WATER LEVEL AS A DATUM IN UPLIFT AND SUBSIDENCE INVESTIGATIONS: PITFALLS AND EFFECTIVE REMEDIES

In paleogeodetic studies where measurements of “sea level” to a precision of 0.1–0.01 m are required, we must understand the nature and causes of sea surface height (SSH) variability. Simply surveying the low or high water mark of a random day to estimate the shape of the tidal curve, as in many previous studies, is insufficient for this level of precision. Instead, I recommend a method in which the water level is surveyed and then a predictive tide model and retroactive corrections are used to calculate how much lower or higher certain water levels are at a site relative to the water level at the time of measurement.

The first step in the process, measuring water level, should be carried out over several minutes or longer, to capture and average over long-period swells. Water heights above a surveyed point can be read off a graduated staff at short intervals.

If a permanent tide gauge is located near a site, the continuous SSH record from the gauge can be used to determine past water levels compared with levels at the time of a survey. At sites far from tide gauges, one can create a “virtual tide gauge” (VTG) based on a harmonic tide model that is subsequently corrected for documented non-harmonic regional sea level anomalies. Unlike real tide gauges, VTGs do not record tectonic uplift or subsidence, but benchmarks or geological features such as microatolls can be used with a VTG to calculate ongoing or past land-level changes.

Numerous harmonic tide models utilize tidal harmonic constituents extracted from satellite observations to predict the ocean tide component of SSH variability at any location at any time in the past or future. The models do a good job of predicting ocean tides, although in complicated coastal settings a model’s resolution is important in determining its accuracy. The models do not, however, consider effects of the pole tide or non-harmonic influences on SSH, such as the inverted barometer effect, drag caused by winds, or longer-term variations such as those associated with the Indian Ocean Dipole (IOD) or El Niño–Southern Oscillation (ENSO).

In this poster, I discuss these tide models, highlight the principal phenomena that influence SSH variability, and explore computational tools (many of which have global application and are available online) that can be used to calculate and compensate for most of this variability.