> ... 90% of the hazard assessments assume coastal sea levels based on geoid models, rather than using actual sea-level measurements. > A geoid is an equipoten

> ... 90% of the hazard assessments assume coastal sea levels based on geoid models, rather than using actual sea-level measurements. > A geoid is an equipotential surface model that approximates MSL based on gravity and the rotation of Earth. As geoid quality depends on gravity observations, uncertainties in global geoid models can range up to several metres in regions that suffer from gravitational data paucity ..., predominantly located in the Global South. Moreover, actual sea-surface height is not just determined by the gravity and rotation of Earth, but also by, for example, ocean currents and large-scale circulation, winds, tides, seawater temperature and salinity. As a result, time-average sea-surface height can deviate strongly (up to several metres) from a geoid, and its difference is the so-called mean dynamic topography (MDT). I don't quite grasp how those flaws in geoid models lead to an overall significant underestimation of sea-level. Shouldn't all those average out? > uncertainties in global geoid models Why would those tend toward underestimation? > sea-surface height is not just determined by the gravity and rotation of Earth, but also by, for example, ocean currents and large-scale circulation, winds, tides, seawater temperature and salinity. Other than temperature and maybe salinity, those factors move water in a closed system: increasing water one place reduces it in another (?). > time-average sea-surface height can deviate strongly Why would average deviation by positive and not zero or negative?