Variations in Sea Surface Temperature (SST) are an important driver of marine species abundance in Large Marine Ecosystems (LMEs). Studies concerned with climate change induced SST trends within these LMEs have so far been relying on satellite data and reanalysis products, with the disadvantages of only having short time-periods available and having to rely on the ability of the models to correctly simulate SST-dynamics, respectively. Here, we provide for the first time a long-term trend analysis of SST for 17 LMEs of the Atlantic Ocean over two different time-periods (1957–2020 and 1980–2020) based on in-situ data gathered from three data collections. We sort our results according to warming categories that were established in an earlier study, i.e., “cooling” (below 0 °C/dec), “slow” (0–0.07 °C/dec), “moderate” (0.07–0.14 °C/dec), “fast” (0.14–0.21 °C/dec) and “superfast” (above 0.21 °C/dec). Our results show a persistent “slow” to “superfast” warming in all considered LMEs. However, the sparse data coverage induces large uncertainties, so that many LMEs cannot uniquely be assigned to one warming category only. We detect no systematic changes in the seasonal SST amplitude of the considered LMEs. We find that the LMEs of the North Atlantic warm faster than those of the South Atlantic and that this difference is increasing with time. Out of the North Atlantic LMEs, the Norwegian Sea, North Sea, Celtic-Biscay Shelf, Gulf of Mexico and the Northeast U.S. Continental Shelf belong exclusively to the superfast warming category for the period 1980–2020.