have been applied to the original UDASH dataset to create UDASH-SalaciaML-2-Arctic:- Single, annual .txt files were concatenated into two single .csv files, one for temperature and one for salinity.- Parameters have been reduced to a minimal set, names and units have been adapted slightly.- Temperature and salinity gradients have been calculated and added. - Missing values are kept as -999.- Traditional QC flags, generated by the classical algorithms, have been added: 0=good, 2=suspect_gradient, 4=spike.