This dataset contains the results of the analysis of standing stocks and activity of prokaryotes during the KOSMOS mesocosm experiment carried out in the framework of the Ocean Artificial Upwelling project. The experiment was carried out in the autumn of 2018 in the oligotrophic waters off Gran Canaria. During the 39 days of experiment nutrient-rich deep water was added to the mesocosms in two modes (singular vs recurring additions), with four levels of intensity. Prokaryotic abundances and viability were quantified with a FACSCalibur flow cytometer (Becton-Dickinson). Prokaryotic activity (as protein synthesis) was assessed by means of BioOrthogonal Non-Canonical Amino acid Tagging (BONCAT), using a Zeiss Axio Imager.Z2m Epifluorescence Microscope (AxioCam MRm, Carl Zeiss MicroImaging). The aim of this dataset was to study the effect of different upwelling regimes on the prokaryotic communities and its potential implications for organic matter cycling.