Construction and analysis of consistent series of sea level measurements

 

According to the Intergovernmental Oceanographic Commission of UNESCO (1985), "The basis for any scientific analysis of sea level must be long series of careful measurements." Studies show that time series of more than 60 years are needed to determine long-term trends on sea level components such as:

  • changes in mean sea level (acceleration?) ;
  • changes and improvement of extreme levels;
  • changes in tidal waves and weather effects;
  • the spatial and temporal variability of sea level trends

Unfortunately, long tide gauge series are rare. However, analogue data do exist; in France, paper tide records have considerable potential.

 

Etudes niveau de la mer

 

Questions about the quality and exploitation of analogue historical sea level data

To answer these questions, it is necessary to digitize old data (at least in part), which are for the time being in analogue form (paper tide curves, paper records) then process and analyse them. Then the long process of constructing a homogeneous sea level series from heterogeneous sets of measurements begins. Reconstruction studies already carried out (detailed below) demonstrate the quality of digitised analogue data.

 

Steps involved in constructing a homogeneous sea level series

 

The steps for constructing a homogeneous sea level series are:

  1. Search for tide archives;
  2. Digitize the data
  3. Search the metadata for reference frames for heights and measurement times;
  4. Adjust the digitized values in heights and times respectively, expressing the water heights relative to the chart datum (usually) and time relative to Coordinated Universal Time (UTC);
  5. Perform typical quality controls: verify the time and height adjustments;
  6. If necessary, search for an explanation in the metadata. "Dubious" data are routinely discarded.

 

Reconstruction of the regional variability of the sea level by combining tide gauge data with spatial structures specific to the ocean derived from ocean models

The regional variability of the sea level in the past (prior to the altimetric period, 1993) can be reconstructed by combining tide gauge data with spatial structures specific to the ocean derived from ocean models. This method makes it possible to reconstruct the sea level in 2 dimensions since 1950, for most of the world, with a resolution close to that of satellite altimetry.

 

To find out more:

 

 

Last updated: 12/12/2012