This paper presents a dataset on the abiotic (oceanographic, atmospheric and global climatic indices) and fishery variables of the marine-coastal area of the Magdalena Province in the area between Taganga and Bahía Concha, located north of Santa Marta in the Colombian Caribbean. The abiotic variables were downloaded from the satellites of the National Aeronautics and Space Administration (NASA) and the meteorological stations of the Institute of Hydrology, Meteorology and Environmental Studies (IDEAM). The fishery variables were obtained through field trips in the study area. A dynamic artificial neural network was implemented to reconstruct the missing data in the fishery variables from the known abiotic variables (Precipitation, North Atlantic Oscillation and Multivariate ENSO Indices). In this way, a dataset was obtained that is important to determine the historical changes of fishery resources for the study area and to make catch forecasts incorporating the variability of the environmental conditions (atmospheric and oceanographic).