To be more specific, 9 feet (2.7m) is the report's worst-case upper bound, the 99.9th-percentile results of running a new simulation model, RCP 8.5, not a predi

To be more specific, 9 feet (2.7m) is the report's worst-case upper bound, the 99.9th-percentile results of running a new simulation model, RCP 8.5, not a prediction of what's likely to happen. That model has a huge range: Its 50th-percentile (median) estimate is only 1 foot (0.3m) of rise, and its 95th-percentile estimate is 7.5 feet (2.3m). There's discussion of that on pp. 31-32 of the summary report here (PDF): http://www.climateassessment.ca.gov/state/docs/20180827-StatewideSummary.pdf http://www.climateassessment.ca.gov/state/docs/20180827-Stat... The chart on p. 32 is a bit tricky to read, because this particular model is estimating maximum sea levels, not mean sea levels. Maximum sea level here is measured relative to a fixed reference of year 2000 mean sea level, and the 20th-century typical max was 1.5m above mean year 2000 sea level. The 99.9th-percentile line at 4.2m is therefore showing a 2.7m rise in the max sea level above pre-2000 norms. The colored blue and red ranges below the 99.9 lines represent the 50th-95th percentile estimates of the two different models. (In this case, at La Jolla specifically.)