Système d'Observation du Niveau des Eaux Littorales (SONEL)

Permanent GPS station at the Saint-Paul island tide gauge, Indian Ocean

Guy Woppelmann — 2012-04-21T16:37:46Z

Researchers from the LEGOS joint research unit (UMR5566, CNRS/Toulouse) in collaboration with the DT/INSU (UPS 855 CNRS/Brest) have installed a permanent GPS station at the Saint-Paul Island (38°41'S - 77°31 E) in the South Indian Ocean, on the GLOSS tide gauge site during the last NIVMER campaign. This installation was done in the framework of the French ROSAME sea level network. It aims at better constraining and explaining the recent estimation of relative sea level trend observed at this site, which is not significantly different from zero.

Source: S. Calmant & Fichen, L., 2012.
NIVMER12 campaign.

Don't hesitate to inform us about your news and results
sonel@sonel.org

Contact: Laurent Testut (UMR LEGOS)
laurent.testut@legos.obs-mip.fr
Tel. 05 61 33 27 85

Main results:


Fig. 1 (click to enlarge)

A data archaeology exercise was carried out on ancient sea level observations recorded during the transit of Venus across the Sun observed in 1874 from Saint-Paul Island, 38°41'S - 77°31 E, in the Southern Indian Ocean (Testut et al. 2010). Historical (1874) and recent (1994-2009) sea level observations were assembled into a consistent time series. A thorough check of the data and its precise geodetic connection to the same datum was only possible thanks to the recent installation of new technologies (GPS buoy and radar tide gauge sensor) and levelling campaigns. The estimated rate of relative sea level change, spanning the last 135 years at Saint-Paul, was not significantly different from zero (-0.1 ± 0.3 mm/yr), a value which could be reconciled with average global estimates of 1.7 mm/yr assuming the DORIS vertical velocity estimate at the nearby Amsterdam Island could be applied to correct for the land motion at the tide gauge (Altamimi and Collilieux, 2010). Considering the scarcity of long-term sea level data in the Southern Hemisphere, the exercise provides an invaluable additional observational constraint for further investigations of geographical sea level rise patterns.

Figure 1. (right) MSL estimates at Saint-Paul relative to the hydrographic zero for the historic observations (1874) and the recent observations (2009). The error bars indicate the uncertainty of the estimates. The grey curve is the annual mean altimetric sea level at Saint-Paul Island from AVISO product and the associated MSL is shown over the period as a rough guide. The orange line shows the GIA-induced relative sea level change rate, and the blue lines indicate the lower, upper and best global sea level trend estimate over the 20th century.

The data from the permanent GPS station that has been installed will provide an invaluable estimate of the Island vertical land movement. This information is critical to better interpret the rather astonishing result of no relative sea level elevation during the past 135 years. It will also be usefull for geodynamic studies on the dormant volcano of Saint-Paul island or on plate tectonics.

Figure 2. The GPS antenna above the tide gauge station in the inundated crater of the Saint-Paul island collapsed volcano.

References:
Altamimi Z. and X. Collilieux, 2010. Quality assessment of the IDS contribution to ITRF2008. Advances in Space Research, 45(12), 1500-1509.
Testut L., B. Martin Miguez, G. Wöppelmann, P. Tiphaneau, N. Pouvreau, M. Karpytchev, 2010.The sea level at Saint-Paul, Southern Indian Ocean, from 1874 to the present. Journal of Geophysical Research, 115, C12028, doi:10.1029/2010JC006404.

Importance of wave age and resonance in storm surges: the case Xynthia, Bay of Biscay

Guy Woppelmann — 2012-01-07T12:50:56Z

Researchers from the joint research unit "Littoral, Environment and Societies" (LIENSs, CNRS / La Rochelle University, team Physical Dynamics of the Littoral) in collaboration with the Portuguese National Civil Engineering Lab (LNEC, Lisbon) have investigated the storm surge associated with the storm Xynthia (February 2010) in the Bay of Biscay, combining in situ measurements with numerical modeling.

