The accuracy of subinertial sea level, as measured by tide gauges in t
he Southern California Bight, is investigated. Sea level differences f
ormed from tide gauges separated by less than 50 km are examined. The
observed differences give an upper bound on errors in the sea level da
ta, provided errors at each station are uncorrelated with each other a
nd with the true sea level signal. Sea level measurements at the San D
iego and La Jolla tide gauges are also compared with simultaneous bott
om pressure measurements made on the shelf 25 km north of San Diego. B
oth comparisons suggest that rms errors in tide-gauge measurements of
sea level are order 1.5 cm. Approximately half the observed error vari
ance has time scales of 1 month and is due, in part, to inaccuracies i
n determining reference levels.