Detection and attribution of human influence on sea level rise are important topics that have not yet been explored in depth. We question whether the sea level changes (SLC) over the past century were natural in origin. SLC exhibit power law long-term correlations. By estimating Hurst exponent through Detrended Fluctuation Analysis and by applying statistics of Lennartz and Bunde [2009], we search the lower bounds of statistically significant external sea level trends in longest tidal records worldwide. We provide statistical evidences that the observed SLC, at global and regional scales, is beyond its natural internal variability. The minimum anthropogenic sea level trend (MASLT) contributes to the observed sea level rise more than 50% in New York, Baltimore, San Diego, Marseille, and Mumbai. A MASLT is about 1 mm/yr in global sea level reconstructions that is more than half of the total observed sea level trend during the XXth century.