Their results show that the exceptional storm surge during Xynthia was amplified by the presence of young and steep waves in the Bay of Biscay, related to the unusual track of Xynthia. This study also shows the occurrence of a resonant phenomenon over the shelf, which explains the development of 6 hour oscillations after the surge peak.

Source: Bertin, X., Bruneau, N., Breilh, J.F., Fortunato, A.B., Karpytchev, M., 2012.
Importance of wave age and resonance in storm surges: the case Xynthia, Bay of Biscay. Ocean Modelling, 42, pp. 16-30.

Don't hesitate to inform us about your news and results
sonel@sonel.org

Contact: Xavier Bertin (UMR LIENSs)
xbertin@univ-lr.fr
Tel. 05 46 50 76 36

Main results:

Due to the combination of a wide continental shelf and their position on the track of mid-latitude winter storms, northwestern European shorelines are exposed to storm surges that often exceed 1 m. Low-lying coasts, such as those of the Wadden Sea, Normandy (English Channel) or Pertuis Charentais (Bay of biscay) are thus vulnerable to marine flooding. The storm Xynthia, which severely hit the coasts of Vendée and Charente Maritime (central part of the Bay of Biscay) in the night between the 27th and the 28th of February 2010, reminded this vulnerability. A 970 mbar low-pressure, accompanied with southwestern winds of the order of 100 to 130 km/h (peaks at 160 km/h on Ré island), induced a storm surge with a maximum measured at 1.50 m at La Pallice tide gauge (La Rochelle, Figure 1). The surge peak matched with a high spring tide, causing the overall water level to reach 8 m above marine Chart datum, a value never attained since the deployment of the numerical tide gauge (28th of April 1997, data available at SONEL/REFMAR). Many natural barriers and dykes that protected low-lying zones breached, causing the flooding of large areas and killing 47 peoples and huge material damages estimated to 1.5 billion euros.

Figure 1. Observed water level (black), astronomic tide (red) and storm surge (blue) at La Pallice tide gauge, showing a 1.5 m storm surge in phase with high spring tides

The study of Bertin et al. (2012) corresponds to the first step of an integrated research project aiming at investigating storm surges and marine flooding in the central part of the Bay of Biscay. A storm surge numerical modelling system was developed based on the 2DH coupling of the hydrodynamic circulation model SELFE (Zhang et Batista, 2008) and the spectral wave model WaveWatchIII (Tolman, 2009). This modeling system was implemented over the North East Atlantic Ocean and resulted in tidal predictions with errors of the order of 1 to 2 % and wave predictions with errors of the order of 15%. In the Bay of Biscay, the storm surge associated with Xynthia was reproduced with an error of the order of 0.1 m

The analysis of numerical results shows that, in the central part of the Bay of Biscay, the dominant mechanism in the storm surge generation was the SW-oriented winds, which drove an Ekman pumping. This phenomenon corresponds to water transport to the right of the wind direction, due to the Coriolis Effect. The comparison between a quadratic parameterization (wind-dependant only) and a wave-dependant parameterization to compute the wind-induced stress revealed that the storm surge was much more realistically predicted when taking waves into account (Figure 2A). This wind stress is computed based on the wind friction velocity U*, also defined as the ratio between the wave phase speed and the wave age. This friction velocity is thus very important in the presence of young and steep waves, which increase the ocean roughness (Figure 2B). Such young waves were related to the unusual track of Xynthia, which crossed the Bay of Biscay from SW to NE, thereby reducing the fetch to a few hundred kilometers only.

Figure 2. (A) Comparison between the storm surge predicted in La Pallice with a quadratic parameterization (red), a wave-dependant parameterization (black), showing that the second matches much better with the observations (blue). (B) Wind friction velocity (black) used to compute the wind stress and wave age (blue) showing a drop during Xynthia.

