Doubtless there is some utility in playing out worst-case scenarios as a way to plan for the future. That being said, the New York Times is reporting a new study, "Millions projected to be at risk from sea level rise in the continental United States," in Nature Climate Change that calculates the number of Americans that would be affected by sea level rises between 3 and 6 feet by 2100 due to man-made global warming. As far as I can tell, the main contribution of the study is new coastal population projections. The researchers calculate that if sea level rises by 3 feet, the land that 4.2 million Americans live on will be at risk of inundation. If sea level rises by 6 feet, then 13.1 million Americans might have to move inland. The costs of relocating that many people could be as much $14 trillion. So researchers can now get their studies published in a prestigious scientific journal and publicized by leading newspapers by merely assuming a catastrophe. Neat trick.
In any case, the researchers cite various estimates for future sea level rise; most of them based on various computer model projections. But what is the actual current rate of sea level rise? According to the National Oceanic and Atmospheric Adminisrtration, average global sea level rise has speeded up and is now increasing at 0.12 inch per year. If that rate were sustained for the next 84 years, that would mean that sea level would increase by just over 10 inches. That's not nothing, but it is not much greater than the approximately 7 inch increase that occurred during the 20th century. Interestingly, millions of Americans did not retreat from the coasts in the past century.
As a final note, surely everyone can agree that it is particularly insane for the federal government to encourage people to move to the coasts by subsidiziing flood insurance.