EM 1110-2-1100 (Part II)
30 Apr 02
may not be comparable with similar measurements at other geographic locations because of the difficulty of
assigning a comparable gauge zero. The most widely accepted of the datums computed from time series are
(2) Mean sea level (MSL) was widely adopted as a primary datum on the assumption that it could be
accurately computed from tidal elevation records measured at any well-exposed tide gauge. MSL
determinations are based on the arithmetic average of hourly water surface elevations observed over a long
period of time. The ideal length of record is approximately 19 years, a period that accounts for the 18- to 19-
year long-term cycle in tides and is sufficient to remove most meteorological effects. In order to fix the
datum in time for a specific location, a common 19-year period is selected for computing MSL. The specific
19-year cycle was adopted by the National Ocean Survey as the official time segment for use in computing
mean values for tidal datums. The 19-year period, called the National Tidal Datum Epoch, is updated
approximately every 25 years.
(3) When estimates of MSL are required, but less than 19 years of data are available, computations
should be based on an integral number of tidal cycles, for example, an integral number of years or 29-day
spring/neap cycles. For gauges where hourly data are not available, or their use is impractical, MSL can be
approximated as the tidal datum midway between MHW and MLW. This datum, referred to as Mean Tide
Level (MTL), may differ from MSL depending on the local relative importance of the diurnal components
of the tide.
(4) Alternate tidal datums are based on low water tidal elevations. These datums provide minimum
depth information for navigational needs. Two commonly used low-water datums in the United States are
the Mean Low Water (MLW) for the Atlantic Coast and the Mean Lower Low Water (MLLW) for the Pacific
coast. The MLLW datum is currently being adopted to several locations along the Atlantic coast. Both
datums are defined as the average tidal height at low water or lower low water during the 19-year period.
Additional datums, applicable to specific locations or purposes, include Mean High Water (MHW) and Mean
Higher High Water (MHHW). These are derived in a manner similar to MLW and MLLW. For areas of
primarily semidiurnal tides, the difference between MHW and MLW is called the mean tidal range. The
difference between MHHW and MLLW gives a corresponding diurnal range or the great diurnal range of the
tide. Both numbers provide an estimate of the magnitude of the local tidal range. An example of the
variability of the above datums is given by Harris (1981) and reproduced in Table II-5-7. Reference tide
stations used in the preparation of Table II-5-7 are shown in Figure II-5-17.
1929 NGVD datum.
(1) One difficulty with using any of the observation-based datums described above is that they vary
considerably with location, as evidenced in Table II-5-7. Also, because each datum can be computed
independently, there is little or no connectivity between datum locations. This lack of a reference elevation
for areas near the coast where no tide observations are available and at interior locations where tide
observations are difficult to obtain led to the establishment of a national fixed datum. This datum does not
account for spatial variability in sea level. The following paragraphs describe the development of the 1929
National Geodetic Vertical Datum (NGVD).
(2) First-order leveling lines established in the mid-1920's provided survey connections between the
Atlantic and Pacific coasts. These surveys indicated that sea levels were higher on the Pacific coast than on
the Atlantic coast and were also higher in the north than in the south on both coasts. The goal of developing
a fixed reference datum was accomplished by defining a geodetic leveling-based datum whose "zero"
coincided with local MSL at locations at which both MSL and geodetic leveling elevations were known.
Water Levels and Long Waves