Finally, field data as well as numerical results show the occurrence of 6-hour oscillations in the storm surge signal, the height of which reaches 0.4 m in the central part of the Bay of Biscay (Figure 1, blue curve). This 6-hour period corresponds to the resonance period of the continental shelf in the central part of the Bay of Biscay. The researchers of LIENSs have interpreted this phenomenon as resulting from the fast displacement of Xynthia (15-20 m/s) in the Bay of Biscay, which induced a perturbation of the free-surface elevation with a significant amount of energy around the resonant frequency. This energy was then amplified by resonance on the shelf. Works in progress suggest that this phenomenon could be valid at other places around the world's coastlines, such as the central part of the Gulf of Mexico, of for other long waves, such as tsunamis.

References:
Tolman H. L., 2009. User manual and system documentation of WAVEWATCH III version 3.14. NOAA/NWS/NCEP/MMAB Technical Note 276, 194 pp.
Zhang, Y. L. and A. M. Baptista, 2008. SELFE: A semi-implicit Eulerian-Lagrangian finite-element model for cross-scale ocean circulation. Ocean Modelling, 21(3-4), 71-96.

Inventory of Sea Level Data in Need of Rescue

Guy Woppelmann — 2011-12-13T16:45:25Z

On behalf of GLOSS (Cover letter below), an effort has been initiated to learn of historic, non-computerized sea level data.
If your affiliation has such historic records, could you please fill in the questionnaire below (in .doc format, clicking on the image or below)?

Or if you know of contacts for such data in need of rescue, could you please provide contact information?
Send either to Mr. Patrick Caldwell from the UHSLC by May 1, 2012.
caldwell@hawaii.edu

Cover letter from GLOSS
Questionnaire

Twelfth Session of the GLOSS Group of Experts, Unesco, Paris, 7-11 November 2011

Guy Woppelmann — 2011-10-05T12:53:21Z

The next meeting of the global sea-level observing programme (GLOSS) is taking place at Unesco, Paris, 9-11 November 2011. Two workshops will precede the event, respectively dealing with:

Storm surge monitoring and extreme sea levels, and the need for high-quality real-time sea level data (7-8 November 2011);

Space-based methods for vertical tide gauge control (afternoon of 8 November and morning of 9 November 2011).

The GLOSS group of experts meeting itself will take place from 9 to 11 November 2011. The agendas are in the process of elaboration (See 1st link below).

Further details :
on these meetings
on the GLOSS programme

Satellite Altimeter Calibration and Deformation Monitoring using GNSS

Guy Woppelmann — 2011-05-03T10:48:00Z

Papers are invited for a special topic issue of Advances in Space Research (ASR) entitled “Satellite Altimeter Calibration and Deformation Monitoring using GNSS”.

Global warming is a key scientific issue nowadays. Over the past decades, satellite altimetry has provided an unprecedented amount of data for monitoring sea level changes and thus global warming. Such scientific results however rely on a significant number of geophysical corrections as well as on continuous monitoring of the satellite altimeter performances using in-situ data. A few calibration sites around the world provide accurate calibration of satellite altimeters using different in-situ techniques involving monitoring of geodetic instruments close to the satellite tracks or at crossing points (platforms, tide gauges, buoys, etc.). While Global Navigation Satellite Systems, such as GPS, are key tools for geodetically monitoring the vertical velocity of such calibration sites, several scientific issues are still under consideration, as vertical positioning is still difficult to obtain in an absolute sense. Furthermore, new satellite altimeters, such as delay-Doppler, wide swath, etc., will soon provide new types of observations for sea-level monitoring and allowing altimetry to be closer at the coastal regions, for which new types of calibration could be foreseen.

This Special Issue is open to all scientists discussing new scientific challenges in satellite altimeter calibration (technical aspects related to the altimeter functioning, description of current calibration sites and current techniques, or global satellite calibration using networks of tide gauges or GNSS networks), as well as current limitations in deformation monitoring (monitoring of the vertical, use of regional and continuously-operating GNSS networks, reference issues when expressing vertical velocities in global, regional or local frames). Paper addressing topics related to natural hazards would also be welcome.

