Performance of modern tide gauges: towards mm-level accuracy

International audience Considerable efforts are being made worldwide to upgrade tide gauge networks using new technologies. Because of the unique location of the Kerguelen Islands, the measurement of sea level there has received particular attention, with up to four systems equipped with modern sens...

Full description

Bibliographic Details
Published in:Scientia Marina
Main Authors: Martin-Miguez, Belen, Testut, Laurent, Woppelmann, Guy
Other Authors: LIttoral ENvironnement et Sociétés (LIENSs), La Rochelle Université (ULR)-Centre National de la Recherche Scientifique (CNRS), Service d'observation SONEL (www.sonel.org) de l'INSU
Format: Article in Journal/Newspaper
Language:English
Published: HAL CCSD 2012
Subjects:
Online Access:https://hal.science/hal-01247934
https://doi.org/10.3989/scimar.03618.18A
Description
Summary:International audience Considerable efforts are being made worldwide to upgrade tide gauge networks using new technologies. Because of the unique location of the Kerguelen Islands, the measurement of sea level there has received particular attention, with up to four systems equipped with modern sensors functioning simultaneously (two pressure tide gauges, a radar tide gauge, and a GPS-equipped buoy). We analysed and compared the sea level data obtained with these systems from 2003 to 2010, together with a time series of tide pole observations. This is the first time that a multi-comparison study with tide gauges has been undertaken over such a long time span and that the stability of modern radar tide gauges has been examined. The multi-comparison enabled us to evaluate the performance of the tide gauges in several frequency ranges, identify errors and estimate their magnitude. The drift of the pressure sensors (up to 8.0 mm/yr) was found to be one of the most relevant sources of systematic error. Other sources of difference such as clock drift, scale error and different locations of the instruments were also detected. After correcting the time series of sea level for these errors we estimated an upper bound for the radar instrumental error in field condition at ~0.3 cm.