Papers must be submitted electronically through the EES/Elsevier Web system at http://ees.elsevier.com/asr. To ensure that all manuscripts are correctly identified for inclusion into the Special Issue, authors must select “Special Issue: Altimetry Calibration”, when they reach the “Article Type” step in the submission process.

Submitted papers must be written in English and should indicate full affiliation addresses for all authors. All manuscripts will be subject to a standard peer-review process, including at least two independent reviewers. There are no page limits, but the length of the paper should be appropriate to the scientific material being presented. Manuscripts should follow the general format used in ASR: (http://www.elsevier.com/wps/find/jo...). The deadline for submission is 15 January 2012.

Manuscripts will be accepted on a paper-by-paper basis and available on-line with a DOI a few days after acceptance. Contributors to this issue will have an opportunity to purchase individual issues once the issue is finalized. All articles will be typeset at no cost to the author, while a nominal fee will be charged to the authors for color figures in print, but not in the electronic version.

Dr. Pascal Willis (pascal.willis@ign.fr) is the Guest Editor for this Special Issue. Authors will be regularly informed on the progress of this Special Issue by email until its final publication. Questions can be directed to Pascal Willis or to the Co-Editor for Special Issues, Dr. Peggy Ann Shea (sssrc@msn.com).

Reorganisation of SONEL, Tide gauge data and the national liaison agency (REFMAR)

Frank — 2011-02-16T14:06:24Z

One of the reasons for which SONEL was founded was to federate the tide gauge networks that exist in France, and ensure that emerging networks were properly incorporated. Although the project was initiated by B. Simon at the SHOM (1999) it subsequently ran into numerous practical difficulties. Finally, it was partially implemented at the University of La Rochelle in 2003 (first on-line observations) and involved just the RONIM and ROSAME networks (cf. actes AEI 2004).

Today, the context is more favourable. The SHOM has been designated as the official national liaison for in situ sea level observations, together with the management and publication of the resulting data: see the "instruction permanente du Premier ministre n°863/SGMer du 20 avril 2010". Today it is tasked with :

coordinating the gathering of sea level data produced by the organisations, agencies and public services that are within the competence of the state authorities ;
Managing and archiving the data in real time and in delayed mode ;
coordinating the publication in real time and in delayed mode of the data while ensuring that the conditions of accessibility and use comply with current legislation pertaining to information in the public domain.

To this end, the SHOM set up REFMAR (opposite) where users can find the data (raw in real time, and quality checked in delayed mode) from tide gauges on French territory. The fact that the SHOM has now taken on the tide gauge data aspect is excellent news and will enable the University of La Rochelle to concentrate its efforts on the other aspects of the SONEL system (see the architecture below). Moreover, it guarantees in particular that tide gauge data is managed by recognised specialists in the field, heirs to an experience which goes back to the first French tide gauge designed by the hydrographic engineer R. Chazallon in 1843.


Architecture SONEL

The infrastructure of SONEL has been reorganised accordingly, particularly the Web and FTP servers of SONEL, in order to bring it into line with the architecture described in the diagram opposite. It incorporates REFMAR in a manner which is optimal and as transparent as possible for users. Numerous links have been created at every level of SONEL, in particular to achieve the initial aim of a synergy of the tide gauge and geodesic data.

In practice

There will be a transition phase which we will ensure to the best of our ability. During this transition phase users will find French tide gauge data both on the ftp.sonel.org and www.sonel.org servers and on the refmar.shom.fr servers . However, they should be aware that they would be well-advised to look for the data at its SHOM (REFMAR) source . The SHOM has both a recognised expertise and centuries-long experience in the field of tide gauges. Moreover, any new development or change in the French tide gauge network will immediately be incorporated in REFMAR given the mission with which it has been entrusted (cf. instruction permanente), and whatever the French tide gauge network (RONIM, ROSAME, etc.).


Phase de transition sur SONEL


A terme sur SONEL

The source of high frequency data (REFMAR) will be identified and accessible from the SONEL Web pages of each station above the current visualisation and download sections during the transition phase (e.g., image on the left), although these visualisation and download sections will eventually disappear from the SONEL pages and only the link toward the source (REFMAR) will be available (e.g., image on the right). The images opposite provide an example of how this will look during the transition phase (which begins today) and going forward.

Agenda

As from today, users of high frequency tide gauge data (hourly, 10min.,...) are invited to register on REFMAR.
Beginning of March, the high frequency tide gauge data will only be accessible on REFMAR, initially on the current provisional server, and eventually on the permanent server, which is under validation (beginning of April).

Daily, monthly and annual mean sea levels

Guy Woppelmann — 2010-12-19T12:44:47Z

The data and graphs concerning daily, monthly and annual mean sea levels have been available on SONEL since October 2010 for the French stations (a link to the PSMSL is provided for international stations when the page specific to the station is identified and the tide gauge is co-located with GPS or DORIS). The means are calculated from the Demerliac filter and the Doodson filter (see details in the article "Calculating the mean sea levels"), with the possibility of a choice for the user.

The graphs

The graphs in the individual pages per station show the means calculated from quality controlled (QC'ed) observations (in blue) and non-QC'ed observations (in red). An example is given on the image below, which is taken from the page of the La Rochelle station.

The non-QC'ed observations are files of raw measurements sampled every 10 minutes. A 10-min raw mesurement usually corresponds to a sea level value averaged over an integration period of two minutes (short-term oscillations such as waves are filtered out) from 1-second samples. They are not in any way verified by SHOM, and should therefore be used with the greatest of precaution, even though a high-tech modern instrument is used (See the results of experiments conducted with radar gauges published in Journal of Coastal Research). The 10-min data files are available roughly one week after the actual date on which the measurement was made.

The QC'ed observations, which are hourly data files, have been verified by the SHOM, after being subjected to different quality controls. The values come from the 10-min. observation files which are resampled on the hour. The QC'ed data are available roughly 2-3 months after the actual date on which the measurements were made. (For details on the quality controls carried out by SHOM).

The files of average values and their format

Only the daily, monthly and annual averages calculated from QC'ed data are available. They are presented in the form of a two-column matrix : the date (in decimal years) and sea level (in millimetres). The format adopted is that of the PSMSL, as are the calculation criteria for the monthly and annual means, in particular for the issue of observation gaps.

They can be downloaded individually from the Web page of the tide gauge by clicking on the button situated under the graph of the means chosen. However, they are also available on the FTP server (ftp.sonel.org) which is very practical when a rapid acces to the data from several stations is required. This is how the names of the data files are defined :

The first letter in lower case :
"d" for the daily means, "m" for the monthly means and "y" for the yearly means
The acronym of the station, in capital letters, is made up of 5 letters (the internal identifier of the SONEL data base)
The extension ".slv", for the type of quantity "sea level".

Example: the daily means file of La Rochelle is called dLROCH.slv

Structure of the file system in the FTP server

|-- legos
 | |-- rosame
 | `-- sao_tome
 |-- shom
 | |-- HAUT_10 => Files of observations at 10 minutes
 | | |-- RECENT -> with the 50000 latest observations at 10 min.
 | | `-- TOTAL -> complete
 | `-- HAUT_HOR => Files of QC'ed hourly observations
 | |-- TOTAL -> complete
 | `-- VALID -> with the observations of the current year
 `-- MSL
 |-- Demerliac => Name of the filter used
 | |-- data -> Files of mean values
 | `-- graphs -> Graphs displayed on the Web site 
 `-- Doodson => Name of the filter used
 |-- data -> Files of mean values
 `-- graphs -> Graphs displayed on the Web site

Reminder of the directories and their content

legos : contains the tide gauge observations of the ROSAME network and that of Sao Tomé run by LEGOS.

shom : contains the tide gauge observations of the RONIM network of SHOM. The sub-directory HAUT_10 contains the observations sampled at 10 min. whereas HAUT_HOR contains the QC'ed hourly observations.

MSL (Mean Sea Level) : contains the daily, monthly and annual means, calculated from the QC'ed hourly observations.